The question that this case presents is whether defendant Pacific Scientific Company ("Pacific") engaged in "exclusionary practices" in violation of the anti-monopoly law, Sherman Act Sec. 2, 15 U.S.C. Sec. 2. The practices at issue are embodied in agreements between Pacific and ITT Grinnell ("Grinnell"), under which Pacific agreed to sell its product (mechanical snubbers) to Grinnell at a specially low price and Grinnell agreed to take nearly all its snubber requirements from Pacific. The district court, 555 F. Supp. 1264, found that the relevant contract provisions did not violate the antitrust laws. We agree with the district court on this matter, and on others less important; and we affirm its judgment.

* On appeal from a judgment in defendant's favor, we must, of course, accept the district court's findings insofar as they determine, or rest upon, the facts, unless they are "clearly erroneous." Fed.R.Civ.P. 52(a). Our reading of the record, in light of this standard, suggests the following somewhat simplified account of the most important factual matters:Pacific produces mechanical snubbers; they are shock absorbers used in building pipe systems for nuclear power plants. No other domestic firm makes mechanical snubbers; foreign mechanical snubbers do not satisfy Nuclear Regulatory Commission standards; and snubber users have found the closest substitute, namely, hydraulic snubbers, to be less reliable than the mechanical version. Thus, in 1976, Pacific's sales accounted for 47 percent of all snubbers sold; in 1977, 83 percent; in 1978, 84 percent, and in 1979, 94 percent.

Grinnell makes and installs nuclear plant pipe systems; it is a major snubber user. Its snubber purchases accounted for 51 percent of all mechanical snubbers and related hardware sold domestically in 1977; 52 percent in 1978; and 43 percent in 1979. By 1976, most of Grinnell's pipe system customers were requiring Grinnell to use mechanical snubbers. Recognizing Pacific's strong market position, Grinnell sought to develop an alternate mechanical snubber source. Hence, it entered into a contract with Barry Wright Corporation ("Barry"), the plaintiff here, under which it would help Barry develop a full mechanical snubber line. Grinnell agreed to contribute to Barry's development costs. It also agreed to use Barry as an exclusive source of supply between 1977 and 1979, promising to buy between $9 million and $15 million worth of snubbers during that period. Barry was to have its full line of six snubber sizes in production by the first quarter of 1977.

While waiting for Barry, Grinnell satisfied its current needs by buying mechanical snubbers from Pacific at Pacific's ordinary "discount" price--20 percent below list. Pacific noticed that Grinnell's orders seemed small in relation to its likely needs. And, by September, 1976, Pacific realized that Grinnell was trying to develop its own supply source through Barry. In August, 1976, Pacific offered Grinnell a special price break--a 30 percent discount from list for small snubbers, 25 percent for larger ones--in return for a large $5.7 million order that would have met Grinnell's snubber needs through 1977. Grinnell, after tentatively accepting this proposal, consulted with Barry and then rejected Pacific's offer. Instead, it placed a smaller $1 million order at the standard 20 percent discount.

Barry could not meet the required January, 1977 production schedule. By mid-January, 1977, it told Grinnell that it would not be able to produce small snubbers until August, 1977 nor large ones until February, 1978. Grinnell then met with Pacific and (at the end of January, 1977) negotiated a $4.3 million snubber contract--enough snubbers to meet Grinnell's estimated needs for the next twelve months. Pacific gave Grinnell the large 30 percent/25 percent discounts. It also gave Grinnell an option--open until July, 1977--to buy its 1978 requirements at the same prices (as long as Grinnell agreed to buy as much as in 1977). Grinnell, in turn, agreed to a non-cancellation clause that would have made it especially onerous for Grinnell to break the agreement.

Grinnell then told Barry that its production delays were unacceptable and that Barry had breached its development contract. Grinnell wanted Barry to continue its efforts, but it would not promise to buy more than $3.6 million worth of snubbers through 1979. Barry said this modification of the development contract was unjustified. It continued to try to develop snubbers. The extent of its progress is in dispute, but there is considerable evidence that it fell further behind its production schedules.

At the end of May, 1977, Grinnell and Pacific agreed further that Grinnell would buy $6.9 million worth of Pacific's snubbers for 1978 and $5 million for 1979. Grinnell predicted snubber "needs" of $6.9 million for each of the two years. On July 5, 1977 and July 14, 1977, Grinnell finalized the agreements for 1978 and 1979 by issuing purchase orders in these amounts. Pacific granted the special 30 percent/25 percent price discounts; and the contracts contained the special cancellation clause.

In June, 1977, Grinnell told Barry that their collaboration was at an end. Barry looked for other potential snubber buyers and then abandoned its snubber efforts. Subsequently, Barry brought this lawsuit against Pacific (and against Grinnell, as well, although Barry and Grinnell have reached a settlement). Barry charged that Pacific's efforts to sell snubbers to Grinnell and the terms of its contracts violated Sections 1 and 2 of the Sherman Act, 15 U.S.C. Secs. 1, 2, and Section 3 of the Clayton Act, 15 U.S.C. Sec. 14. It added a claim that Pacific had tortiously interfered with Barry's snubber-development contract with Grinnell.

The district court found that Barry failed to establish that Pacific's conduct was improper. The court also found that Barry had breached its contract with Grinnell before Grinnell placed its 1977 order with Pacific (in late January, 1977). And, it found that Barry would not have been capable of supplying Grinnell in 1977-79 regardless. Barry appeals all these findings. We need not consider the latter two, however, for we find the district court's conclusion about the lawfulness of Pacific's behavior adequately supported. And that conclusion is dispositive.

Barry's central claim is that Pacific's conduct violates Sherman Act Sec. 2, which makes it unlawful to "monopolize ... any part of the trade or commerce among the several States." 15 U.S.C. Sec. 2. Monopolization has two elements: first, the "possession of monopoly power in the relevant market" and, second, the "acquisition or maintenance of that power" by other than such legitimate means as patents, "superior product, business acumen, or historic accident." United States v. Grinnell Corp., 384 U.S. 563, 570-71, 86 S. Ct. 1698, 1703-04, 16 L. Ed. 2d 778 (1966). On this appeal, the parties do not dispute Pacific's monopoly power in the relevant market, which the district court identified as the domestic market for snubbers (whether mechanical or hydraulic). Nor do they dispute the legitimacy of Pacific's acquisition of this power. Whether its acquisition rested upon the fact that Pacific possessed an important mechanical snubber patent, upon a "superior product," or upon the "historic accident" that users began to find hydraulic snubbers unsatisfactory, no one alleges that Pacific acted unlawfully in acquiring its dominant position. Accepting these two assumptions, we turn to the issue that is in dispute, whether Pacific maintained its monopoly position against the threat of Barry's entry through improper means.

In this context, a practice, a method, a means, is "improper" if it is "exclusionary." United States v. United Shoe Machinery Corp., 110 F. Supp. 295, 342 (D.Mass.1953), aff'd per curiam, 347 U.S. 521, 74 S. Ct. 699, 98 L. Ed. 910 (1954). To decide whether Pacific's conduct was exclusionary, we should ask whether its dealings with Grinnell went beyond the needs of ordinary business dealings, beyond the ambit of ordinary business skill, and "unnecessarily excluded competition" from the snubber market. Greyhound Computer Corp. v. International Business Machines Corp., 559 F.2d 488, 498 (9th Cir.1977), cert. denied, 434 U.S. 1040, 98 S. Ct. 782, 54 L. Ed. 2d 790 (1978). Professors Areeda and Turner have put the matter nicely: " 'Exclusionary' conduct is conduct, other than competition on the merits or restraints reasonably 'necessary' to competition on the merits, that reasonably appears capable of making a significant contribution to creating or maintaining monopoly power." 3 P. Areeda and D. Turner, Antitrust Law p 626 at 83 (1978). Was Pacific's conduct reasonable in light of its business needs or did it unreasonably restrict competition? California Computer Products, Inc. v. International Business Machines Corp., 613 F.2d 727, 735-36 (9th Cir.1979). Barry points to three specific aspects of Pacific's conduct--its offer of special discounts to Grinnell, its insistence on a long-term large-volume contract, and its inclusion of the special non-cancellation clause--which it claims show that Pacific acted in an exclusionary manner.

* Barry first attacks the special 30 percent/25 percent discounts that Pacific granted Grinnell. It argues that Pacific's discounted prices were unreasonably low. This argument founders, however, on the district court finding that these prices, while lower than normal, nonetheless generated revenues more than sufficient to cover the total cost of producing the goods to which they applied. Barry does not attack that finding; but, instead, it argues that price cutting by a monopolist may still prove unlawful, even if prices remain above total cost. While some circuits have accepted a form of Barry's argument, see, e.g., Transamerica Computer Co. v. International Business Machines Corp., 698 F.2d 1377 (9th Cir.), cert. denied, --- U.S. ----, 104 S. Ct. 370, 78 L. Ed. 2d 329 (1983); International Air Industries, Inc. v. American Excelsior Co., 517 F.2d 714, 724 (5th Cir.1975), cert. denied, 424 U.S. 943, 96 S. Ct. 1411, 47 L. Ed. 2d 349 (1976), we do not.

To understand the basis of our disagreement, one must ask why the Sherman Act ever forbids price cutting. After all, lower prices help consumers. The competitive marketplace that the antitrust laws encourage and protect is characterized by firms willing and able to cut prices in order to take customers from their rivals. And, in an economy with a significant number of concentrated industries, price cutting limits the ability of large firms to exercise their "market power," see J. Bain, Industrial Organization ch. 5 (2d ed. 1968); F. Scherer, Industrial Market Structure and Economic Performance 56-70, 222-25 (2d ed. 1980); at a minimum it likely moves "concentrated market" prices in the "right" direction--towards the level they would reach under competitive conditions. See 2 P. Areeda & D. Turner, Antitrust Law p 404; J. Bain, supra, at 118-23; F. Scherer, supra, ch. 5. Thus, a legal precedent or rule of law that prevents a firm from unilaterally cutting its prices risks interference with one of the Sherman Act's most basic objectives: the low price levels that one would find in well-functioning competitive markets. See, e.g., National Society of Professional Engineers v. United States, 435 U.S. 679, 692-96, 98 S. Ct. 1355, 1365-67, 55 L. Ed. 2d 637 (1978); MCI Communications Corp. v. American Telephone and Telegraph Co., 708 F.2d 1081, 1114 (7th Cir.), cert. denied, --- U.S. ----, 104 S. Ct. 234, 78 L. Ed. 2d 226 (1983); 1 P. Areeda and D. Turner, Antitrust Law paragraphs 103-05, 111-12.

Despite these considerations, courts have reasoned that it is sometimes possible to identify circumstances in which a price cut will make consumers worse off, not better off. See, e.g., Northeastern Telephone Co. v. American Telephone and Telegraph Co., 651 F.2d 76, 88 (2d Cir.1981), cert. denied, 455 U.S. 943, 102 S. Ct. 1438, 71 L. Ed. 2d 654 (1982). Suppose, for example, a firm cuts prices to unsustainably low levels--prices below "incremental" costs. Suppose it drives competitors out of business, and later on it raises prices to levels higher than it could have sustained had its competitors remained in the market. Without special circumstances there is little to be said in economic or competitive terms for such a price cut. Yet, how often firms engage in such "predatory" price cutting, whether they ever do so, and precisely when, is all much disputed--a dispute that is not surprising given the difficulties of measuring costs, discerning intent, and predicting future market conditions. See, e.g., Koller, The Myth of Predatory Pricing: An Empirical Study, 4 Antitrust L. & Econ.Rev. 105 (1971); McGee, Predatory Pricing Revisited, 23 J.L. & Econ. 289 (1980); Posner, Exclusionary Practices and the Antitrust Laws, 41 U.Chi.L.Rev. 506, 516-17 (1974); F. Scherer, supra, at 335-40.

Despite this dispute, there is general agreement that a profit-maximizing firm might sometimes find it rational to engage in predatory pricing; it might do so if it knows (1) that it can cut prices deeply enough to outlast and to drive away all competitors, and (2) that it can then raise prices high enough to recoup lost profits (and then some) before new competitors again enter the market. See Areeda & Turner, Predatory Pricing and Related Practices Under Section 2 of the Sherman Act, 88 Harv.L.Rev. 697, 698-99 (1975) [hereafter cited as Areeda & Turner, Predatory Pricing ]. There is also general agreement that the antitrust courts' major task is to set rules and precedents that can segregate the economically harmful price-cutting goats from the more ordinary price-cutting sheep, in a manner precise enough to avoid discouraging desirable price-cutting activity. See P. Areeda & D. Turner, Antitrust Law paragraphs 711.1a-b, 714.1b, 714.5 (1982 Supp.); Joskow & Klevorick, A Frameworker for Analyzing Predatory Pricing Policy, 89 Yale L.J. 213 (1979).

Barry, of course, suggests that Pacific's price cut is a "goat," arguing that Pacific "intended" to drive Barry from the market place. Some courts have written as if one might look to a firm's "intent to harm" to separate "good" from "bad." See, e.g., D.E. Rogers Associates, Inc. v. Gardner-Denver Co., 718 F.2d 1431, 1435-36 (6th Cir.1983); Forster Manufacturing Co. v. FTC, 335 F.2d 47 (1st Cir.1964), cert. denied, 380 U.S. 906, 85 S. Ct. 887, 13 L. Ed. 2d 794 (1965). But "intent to harm" without more offers too vague a standard in a world where executives may think no further than "Let's get more business," and long-term effects on consumers depend in large measure on competitors' responses. See Zerbe & Cooper, An Empirical and Theoretical Comparison of Alternative Predation Rules, 61 Tex.L.Rev. 655, 659-677 (1982). Moreover, if the search for intent means a search for documents or statements specifically reciting the likelihood of anticompetitive consequences or of subsequent opportunities to inflate prices, the knowledgeable firm will simply refrain from overt description. If it is meant to refer to a set of objective economic conditions that allow the court to "infer" improper intent, see, e.g., D.E. Rogers Associates, Inc. v. Gardner-Denver Co., 718 F.2d at 1436; cf. O. Hommel Corp. v. Ferro Corp., 659 F.2d 340, 347 (3rd Cir.1981), cert. denied, 455 U.S. 1017, 102 S. Ct. 1711, 72 L. Ed. 2d 134 (1982), then, using Occam's razor, we can slice "intent" away. Thus, most courts now find their standard, not in intent, but in the relation of the suspect price to the firm's costs. And, despite the absence of any perfect touchstone, modern antitrust courts look to the relation of price to "avoidable" or "incremental" costs as a way of segregating price cuts that are "suspect" from those that are not. See, e.g., Northeastern Telephone Co. v. American Telephone and Telegraph Co., 651 F.2d at 86-91; cf. Americana Industries, Inc. v. Wometco de Puerto Rico, Inc., 556 F.2d 625, 628 (1st Cir.1977); see generally Areeda & Turner, Predatory Pricing.

One can understand the intuitive idea behind this test by supposing, for example, that a firm charges prices that fail to cover these "avoidable" or "incremental" costs--the costs that the firm would save by not producing the additional product it can sell at that price. Suppose further that the firm cannot show that this low price is "promotional," e.g., a "free sample." Nor can it show that it expects costs to fall when sales increase. Then one would know that the firm cannot rationally plan to maintain this low price; if it does not expect to raise its price, it would do better to discontinue production. Moreover, equally efficient competitors cannot permanently match this low price and stay in business. Further, competitive industries are typically characterized by prices that are roughly equal to, not below, "incremental" costs. See F. Scherer, supra, at 13-14; Areeda & Turner, Predatory Pricing at 702-03, 712. At a minimum, one would wonder why this firm would cut prices on "incremental production" below its "avoidable" costs unless it later expected to raise its prices and recoup its losses. When prices exceed incremental costs, one cannot argue that they must rise for the firm to stay in business. Nor will such prices have a tendency to exclude or eliminate equally efficient competitors. Moreover, a price cut that leaves prices above incremental costs was probably moving prices in the "right" direction--towards the competitive norm. See J. Bain, supra, at 14; Areeda & Turner, Predatory Pricing, at 704-07. These considerations have typically led courts to question, and often to forbid, price cuts below "incremental costs," (or "avoidable costs"), while allowing those where the resulting price is higher. See, e.g., Northeastern Telephone Co. v. American Telephone and Telegraph Co., 651 F.2d at 88; International Air Industries, Inc. v. American Excelsior Co., 517 F.2d at 723-24; cf. William Inglis & Sons Baking Co. v. ITT Continental Baking Co., Inc., 668 F.2d 1014, 1035-36 (9th Cir.1981) (price below average variable cost supports "inference" of predatory pricing), cert. denied, 459 U.S. 825, 103 S. Ct. 57, 74 L. Ed. 2d 61 (1982).

In fact, the use of cost-based standards is more complicated than this brief discussion suggests. But, we need not explore here the arguments about how best to measure "incremental" or "avoidable" costs (e.g., whether "average variable cost" is an appropriate surrogate). Nor need we consider the theoretical difficulties that arise when prices fall between "incremental costs" and "average (total) costs," a circumstance that can arise either when production is at a level below full capacity and the firm lowers prices to levels that do not cover a "fair share" of fixed costs or when a plant is pushed beyond its "full" capacity at prices that do not cover the specially high costs of the extraordinary production levels. See, e.g., William Inglis & Sons Baking Co. v. ITT Continental Baking Co., 668 F.2d at 1034-36; Pacific Engineering and Production Co. v. Kerr-McGee Corp., 551 F.2d 790, 795-97 (10th Cir.), cert. denied, 434 U.S. 879, 98 S. Ct. 234, 54 L. Ed. 2d 160 (1977); Areeda & Turner, Predatory Pricing, at 709-12; Scherer, Predatory Pricing and the Sherman Act: A Comment, 89 Harv.L.Rev. 869 (1976); Zerbe & Cooper, supra, at 681-84. Here we have a price that exceeds both "average cost" and "incremental cost"--that exceeds cost however plausibly measured. And as to those prices, "virtually every court and commentator agrees" that they are lawful, "perhaps conclusively, but at least presumptively." P. Areeda & D. Turner, Antitrust Law p 711.1c at 118 (1982 Supp.).

Barry points, however, to a possible exception to this rule--an "exception" created by the Ninth Circuit making certain price cuts unlawful even when the resulting revenues exceed total costs. In Inglis, that circuit held that a price cut is unlawful if

the anticipated benefits of defendant's price depended on its tendency to discipline or eliminate competition and thereby enhance the firm's long-term ability to reap the benefits of monopoly power.

668 F.2d at 1035. In the Ninth Circuit's view, prices below "average variable costs" (a surrogate for "incremental costs") produce a presumption of "predatory pricing." When prices exceed "average variable cost," but are "below average total cost," the plaintiff must prove by a preponderance of the evidence that the defendant's pricing policy depends on its exclusionary or disciplinary tendency. And, in a case like this one--a case that the Ninth Circuit would describe as "prices above average total cost"--the plaintiff can still win if it proves "by clear and convincing evidence--i.e., that it is highly probably true--that the defendant's pricing policy was predatory," in the sense defined in Inglis. Transamerica Computer Co. v. International Business Machines Corp., 698 F.2d at 1388.

The virtue of the Ninth Circuit test is that it recognizes an economic circumstance in which even "above total cost" price cutting might not be procompetitive and might, in theory, hurt the consumer. See Transamerica Computer Corp. v. International Business Machines Corp., 698 F.2d at 1387. For instance, if a dominant firm's costs are lower than its competitors', it could use an "above cost" price cut to drive out competition, and then later raise prices to levels higher than they otherwise would be. Moreover, if the price cut meant less profit for the firm unless (1) it drove out competitors and (2) higher prices later followed, the cut might be viewed as lying outside the range of normal, desirable, competitive processes. Even though such a price cut would only injure or eliminate firms that were less efficient than the price-cutter, one could argue that, other things being equal, their continued presence helps the competitive process (say, by constraining price rises) and may lead to greater efficiency in the future. Why should the antitrust laws not forbid this potentially harmful behavior? Indeed, economists have identified this type of pricing behavior (and certain other forms of above-cost pricing behavior) as potentially harmful, see, e.g., Brodley & Hay, Predatory Pricing: Competing Economic Theories and the Evolution of Legal Standards, 66 Cornell L.Rev. 738, 743-46 (1981); Scherer, supra (discussing other, more complex anticompetitive pricing strategies involving above-cost prices); Williamson, Predatory Pricing: A Strategic Welfare Analysis, 87 Yale L.J. 284 (1977) (same).

Nonetheless, while technical economic discussion helps to inform the antitrust laws, those laws cannot precisely replicate the economists' (sometimes conflicting) views. For, unlike economics, law is an administrative system the effects of which depend upon the content of rules and precedents only as they are applied by judges and juries in courts and by lawyers advising their clients. Rules that seek to embody every economic complexity and qualification may well, through the vagaries of administration, prove counter-productive, undercutting the very economic ends they seek to serve. Thus, despite the theoretical possibility of finding instances in which horizontal price fixing, or vertical price fixing, are economically justified, the courts have held them unlawful per se, concluding that the administrative virtues of simplicity outweigh the occasional "economic" loss. See United States v. Trenton Potteries Co., 273 U.S. 392, 47 S. Ct. 377, 71 L. Ed. 700 (1927) (horizontal price fixing); United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 60 S. Ct. 811, 84 L. Ed. 1129 (1940) (same); Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S. Ct. 376, 55 L. Ed. 502 (1911) (vertical price fixing); see also National Society of Professional Engineers v. United States, 435 U.S. 679, 687-92, 98 S. Ct. 1355, 1363-65, 55 L. Ed. 2d 637 (1978); Bok, Section 7 of the Clayton Act and the Merging of Law and Economics, 74 Harv.L.Rev. 226 (1960). Conversely, we must be concerned lest a rule or precedent that authorizes a search for a particular type of undesirable pricing behavior end up by discouraging legitimate price competition. Indeed, it is this risk that convinces us not to follow the Ninth Circuit's approach.

Thus, we believe we should not adopt the Ninth Circuit's exception because of the combined effect of the following considerations. For one thing, a price cut that ends up with a price exceeding total cost--in all likelihood a cut made by a firm with market power--is almost certainly moving price in the "right" direction (towards the level that would be set in a competitive marketplace). The antitrust laws very rarely reject such beneficial "birds in hand" for the sake of more speculative (future low-price) "birds in the bush." To do so opens the door to similar speculative claims that might seek to legitimate even the most settled unlawful practices. (Should a price-fixer be allowed to argue that a cartel will help weaker firms survive bad times, leaving them as a competitive force when times are good? Suppose the price-fixer offers to "prove it" by "clear and convincing evidence?" See United States v. Socony-Vacuum Oil Co., 310 U.S. at 220-23, 60 S.Ct. at 842-44.)

For another thing, the scope of the Ninth Circuit's test is vague. Is it meant to include, for example, "limit pricing"--the common practice of firms in concentrated industries not to price "too high" for fear of attracting new competition? See Transamerica Computer Corp. v. International Business Machines Corp., 698 F.2d at 1387; Areeda & Turner, Predatory Pricing, at 705-06. The "anticipated benefits" of such a price arguably depend "on its tendency to discipline or eliminate competition" thereby enhancing "the firm's long-term ability to reap the benefits of monopoly power." Does the test mean to include every common instance of a firm (with market power) deciding not to raise its prices? If it means to include either of these sorts of circumstances, the rule risks making of the antitrust laws a powerful force for price increases. But, if the rule does not mean to include these sorts of circumstance, which prices do, and which do not, fall within the test's proscription?

Further, even were the test more specific, it seems to us as a practical matter most difficult to distinguish in any particular case between a firm that is cutting price to "discipline" or to displace a rival and one cutting price "better to compete." No one would condemn a price cut designed to maximize profits in the short run, i.e., by increasing sales at the lower price, not by destroying competition and then raising prices. But the general troubles surrounding proof of firm costs, see, e.g., Transamerica Computer Co. v. International Business Machines Corp., 698 F.2d at 1387; P. Areeda & D. Turner, Antitrust Law paragraphs 711.1d, 715.2 (1982 Supp.), only hint at the difficulty of deciding whether or not a firm's price cut is profit-maximizing in the short-run, a determination that hinges not only on cost data, but also on elasticity of demand, competitors' responses to price shifts, and changes in unit costs with variations in production volume. Direct statements by firm executives concerning their expectations will probably not be found; and, one might ask, in light of uncertain and changing market conditions, how much will the firm itself know? One can foresee conflicting testimony by economic experts, with the eventual determination made, not by economists or accountants, but by a jury. Of course, one might claim that such are the dangers inherent in many antitrust cases. But the consequence of a mistake here is not simply to force a firm to forego legitimate business activity it wishes to pursue; rather, it is to penalize a procompetitive price cut, perhaps the most desirable activity (from an antitrust perspective) that can take place in a concentrated industry where prices typically exceed costs. See United States v. Container Corporation of America, 393 U.S. 333, 89 S. Ct. 510, 21 L. Ed. 2d 526 (1969); O. Hommel Co. v. Ferro Corp., 659 F.2d at 346-47; Pacific Engineering & Production Co. v. Kerr-McGee Corp., 551 F.2d at 796-97; 2 P. Areeda & D. Turner, Antitrust Law paragraphs 404-05; J. Bain, supra, at 118-23; F. Scherer, supra, ch. 5.

Additionally, if private plaintiffs are allowed to attack the "above total cost disciplinary price," we are unlikely to lack for plaintiffs willing to make the effort. After all, even the most competitive of price cuts may hurt rivals; indeed, such may well be its object. And those rivals, if seriously damaged, may well bring suit, hoping or believing they can fit within the Inglis/Transamerica standard.

Finally, we ask ourselves what advice a lawyer, faced with the Transamerica rule, would have to give a client firm considering procompetitive price-cutting tactics in a concentrated industry. Would he not have to point out the risks of suit--whether ultimately successful or not--by an injured competitor claiming that the cut was "disciplinary?" Price cutting in concentrated industries seems sufficiently difficult to stimulate that we hesitate before embracing a rule that could, in practice, stabilize "tacit cartels" and further encourage interdependent pricing behavior. See 2 P. Areeda & D. Turner, Antitrust Law p 404b3; F. Scherer, supra, at 190-93; Turner, Definition of Agreement under the Sherman Act: Conscious Parallelism and Refusals to Deal, 75 Harv.L.Rev. 655 (1962); cf. Hay & Kelley, An Empirical Survey of Price Fixing Conspiracies, 17 J.L. & Econ. 13, 20-24 (1974). This risk could be minimized only if the conditions imposed by the words "clear and convincing evidence" were so stringent that the claim could almost never be proved. But then, one might ask, would the Transamerica "exception" be worth the trouble? See Transamerica Computer Co. v. International Business Machines Corp., 698 F.2d at 1390-91 (Lucas, J., dissenting).

We reiterate that these considerations might not prove sufficient to make the difference were we dealing with a price that, although above average total costs, was below incremental costs--a price that in the absence of special circumstances proves unsustainable. But, here we deal with a price that is above both incremental and total cost.

In sum, we believe that such above-cost price cuts are typically sustainable; that they are normally desirable (particularly in concentrated industries); that the "disciplinary cut" is difficult to distinguish in practice; that it, in any event, primarily injures only higher cost competitors; that its presence may well be "wrongly" asserted in a host of cases involving legitimate competition; and that to allow its assertion threatens to "chill" highly desirable procompetitive price cutting. For these reasons, we believe that a precedent allowing this type of attack on prices that exceed both incremental and average costs would more likely interfere with the procompetitive aims of the antitrust laws than further them. Hence, we conclude that the Sherman Act does not make unlawful prices that exceed both incremental and average costs.

Even if we are wrong, however, and Inglis and Transamerica contain the correct legal standard, Barry would fail. Barry has not demonstrated by "clear and convincing evidence" that Pacific offered Grinnell an additional 5 or 10 percent price discount only because it wished to keep Barry out of the market. Rather, Pacific testified that the additional discount was related to additional cost savings. Grinnell's firm order for snubbers in 1977 was larger than the total volume of Pacific sales of snubbers in any previous year. Pacific had excess snubber capacity. The price discount, by securing the firm order, allowed Pacific to operate this capacity more efficiently, saved Pacific money, and thereby produced more profit than a higher price (without the firm order) could have done, without regard to any impact on Barry. At least there is evidence this was so--that this was the sort of price cut a firm would have made under competitive conditions in a fully competitive industry. And, the evidence in the record to this effect precludes the possibility of "clear and convincing evidence" to the contrary. Thus, even under the Transamerica test, Pacific's price cut would not be found anticompetitive or exclusionary.

Barry argues next that Pacific's "requirements contract" with Grinnell was exclusionary, that it was more restrictive of competition than legitimate business considerations could justify. The district court, however, disagreed. And so do we.

The antitrust problem that courts have found lurking in requirements contracts grows out of their tendency to "foreclose" other sellers from the market by "tying up" potential purchases of the buyer. Arguably, under certain circumstances substantial foreclosure might discourage sellers from entering, or seeking to sell in, a market at all, thereby reducing the amount of competition that would otherwise be available. Of course, the connection between "foreclosure" and lessened entry is somewhat distant, since it depends on how potential competitors perceive the foreclosure's likely effects on their opportunities. Moreover, virtually every contract to buy "forecloses" or "excludes" alternative sellers from some portion of the market, namely the portion consisting of what was bought. It is not surprising, then, that courts have judged the lawfulness of contracts to purchase not under per se rules but under a "rule of reason." Tampa Electric Co. v. Nashville Coal Co., 365 U.S. 320, 333, 81 S. Ct. 623, 631, 5 L. Ed. 2d 580 (1961); Bob Maxfield, Inc. v. American Motors Corp., 637 F.2d 1033, 1036 (5th Cir.), cert. denied, 454 U.S. 860, 102 S. Ct. 315, 70 L. Ed. 2d 158 (1981); United States v. American Can Co., 87 F. Supp. 18, 31 (N.D.Cal.1949). And this is so whether those contracts involve a buyer's promise to buy "exclusively" from the seller, see, e.g., Bob Maxfield, Inc. v. American Motors Corp., supra; American Motor Inns, Inc. v. Holiday Inns, Inc., 521 F.2d 1230, 1246-53 (3d Cir.1975), a promise to buy all his "requirements" from the seller for a specified time, see, e.g., Tampa Electric Co. v. Nashville Coal Co., supra, or a promise to buy a large dollar amount, see, e.g., Magnus Petroleum Co. v. Skelly Oil Co., 599 F.2d 196 (7th Cir.), cert. denied, 444 U.S. 916, 100 S. Ct. 231, 62 L. Ed. 2d 171 (1979); White & White, Inc. v. American Hospital Supply Corp., 540 F. Supp. 951, 1032 (W.D.Mich.1982). See Perington Wholesale, Inc. v. Burger King Corp., 631 F.2d 1369, 1374 (10th Cir.1979); 3 P. Areeda & D. Turner, Antitrust Law p 732a. Thus, in determining "the probable effect of the contract on the relevant area of effective competition," Tampa Electric Co. v. Nashville Coal Co., 365 U.S. at 329, 81 S.Ct. at 629, we are to take into account both the extent of the foreclosure and the buyer's and seller's business justifications for the arrangement. Id. at 333-34, 81 S.Ct. at 631. We must look both to the severity of the foreclosure (a fact which, other things being equal, suggests anticompetitive harm) and the strength of the justifications in determining whether the "size" of the contract to purchase is reasonable.

Barry would like to characterize the agreements before us as consisting of a Grinnell promise to buy all its snubber requirements from Pacific for three years. And, if Barry correctly describes Grinnell as accounting for half the snubber market, this characterization suggests a three-year "foreclosure" of 50 percent of the relevant market. In terms of the case law, this sounds like a significant foreclosure. See, e.g., Standard Oil Co. v. United States, 337 U.S. 293, 69 S. Ct. 1051, 93 L. Ed. 1371 (1949); Lessig v. Tidewater Oil Co., 327 F.2d 459, 468 (9th Cir.), cert. denied, 377 U.S. 993, 84 S. Ct. 1920, 12 L. Ed. 2d 1046 (1964); Mytinger & Casselberry, Inc. v. FTC, 301 F.2d 534 (D.C.Cir.1962). But this description considerably overstates the size of the foreclosure and its likely anticompetitive effect for several reasons.

First, Grinnell did not actually promise to buy all its requirements from Pacific; it entered into a contract for a fixed dollar amount. There is "an important difference between a 'requirements' contract and a contract which calls for the purchase of a definite quantity over a period of time which the buyer estimates to be sufficient to meet his requirements." Tampa Electric Co. v. Nashville Coal Co., 276 F.2d 766, 771 (6th Cir.1960), rev'd on other grounds, 365 U.S. 320, 81 S. Ct. 623, 5 L. Ed. 2d 580 (1961). A true requirements contract flatly eliminates the buyer from the market for its duration; a fixed quantity contract leaves open the possibility that the buyer's needs will exceed his contractual commitment; he is free to purchase from others any excess amount that he may want. This flexibility is important here, for it left Grinnell the legal power to buy small (and then in 1979, larger) amounts from Barry should they have become available.

Second, the Grinnell-Pacific contract was not a single three-year agreement. The district court found that the contract consummated in late January, 1977, bound Grinnell to take $4.3 million worth of snubbers--its estimated needs for the next twelve months. Grinnell then separately promised--in May 1977--to buy $6.9 million worth of snubbers for 1978 (its estimated requirements) and $5 million for 1979 (considerably less than its estimated requirements). Even if we view the 1978 and 1979 purchases as a single agreement, the scope of this agreement's preclusive effect then extended over something less than Grinnell's expected requirements and lasted about two years.

Third, Grinnell's contract with Barry suggests that a snubber buyer typically places orders for snubbers well in advance of expected delivery. In fact, normally Grinnell would be required to place orders with Barry (once Barry developed the product) at least six months before delivery. Moreover, the record demonstrates the long delay that any new entrant would have to anticipate between the time it decided to enter the snubber market and the time it would be ready to deliver a product. Under these circumstances, a Grinnell decision to buy a year or two's worth of snubbers from Pacific at one shot instead of from time to time seems likely, as a practical matter, to have had, at most, limited anticompetitive effects.

At the same time, the record suggests the existence of legitimate business justifications for the agreements from the perspectives of both buyer and seller. For Grinnell, the contracts guaranteed a stable source of supply, and, perhaps, more important, they assured Grinnell a stable, favorable price. For Pacific, they allowed use of considerable excess snubber capacity; and they allowed production planning that was likely to lower costs. (At least Pacific executives testified as much and the record contains no refutation of this testimony.)

Finally, Grinnell is not a small firm that Pacific could likely bully into accepting a contract that might foreclose new competition. See Tampa Electric Co. v. Nashville Coal Co., 365 U.S. at 329, 81 S.Ct. at 629; cf. Standard Oil Co. v. United States, 337 U.S. at 308, 69 S.Ct. at 1059; Lessig v. Tidewater Oil Co., 327 F.2d at 467; United States v. American Can Co., supra. To the contrary, it was Grinnell, not Pacific, that sought the extensions for 1978 and 1979. Moreover, Grinnell had every interest in promoting new competition. It agreed to provide Barry with several hundred thousand dollars expressly to develop a new source of supply. Grinnell could have obtained snubbers without placing such large orders had it given up the "special" extra 5 to 10 percent price discount, a matter of a few hundred thousand dollars per year. Had Grinnell believed that the long-term nature of the contracts significantly interfered with new entry, or inhibited the development of a new source of supply, it is difficult to understand why it would have sought the agreements.

In sum, in light of the nature of the contracts and the market, their fairly short time period, their business justifications, the characteristics of the parties, and their likely motives as revealed by their business interests, we believe that the district court could reasonably have concluded they were not "exclusionary."

Barry also argues that the "noncancellation" clauses were exclusionary. The clauses stated that "quantities [mentioned in the contract] ... are considered to be minimum with pricing based accordingly. For this reason, full cancellation charges would apply if all or any portion of the requirements were cancelled or rescheduled by ITT Grinnell beyond [the contract period]." Those clauses, as the parties (and, we think, as the district court) interpreted them, would require Grinnell to pay the entire price of the yearly order (i.e. $4.3 million for 1977, $6.9 million for 1978, etc.) whether Grinnell took all, some or none of the snubbers that the order covered. The noncancellation clause thus acted as a powerful economic incentive for Grinnell to stick to its bargain; and Barry says this incentive was "too powerful" to the point where it is "exclusionary."

Of course, the parties to a contract have a right to collect damages for breach. And, they can insert a liquidated damages provision in the contract as a measure of damages. See U.C.C. Sec. 2-718. Yet, "[t]he central objective behind the system of contract remedies is compensatory, not punitive." Restatement (Second) of Contracts Sec. 356 comment (a) (1979). Thus, a liquidated damages provision is enforceable only if it "is reasonable in the light of the anticipated or actual harm caused by the breach, the difficulties of proof of loss, and the inconvenience or non-feasibility of otherwise obtaining an adequate remedy. A term fixing unreasonably large liquidated damages is void as a penalty." U.C.C. Sec. 2-718(1). If the cancellation clause provides only for lawful liquidated damages, it is no more "exclusionary" than the underlying substantive agreement that it helps to enforce. But Barry argues that the cancellation clause is a "penalty" and that it therefore has no legitimate place in the substantive agreement.

We shall assume for the sake of argument that Barry is correct. If so, the clause in principle might have an unjustified anticompetitive effect. It might discourage a buyer from pursuing a course of action otherwise open to him under the law of contracts, namely to breach the purchase agreement, to pay damages, and to buy from a new entrant instead. While the presence of the clause could not legally forbid this course of action--as it would be unenforceable as a penalty at law--its presence does still threaten the buyer with the lawsuit that would be needed to prove that it is unenforceable. And it is this threat, and the consequent additional deterrence to the "breach and pay damages" course of action that constitutes the "unreasonably anticompetitive" aspect of the clause. Cf. United States v. United Shoe Machinery Corp., 110 F. Supp. 295, 320, 344 (D.Mass.1953) (lease provisions deterring early terminations are exclusionary), aff'd per curiam, 347 U.S. 521, 74 S. Ct. 699, 98 L. Ed. 910 (1954).

This argument, while logical, strikes us as of virtually no practical importance in this case. Even if one heroically assumed that Grinnell might have wished to breach and to buy elsewhere in 1977, 1978 or 1979, it is virtually impossible to believe that the presence of this clause could have stopped it from doing so. Given Grinnell's size and the competence of its legal staff, it is most unlikely to have been deterred by Pacific's assertion of unusually high damages resting upon a legally invalid provision in the contract. (And, if the provision is not legally invalid--that is, if it does reasonably reflect Pacific's likely actual damages--then it is not, from an antitrust perspective, unreasonable). This is simply to say that the anticompetitive consequence to which Barry might point is too remote, too speculative in the context of this case, to warrant classifying the clause as significantly anticompetitive. And, the district court's conclusion that its presence did not transform otherwise lawful purchase agreements into unlawful, exclusionary ones, is adequately supported.

If Pacific's challenged conduct is not "exclusionary," for purposes of Sherman Act Sec. 2, then a fortiori, it does not violate the other provisions of law that Barry cites. A monopolist's conduct that from a competitive point of view is unreasonable, is "exclusionary." California Computer Products, Inc. v. International Business Machines Corp., 613 F.2d 727, 735-36 (9th Cir.1979); 3 P. Areeda & D. Turner, Antitrust Law p 626 at 83; see United States v. United Shoe Machinery Corp., 110 F. Supp. at 342. Thus, conduct that is not "exclusionary" is not "unreasonable." And we do not see how ordinarily that conduct could be forbidden by Sherman Act Sec. 1's prohibition of "unreasonable" restraints of trade. See Standard Oil Co. v. United States, 221 U.S. 1, 61-64, 31 S. Ct. 502, 516-517, 55 L. Ed. 619 (1911) (Sherman Act Sec. 1's "restraint of trade" language interpreted according to "rule of reason"). We see no conduct here that is subject to any section 1-based rule of "per se" illegality. See, e.g., United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 60 S. Ct. 811, 84 L. Ed. 1129 (1940) (price fixing); Addyston Pipe & Steel Co. v. United States, 175 U.S. 211, 20 S. Ct. 96, 44 L. Ed. 136 (1899), aff'g 85 F. 271 (6th Cir.1898) (market division). Barry cites Engine Specialties, Inc. v. Bombardier Ltd., 605 F.2d 1 (1st Cir.1979), cert. denied, 446 U.S. 983, 100 S. Ct. 2964, 64 L. Ed. 2d 839 (1980), in an effort to find precedent for application of a per se rule, but that case involved competitors who explicitly divided territories--a circumstance absent here.

Nor can Barry show tortious interference with Barry's Grinnell contract. Under Massachusetts law, one essential element of the tort is that defendant's conduct be "either per se unlawful" or "conducted with malice." Araserv, Inc. v. Bay State Harness Horse Racing and Breeding Association, Inc., 437 F. Supp. 1083, 1094 (D.Mass.1977) (quoting Old Colony Donuts, Inc. v. American Broadcasting Companies, 368 F. Supp. 785, 787 (D.Mass.1974)). The only basis for claiming "per se" unlawfulness that we can find is "antitrust" unlawfulness--a matter that we have disposed of.

Barry cannot show "malice" without showing that Pacific knew its conduct "would result in injury to [Barry's] contract rights." Goldsmith v. Traveler Shoe Corp., 236 Mass. 111, 116, 127 N.E. 606 (1921); cf. Beekman v. Marsters, 195 Mass. 205, 212, 80 N.E. 817 (1907) (showing of knowledge of contract necessary to obviate showing of express malice). And the district court correctly found the evidence here insufficient. The trial court found that, although Pacific officers became aware, by September, 1976, of the existence of an agreement between Barry and Grinnell for the development of a mechanical snubber, they did not know the terms of that agreement. Moreover, the terms of the Barry-Grinnell contract permitted Grinnell to purchase from other suppliers if "Barry is unable to fulfill required delivery schedules." This provision at least arguably could have protected Grinnell from a breach of contract claim by Barry on account of Grinnell's purchases from Pacific; and the simple possibility that it would have done so precludes attribution to Pacific of knowledge that Grinnell's purchases from it would breach the Barry-Grinnell contract. Similarly, the trial court found that Barry was in breach at the time of the first Pacific-Grinnell agreement. Even if we assume purely for the sake of argument that it was wrong, the claim of breach is strong enough to preclude the requisite finding of knowing interference by Pacific. In short, while Pacific might have been aware that its sales to Grinnell would undermine Barry's prospects for its new line of snubbers, there was no basis for concluding that Pacific knew that those sales would interfere with Barry's contractual rights.

Affirmed.