GARWOOD, Circuit Judge:

Plaintiff-appellant Stearns Arp or t Equipment Co., Inc. (Stearns) brought this suit against defendant-appellee FMC Corporation (FMC), claiming FMC had violated the Sherman Act, the Robinson-Patman Act, and Texas state law. Stearns appeals the district court’s grant of summary judgment to FMC, and also challenges certain expenses awarded to FMC as costs. We affirm.

Facts and Proceedings Below

Stearns and FMC are both manufacturers of boarding bridges, the devices that allow passengers to enter and exit passenger airplanes. Historically, the domestic market has been dominated by Jetway, a brand previously produced by a division of a company not a party to this case. In 1994, the Jetway division was purchased by FMC, which continued its operation. Stearns, a wholly-owned subsidiary of Trinity Industries, has been producing bridges since the beginning of the 1980s. Both parties export their bridges around the world, and about a dozen manufacturers produce bridges abroad. *521 While foreign competitors have bid on some projects and sold a handful of bridges here, during the relevant time frame actual foreign penetration in the North American market was minimal. The record does show that foreign producers sporadically expressed interest in the market, and one recently opened up a sales office in the United States.

FMC and Stearns utilize competing technologies in their bridges. Stearns relies on hydraulic systems for its bridges, while FMC uses an electromechanical system. The record establishes that at least some bridge purchasers felt that there were substantial differences between the two systems under various circumstances. In addition, FMC was in the process of developing and introducing computerized controls in some of its models, called “smart bridges,” during the relevant time frame. The “smart-bridge” technology — which had some teething troubles — was significantly different from the mechanism used by Stearns.

. Prior to the mid-1980s, the dominant purchasers of bridges in the United States had been airlines. The airlines had frequently dealt exclusively with Jetway. However, during that period the market began to shift and municipal airport authorities became the primary purchasers of bridges. This shift led to most sales in the industry being governed by competitive bid processes. After some initial successes in this new market, Steams began to lose market share to FMC. Stearns alleges that its loss of sales to municipal bidders was the product not of vigorous competition, but rather of an orchestrated program by FMC to avoid fair competition through a combination of exclusionary manipulation of municipal bids and predatory pricing.

Stearns filed an antitrust action against FMC on December 4, 1995. The complaint initially alleged violation of Sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1-2, Section 2(a) of the Robinson-Patman Act, 15 U.S.C. § 13(a), and tortious interference and unfair competition under state law. The district court granted FMC’s motion for summary judgment on the Section 1 Sherman Act claims on May 31, 1996. See Stearns Airport Equipment Co., Inc. v. FMC Corp., 977 F.Supp. 1263 (N.D.Tex.1996). Stearns does not appeal that ruling. Discovery continued on the other claims, and FMC filed another motion for summary judgment on December 20, 1996. Stearns requested an extension of time for its response, which was granted, and also filed a motion under Rule 56(f) to delay summary judgment until the completion of discovery. The district court denied the Rule 56(f) motion, but allowed discovery to continue until March 26, 1997, when it granted summary judgment to FMC on all claims. Stearns moved to reconsider and offered additional evidence. This motion was denied and this appeal followed.

I. Standard of review.

We review a district court’s grant of summary judgment employing the same standard it was required ' to apply in granting the motion. Dutcher v. Ingalls Shipbuilding, 53 F.3d 723, 725 (5th Cir.1995). Summary judgment must be affirmed when the moving party has identified material facts not in genuine dispute and the nonmoving party fails to produce or identify in the record summary judgment evidence sufficient to sustain a finding in its favor respecting such of those facts as to which it bears the trial burden of proof. In reviewing the record, we must view all facts in the light most favorable to the nonmovant. We review questions of law de novo. Id. We no longer maintain that summary judgment is especially disfavored in categories of eases. See Little v. Liquid Air Corporation, 37 F.3d 1069, 1075 n. 14 (5th Cir.1994) (en banc) (“we reject any suggestion that the appropriateness of summary judgment can be determined by the case classification.”). Stearns’ attempt to invoke earlier cases in which we suggested that summary judgment should be shunned when complex antitrust claims are involved thus fails.

Stearns on this appeal treats its Robinson-Patman and state law claims as derivative of its Sherman Act section 2 claim. Accordingly, if we find that summary judgment should be affirmed on the Section 2 claims, we must also affirm the dismissal of these claims.

*522 II. Exclusionary Conduct

A violation of section 2 of the Sherman Act is made out when it is shown that the asserted violator 1) possesses monopoly power in the relevant market and 2) acquired or maintained that power wilfully, as distinguished from the power having arisen and continued by growth produced by the development of a superior product, business acumen, or historic accident. United States v. Grinnell Corp., 384 U.S. 563, 86 S.Ct. 1698, 1704, 16 L.Ed.2d 778 (1966). For the purpose of this summary judgment, we will assume, as the district court did, that FMC does possess monopoly power in the North American market for boarding bridges. Ex clusionary conduct under section 2 is the creation or maintenance of monopoly by means other than the competition on the merits embodied in the Grinnell standard. See Aspen Skiing Co. v. Aspen Highlands, 472 U.S. 585, 105 S.Ct. 2847, 2859, 86 L.Ed.2d 467 (1985) (attempting to exclude on grounds other than efficiency); C.E. Services, Inc. v. Control Data Corporation, 759 F.2d 1241, 1247 (5th Cir.1985) (quoting 3 P. Areeda and D. Turner, Antitrust Law p. 626, at 83 (1978)). The key factor courts have analyzed in order to determine whether challenged conduct is or is not competition on the merits is the proffered business justification for the act. If the conduct has no rational business purpose other than its adverse effects on competitors, an inference that it is exclusionary is supported. See Aspen, 105 S.Ct. at 2860 (finding failure to offer persuasive business justification “most significant”). Summary judgment is appropriate in some cases where defendant’s business justification is unchallenged. See Bell v. Dow Chemical Co., 847 F.2d 1179, 1185-86 (5th Cir.1988) (in a refusal-to-deal case, rejecting contention that Aspen’s procedural posture indicated that business justification was a matter for the jury but going on to reject the proffered justification).

Stearns contends that FMC, threatened by the switch of purchasing from the airlines to municipal airport authorities, adopted a plan to avoid competition on the merits, and specifically competition on price. The heart of this alleged plot is contained in an FMC presentation directed to its marketing and sales personnel. The presentation urged FMC’s employees to use four strategies in pursuing sales to municipalities. First, FMC was to attempt to convince municipalities that they should avoid competitive bidding and strike a purchase agreement with FMC directly — so called “sole-sourcing.” Second, if bidding appeared inevitable, FMC should strive to drive the criteria for the award away from price alone by requesting various product features be weighted against cost in the final calculation of the best bid. Third, efforts were to be made to insure that the specifications adopted by a municipality were tailored to fit FMC’s product and exclude Steams. Lastly, FMC would “induce complexities in the bidding process” by suggesting certain certifications and restrictions be added that worked to the detriment of Stearns. 1 Taken together, Stearns argues that these strategies constituted a deliberate plan to exclude Stearns from competing in the municipal bridge market, thus harming consumers by robbing them of a true competitive process.

The key point uniting these allegations is that they all involved FMC’s attempts to persuade buyers to favor their product prior to the actual bid. Courts that have considered whether attempts to convince independent government purchasers to adopt specifications in their favor prior to bidding are a violation of the antitrust laws have uniformly found such behavior not to be a violation. The Ninth Circuit, presented with a claim that a monopolist’s contacts with county officials and architects led to the specification of its product prior to a bid rejected the contention that such contacts violated the Sherman Act. Security Fire Door Co. v. County of Los Angeles, 484 F.2d 1028, 1030-31 (9th Cir.1973). The Security Fire Door court found that there had been no injury to competition through these contacts since the *523 competitor was free to engage in similar persuasive efforts with the relevant officials. Competition on the merits was assured as long as the plaintiff had been “free to tout the virtues of his particular [product] in an effort to secure favorable specifications.” Id. at 1031. Other courts have agreed with this reasoning. See Richard Hoffman Corp. v. Integrated Building Systems, 610 F.Supp. 19, 23 (N.D.Ill.1985) (reversing prior determination that contractor had violated antitrust laws by specifying product it distributed when drawing up specifications used by the county after it was shown competitor had ample opportunity to challenge specification and tout the virtues of its product); Super-Turf, Inc. v. Monsanto Co., 660 F.2d 1275, 1280 (8th Cir.1981) (“Even if one accepts SuperTurfs argument that the adoption of one product’s specifications precludes further competition, it is also true that SuperTurf is free to press for the adoption of its own product specifications.”); Triple M Roofing Corp. v. Tremco, Inc., 753 F.2d 242, 246 (2nd Cir.1985) (The court found that defendant’s efforts to inform government of its product, which led to its being specified by brand in the contract, “exemplified those expected of an aggressive sales representative.” It noted that these activities “promote rather than hinder competition.”); 2 Whitten v. Paddock Pool Builders, Inc., 508 F.2d 547, 558 (1st Cir.1974) (endorsing lower court’s finding that efforts to convince architects to include propriety specifications in a contract was simply “salesmanship”).

While all of these cases involved section 1 of the Sherman Act rather than section 2, and several also were in a different procedural posture than we face here, their logic properly applies to our inquiry. Under Aspen, we ask in section 2 exclusionary conduct cases whether the challenged conduct involved competition on the merits. Security Fire Door and its kin clarified that, in the municipal bidding context, permissible competition is not restricted to the bid itself but can also occur in the process of “selling” specifications and contract forms, when companies “tout the virtues” of their product. The choice of the consumer can be expressed in specifications as well as the final bid. See Security Fire Door, 484 F.2d at 1031. We therefore will examine FMC’s behavior throughout the municipal contracting process to determine whether it relied on a superior product or business acumen in pursuing its goals, or had recourse to methods beyond competition on the merits.

The first point that separates FMC’s behavior in the contracting process from section 2 cases of exclusionary conduct is its economic rationality. Generally, a finding of exclusionary conduct requires some sign that the monopolist engaged in behavior that— examined without reference to its effects on competitors — is economically irrational. When there is no other possible explanation for an action, there is a strong inference that it was taken for the purpose of harming competitors rather than otherwise advancing the monopolist’s business. For example, in the leading modern case on exclusionary conduct, Aspen Skiing, two companies ran ski lifts on several different mountains in the same resort area.. Traditionally the companies had honored ski passes that were good on all mountains. The larger company stopped honoring the joint passes, instead setting up tickets that only covered its mountains. This decision violated long-standing industry practice and “infuriated” the larger mountain’s customers. The Supreme Court found that the defendant “was apparently willing to forgo daily ticket sales” to these customers “because it was more interested in reducing competition .... by harming its smaller competitor” and that the logical inference was that the defendant “was not motivated by efficiency concerns and that it was willing to sacrifice short-run benefits and consumer goodwill in exchange for a per *524 ceived long-run impact on its smaller rival.” Aspen, 105 S.Ct. at 2860, 2861.

In short, Aspen involved a company willingly accepting a real loss because it represented a relative gain. 3 Here, the business justification — independent of harm to competitors — for FMC’s actions is obvious: it was trying to sell its product. While Stearns may feel very much aggrieved at their success, the tactics it complains of were all fairly simple attempts to generate sales by “touting the virtues” of its bridges. “Acts which are ordinary business practices typical of those used in a competitive market do not constitute anti-competitive conduct violative of Section 2.” Trace X Chemical, Inc. v. Canadian Industries, Ltd., 738 F.2d 261, 266 (8th Cir.1984). FMC’s sales efforts produced real, not merely relative, gains for the company. Certainly we have nothing akin to the baffling (until the effect on competitors is examined) request in Great Western that a supplier raise the prices it charged to the monopolist, or the withdrawal of a valued consumer item in Aspen.

The second distinguishing feature of this case is that all of the alleged exclusionary conduct required the active approval of the consumer — the party the Act protects. Aspen and other “refusal-to-deal” exclusionary cases involve a unilateral decision by the monopolist. The consumer has no input on a decision that affects his interest. But here, the decision to sole-source a contract or adopt a particular specification was always ultimately in the hands of the consumer. The record indicates that FMC felt itself obligated to come up with “selling points” to accompany its strategies. Thus when attempting to obtain a sole-sourcing agreement, FMC would stress its technological superiority. Having convinced a municipality on this predicate merits argument, FMC would then argue that sole-sourcing was a cheaper option since a full bidding process involved substantial costs and legal complications. Similarly, when “introducing complexities into the bidding process” FMC agents were told to point out the advantages for municipalities in possible product liability actions if outside certifications were maintained or if independent distributors were barred from bidding. And the record is full of evidence that FMC aggressively touted its electromechanical and “smart-bridge” technology as qualitatively superior to Stearns’ product.

All of these arguments made by FMC to its potential customers may have been wrong, misleading, or debatable. But they are all arguments on the merits, indicative of competition on the merits. To the extent they were successful, they were successful because the consumer was convinced by either FMC’s product or FMC’s salesmanship. FMC — unsurprisingly—wanted to be picked over Stearns on a contract. Also unsurprisingly, for that purpose it calculated carefully what kind of specifications would insure that it would get the contract because Stearns could not bid on a project. 4 But it could not ask municipalities to enter into a sole-sourcing agreement or specify smart-bridge technology merely by asking them to hurt Stearns. FMC had to convince the customer that FMC’s approach was best for the customer, not best for FMC. Inferring an attempt to circumvent competition on the *525 merits is extraordinarily difficult when the alleged violator takes the facially rational and unproblematic step of attempting to sell its product, couches its arguments to the customer in favor of a sale on the merits of the product and procedures it recommends, and the consumer agrees. Without a showing of some other factor, we can assume that a consumer will make his decision only on the merits. To the extent a competitor loses out in such a debate, the natural remedy would seem to be an increase in the losing party’s sales efforts on future potential bids, not an antitrust suit.

The only factor temporarily obscuring the flaws in Stearns’ argument is the municipal bidding context. 5 This is because in the municipal market, bidding statutes generally forbid considerations other than price once a certain point in the process has been reached — municipalities, unlike ordinary consumers, cannot decide at the last minute to purchase a more expensive but higher quality product. But we do not find that the form of these statutes alters the inquiry demanded by Aspen — whether competition is or is not on the merits. Nor do they indicate that “merit” in municipal bidding can only be measured in terms of price. 6 Competition grounded in nonprice considerations such as reliability, maintenance support, and general quality is competition on the merits. The municipal bidding process merely mandates a bifurcation of the consumer’s decision on the merits. During the first, pre-bid stage the municipality must attempt to insure that its nonprice considerations are adequately addressed — and sales efforts at this stage can enlighten a municipal consumer of new advances. See Triple M, 753 F.2d at 246-247 (alternative restorative method would have been unknown to contractor without specification push by supplier, and thus unavailable to ultimate consumer, the State of Georgia). The bidding itself can only resolve a limited portion of the merits — the issue of price.

To be sure, if FMC is successful in its initial efforts, Stearns may be effectively excluded from the final bid. 7 But if FMC fails *526 in persuading officials or Stearns intervenes, FMC’s chief selling point is similarly barred from the final consideration. Municipal contracting will always produce distortions like this. The central insight of Security Fire Door and the other section 1 cases is that jockeying over specifications and bid procedures is a valid form of competition. There is ■ no indication that anything prevented Stearns from doing the necessary research and finding what airports were beginning to prepare a contract, and pushing its arguments at the specifications phase. We decline to find that FMC violated section 2 of the antitrust law by vigorously stressing the qualitative merits of its product during the sole window in which municipalities allowed it to present these nonprice arguments. This behavior was “simple salesmanship” that enhanced rather than subverted competition on the merits. If Stearns was “excluded,” it was excluded by FMC’s superior product or business acumen.

Of course, this conclusion would be called into question if there was evidence that the municipal consumer’s agents had been co-opted by the monopolist to a degree that it could be inferred they were ’ not acting in what they thought was the best interest of the municipality as respects the particular decision being made. Bribery and threats are not competition on the merits. Several cases have found violations of section 2 when a monopolist engages in what appears to be normal competitive behavior, but has manipulated representatives of the consumer to the point that the integrity of the decisional process has been violated. See, e.g., Indian Head, Inc. v. Allied Tube & Conduit Corp., 817 F.2d 938, 947 (2d Cir.1987). Thus courts have found that an exclusionary claim can lie when the monopolist has bribed the officials evaluating the contract. See Buddie Contracting v. Seawright, 595 F.Supp. 422, 425 (N.D.Ohio 1984) (monopolist previously pleaded guilty to criminal charges of unlawful interest in a public contract). And in Indian Head, a manufacturer of traditional metal pipe paid for the enrollment of hundreds of interested individuals in order to “pack” a vote of a building association on whether to approve specification of PVC pipe. Indian Head, 817 F.2d at 947. The bought voters duly blocked the approval of the competitor’s PVC product. The Second Circuit found that this behavior constituted exclusionary conduct — while it might be permissible to argue the ease against PVC before the association, it was impermissible to buy the jury.

Stearns has failed to introduce evidence that the independence of the consumers’ judgment had been tainted, by FMC. To bring the case within the ambit of Indian Head, Stearns must allege that there was a conspiracy or self-interest present strong enough to overcome our assumption that agents will act with the purpose of furthering the interest of their principal. To survive summary judgment, an inference of conspiracy must be backed by evidence that tends to disprove the assumption of independent action. See Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S.Ct. 1348, 1356-57, 89 L.Ed.2d 538 (1986) (discussing standard under section 1 of the Sherman Act). Stearns’ argument is filled with ominous sounding phrases suggesting that lower-echelon officers of the municipalities were “induced” by FMC to adopt its nefarious scheme 8 , and that the airlines were “friends” of FMC and exerted pressure on its behalf. But constant repetition does not alter the fact that Stearns could introduce no evidence of improper, disloyal motive.

Stearns admitted at oral argument that there was no indication that the employees of the consumer were driven by anything other than the desire to obtain the best product possible. As for the airlines, since they depend on the smooth functioning of bridges to service their customers, they naturally expressed their preference to the municipalities. While employees of airlines might indeed be “friends” of any party that could provide them with reliable equipment crucial to their business, there was nothing in the *527 record that indicated that the “friendship” could be traced to anything other than their belief FMC produced a superior product. Cf. Instructional Systems Development Corp. v. Aetna Casualty and Surety Co., 817 F.2d 639, 647 (10th Cir.1987) (evidence indicated company exerted influence on municipality in favor of monopolist as part of agreement to avoid monopolist challenging it in another market). Any influence FMC had over these parties was won because it convinced them of the virtues of its product — it competed on the merits. Certainly we have no allegations of bribery, and nothing here is akin to Indian Head, where the method of competition was more ward boss than businessman.

Ultimately, Stearns does not and cannot claim that it has been excluded from competing on the merits. Every sales pitch and every suggestion that FMC made was evaluated by independent municipal actors who were concerned solely with the merits of the product they were charged with evaluating. Stearns was free to engage in identical tactics and tout the virtues of its product. Steams is really complaining that its municipal consumers keep picking the “wrong” product. Thus it introduces evidence that its technology is sound and FMC’s sales pitches touting its product are misleading. It appears to be assuming that if FMC’s product was not objectively superior, then its victories were not on the merits. But this Court is ill-suited to attempt to judge the relative merits of electromechanical bridges versus hydraulic bridges. That decision is left in the hands of the consumer, not the courts, and to the extent this judgment is “objectively” wrong, the inference is not that there has been a violation of section 2, but rather that the winning party displayed superior business acumen in selling its product. See Triple M, 753 F.2d at 246 (success enjoyed by embedding specifications in municipal contract was “obtained by commercial initiative and skillful marketing”). Competition, even the maintenance of monopoly, through superior business acumen is allowed under section 2. See Grinnell, 86 S.Ct. at 1704. Thus regardless of whether its success can be traced to a “truly” superior product or persuasive business acumen, FMC competed on the merits and has not engaged in exclusionary conduct. -

In the same vein, Stearns attempts to justify its request that we overrule the consumer’s verdict by claiming that these municipal consumers are too unsophisticated to make an unguided decision. Again, this Court is ill-suited to attempt to judge the competence of municipal purchasing agents. Further, while there was evidence that FMC believed that the municipalities were less sophisticated purchasers than the airlines had been, one of Stearns’ main complaints was that the sophisticated airlines weighed in against it as well. To the extent municipal officers may have lacked perfect information, Stearns could have supplied them with the missing perspective by matching FMC’s sales efforts. The municipal authorities in question must be treated as if they were capable of running the construction of multi-million dollar airports.

III. Predatory Pricing.

Stearns’ other contention alleges that when the exclusionary strategy discussed above failed to work, FMC resorted to predatory pricing of its bridges. Once Stearns has been vanquished by this combined attack, it is claimed FMC will use its monopoly power to raise prices again. We find Stearns’ evidence of the existence of a predatory pricing scheme unconvincing, both because there has been an inadequate showing that any under-cost bids occurred and because there has been no showing that recoupment of the putative FMC losses is possible in the bridge market.

The Supreme Court has expressed extreme skepticism of predatory pricing claims. The central difficulty with such actions is that the conduct alleged is difficult to distinguish from conduct that benefits consumers. See Matsushita, 106 S.Ct. at 1360 (“[CJutting prices in order to increase business often is the very essence of competition. Thus, mistaken inferences in cases such as this one are especially costly, because they chill the very conduct the antitrust laws are designed to protect.”). Moreover, the Court has noted the consensus among economists that such schemes are difficult if not impossible to *528 successfully complete and thus unlikely to be attempted by rational businessmen. See id. at 1357 (“[T]here is a consensus among commentators that predatory pricing schemes are rarely tried, and even more rarely successful.”); Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 113 S.Ct. 2578, 2590, 125 L.Ed.2d 168 (1993) (“general implausibility of predatory pricing”); Car gill, Inc. v. Monfort of Colorado, 479 U.S. 104, 107 S.Ct. 484, 495 n. 17, 93 L.Ed.2d 427 (1986) (“Although the commentators disagree as to whether it is ever rational for a firm to engage in such conduct, it is plain that the obstacles to the successful execution of a strategy of predation are manifold, and that the disincentives to engage in such a strategy are accordingly numerous.”)

Accordingly, the standard for inferring an impermissible predatory pricing scheme is high. To succeed, such a claim must demonstrate both that 1) the prices complained of are below an appropriate measure of the alleged monopolist’s costs and 2) that the alleged monopolist has a reasonable chance of recouping the losses through below-cost pricing. Brooke Group, 113 S.Ct. at 2587-88 (establishing unitary standard that includes section 2 claims). 9 In other words, before a violation is found, a claimant must be able to demonstrate both that there has been a specific incident of underpricing and that the claimed scheme makes economic sense. Of course, the Court’s skepticism towards these claims has not altered the standards for summary judgment. See Eastman Kodak Co. v. Image Technical Serv., Inc., 504 U.S. 451, 112 S.Ct. 2072, 2083, 119 L.Ed.2d 265 (1992). But the standard adopted by the Court incorporates this skepticism, and to survive summary judgment a plaintiff must have evidence that the predation scheme is economically rational. See id. (“If the plaintiffs theory is economically senseless, no reasonable jury could find in its favor, and summary judgment should be granted.”); Advo, Inc. v. Philadelphia Newspapers, Inc., 51 F.3d 1191, 1196-97 (3rd Cir.1995). We find that Stearns has failed to present evidence to satisfy either prong of this test.

A. Recoupment

We begin by examining the possibility of recoupment. This inquiry is really into the economic rationality of the challenged conduct. If there is no likelihood of recoupment, it would seem improbable that a scheme would be launched. Given the high error cost of finding companies liable for cutting prices to the consumer, the court should thus refuse to infer predation. See Matsushita, 106 S.Ct. at 1360 (summary judgment is appropriate if recoupment is unlikely and the monopolist thus had no motive to engage in the alleged activity unless proper direct evidence of the scheme is introduced). To achieve the recoupment requirement of Brooke Group, a claimant must meet a two-prong test. First, a claimant must demonstrate that the scheme could actually drive the competitor out of the market. Sec *529 ond, there must be evidence that the surviving monopolist could then raise prices to consumers long enough to recoup his costs without drawing new entrants to the market. Brooke Group, 113 S.Ct. at 2589.

1. Possibility of eliminating Stearns

In examining whether an alleged scheme could actually succeed in eliminating a competitor, we must look to “the extent and duration of the alleged predation” and the parties’ relative strength. Brooke Group, 113 S.Ct. at 2589. Stearns has only attempted to introduce evidence of underpricing in five bids spread out over four years, and in four of these cases the bids were for only two bridges apiece. A Stearns executive admitted that there are an average of 60-100 bidding opportunities — usually for many more than two bridges — annually in the domestic market alone. Yet Stearns contends that this rare and intermittent underpricing could somehow bring it to its knees. It certainly has not yet — Stearns remains in the market — and it is difficult to see how these rare, isolated incidents could have a serious effect on its health. Boarding bridges constitute only forty percent of Stearns’ business — it also produces baggage-handling equipment — and its corporate parent is a strong company.

Stearns claims that “even small amounts of predation are not permissible under the antitrust laws.” This is true in an abstract sense, but in applying the concept to this case Stearns completely ignores the insight of Matsushita and Brooke Group — unless there is a showing of reasonably possible success using the scheme, there is no predation. “If market circumstances or deficiencies in proof would bar a reasonable jury from finding that the scheme alleged would likely result in sustained supraeompetitive pricing, the plaintiffs case has failed.” Brooke Group, 113 S.Ct. at 2589. If FMC’s pricing cannot drive Stearns out of the market, then it will never have a chance to charge supraeompetitive prices, let alone sustain those levels.

Stearns relies on a case in which this Court found allegations of predation involving only five service contracts could survive summary judgment. C.E. Services, Inc. v. Control Data Corp., 759 F.2d 1241, 1247 (5th Cir.1985). We initially note that this case was decided under different legal standards. Not only did Control Data predate the Supreme Court’s renewed examination of predation claims that began in Matsushita, but it also came before the Court’s contemporaneous clarification of the summary judgment threshold. See, e.g., Celotex Corp. v. Catrett, 477 U.S. 317, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986); Liquid, Air Corporation, 37 F.3d at 1075 (acknowledging this Circuit’s incorrect application of Rule 56 prior to Celotex). But to the extent Control Data remains persuasive, it highlights Stearns’ failure to meet its burden under Brooke Group. While Control Data involved predation in only five contracts, these were the plaintiffs only service contracts at the time. The loss of these contracts led to the plaintiffs immediate bankruptcy. Control Data, 759 F.2d at 1243. Here, of course, only five bid opportunities are involved, in a market where 240-400 such chances were available during the alleged predation. Stearns is citing a case in which one hundred percent of the relevant bids were below-cost to support an allegation that below-cost bidding in — at most — two percent of the bids constituted predation.

Stearns’ response is to plead that the predation must be understood in light of the exclusionary conduct — predation is FMC’s backup strategy. But this begs the question. If FMC had engaged in exclusionary conduct, Stearns would win in any case. But as discussed above, we find that the challenged conduct is permissible. Stearns has introduced no evidence that its survival is threatened by the sales lost to the rare, sporadic predation that it alleges, and does not claim that the continuation of below-cost bids at this level — two percent at most — will drive it out of business. Instead, it claims that it faces ruin because of FMC’s pursuit of tactics we have found unobjectionable, coupled with the alleged rare predation. We decline to find predation when there has been no showing that the alleged below-cost pricing campaign was of a sufficient duration and extent to independently force Stearns out of *530 the market and there has been no other valid claim of antitrust-impermissible conduct.

2. Barriers to entry

Even if Stearns had advanced evidence that the alleged predation could drive it out of the market, it has failed to meet the second prong of the recoupment test. A competitor must be able to not only eliminate its competitors through predation, but also be able to maintain supracompetitive prices long enough to recoup the losses it incurred in the predation campaign. If barriers to entry in an industry are low, new entrants into the industry will appear when the monopolist raises its prices, and the net effect of the campaign will be a loss to the predator and a windfall for consumers who paid the subcompetitive predatory price. See C.A.T. Industrial Disposal, Inc. v. Browning-Ferris Industries, Inc., 884 F.2d 209, 211 (5th Cir.1989); Advo, 51 F.3d at 1200 (“Such futile below-cost pricing effectively bestows a gift on consumers, and the Sherman Act does not condemn such inadvertent charity.”).

Stearns claims that recoupment will be possible because the industry’s high barriers to entry prevent the emergence of new challengers if FMC succeeds in disposing of Stearns. This would seem rather difficult to credit. The record shows that a large number of foreign firms produce bridges, and one of the most recently successful was formed only in the 1980s. Putting aside the lack of evidence that new American entrants could not take up the standard if Stearns falters, it would seem probable that these already established foreign manufacturers would leap into the United States market if FMC began to charge supracompetitive prices. There was no evidence of special industry conditions such as special tariffs or domestic purchase limitations on municipalities that would block such an entry.

Stearns asks the court to infer the existence of entry barriers from the historical lack of success foreign firms have had in the domestic market. This is irrelevant. “In evaluating entry barriers in the context of a predatory pricing claim, however, a court should focus on whether significant entry barriers would exist after the merged firm had eliminated some of its rivals, because at that point the remaining firms would begin to charge 'supracompetitive prices, and the bar-riersbhat existed during the competitive conditions might well prove insignificant.” Car-gill, 107 S.Ct. at 494 n. 15. Once FMC’s alleged plot has succeeded, according to Stearns’ logic it will raise prices. The question is what will stop foreign firms from appearing on the scene, pointing out to municipalities the supracompetitive prices, and providing an alternative. The only specific barriers to foreign entry mentioned by Stearns are transportation costs, manufacturing costs, and the “demonstrated ability of the dominant firm to charge supracompeti-tive prices.”

All imports will face transportation costs. However, domestic producers will also incur some analogous costs shipping products from their plants to end users. For transport costs to represent a true barrier to entry, there must be a showing that in a particular industry the costs incurred by new entrants significantly exceed the transport costs incurred by the monopolist. While the record indicates that costs of exporting in the industry are substantial, we have no evidence of the costs to FMC of shipping its bridges around the country. Moreover, transport costs have not prevented both' Stearns and FMC from successfully competing in Europe and Asia against native bridge manufacturers. Stearns’ expert admitted this, noting that in many of these cases the American companies shipped components over to be fabricated by local sub-contractors. It is unclear why a foreign corporation could not use the same strategy in the American market, and we have no evidence of how much this practice might affect costs. The evidence in the record is for the export of complete bridge units. The report relied on by Stearns’ expert also noted that foreign manufacturers are intimidated by customer satisfaction with Stearns and FMC — a situation which would likely change when Stearns is gone and FMC raises its prices. In any case, Stearns’ expert also admitted that a foreign firm could circumvent the transport issue entirely by setting up manufacturing *531 plants in the United States. 10 Given the large percentage of world sales America represents, it would not seem unreasonable to assume that they would do so once FMC began to raise prices.

Stearns’ briefs referred to “other costs,” which appears to refer to its expert’s brief mention of manufacturing costs and the airlines’ familiarity with FMC’s “brand.” Stearns’ expert conceded that manufacturing setup was not particularly onerous in the industry. “The principal barrier to entry into the North American PBB business is not the scale of manufacturing required.” New entrants to a market will always face these kinds of entry costs. They will also always face barriers stemming from consumer inertia and unfamiliarity with its products. “New entrants and customers in virtually any market emphasize the importance of a reputation for delivering a quality good or service.... [Plaintiff’s argument that reputation is entry barrier], without some limiting principle (that it fails to supply), implies that there are barriers to entry, significant in an antitrust sense, in all markets.” Advo, 51 F.3d at 1201-02. The question is not whether there are barriers to entry, but rather whether the barriers in a particular industry are large enough to trigger judicial concern. See Matsushita, 106 S.Ct. at 1348 n. 15 (noting lack of evidence that entry into the market was “especially” difficult); Cargill, 107 S.Ct. at 494 n. 15 (issue is whether barriers are “significant”). There was no evidence presented that industrial set-up costs or the costs associated with overcoming consumers’ settled preferences created unusual barriers to entry in the bridge market. 11

The barrier to entry that Stearns’ expert focused on was the same conduct that gave rise to exclusionary conduct claims. Just as the core of the claimed predation threat to Stearns’ survival was in actuality a restatement of these claims, Stearns’ allegation here is merely another attempt to repackage these same allegations as a barrier to entry. Since FMC can “induce” municipalities and FMC’s “friends” in the airline industry to ignorantly (though honestly) act against their own economic interests, it is claimed that new competitors will be scared off. We have discussed above that the conduct at issue did not violate the antitrust laws. It was merely vigorous competition, and the ultimate consumer of the product at all times retained the power of choice. 12 We decline to find this unobjectionable conduct constitutes a barrier to market entry. Moreover, while outside observers may indeed hesitate when viewing FMC’s current success in wooing municipalities, the whole theory of predation postulates that FMC’s behavior will significantly change once Stearns is eliminated. When FMC is charging supraeompetitive prices, its quality arguments will become less persuasive. A competitor could either match the quality standards that FMC has convinced the municipalities to adopt and underbid, or show them that the quality differential that justified adoption of certain specifications at a lower price cannot serve to mandate the same result at the supracompetitive price. 13

*532 B. Below-Cost Pricing

The above analysis, which we find determinative, assumes arguendo that all of Stearns’ allegations of below-cost pricing were supported by evidence. Strengthening our above conclusion and providing an independent ground for rejecting Stearns’ claim is our support for the district court’s conclusion that Stearns had not in fact put forth evidence of below-cost pricing. Under Brooke Group, a claimant must demonstrate that the prices at issue were below an appropriate measure of its rival’s costs. Brooke Group, 113 S.Ct. at 2587-88 (declining to resolve conflict among circuits over what constitutes a proper measure of cost, but finding only below-cost prices can lead to liability). Stearns incorrectly relies on cases in this Circuit predating Brooke Group, in which we left open the possibility that prices above a monopolist’s variable costs could be predatory under certain circumstances. See, e.g., Adjusters Replace-A-Car, Inc. v. Agency Rent-A-Car, Inc., 735 F.2d 884, 889-91 (5th Cir.1984) (allowing for predation when prices were above cost but barriers to entry in an industry were high). The district court case from which Stearns extracts its predatory pricing standard was decided in the same month as, but before, Brooke Group. See Continental Airlines, Inc. v. American Airlines, Inc., 824 F.Supp. 689 (S.D.Tex.1993) (discussing pricing above variable cost but below short-run profit maximizing price). In the wake of Brooke Group’s clarification of the standard, a plaintiff must show pricing below the standard this Court has long embraced as an appropriate measure of cost-average variable cost. See Adjusters, 735 F.2d at 891 (pricing below cost is pricing below average variable cost); International Air Industries, Inc. v. American Excelsior Co., 517 F.2d 714, 724 (5th Cir.1975) (embracing commentator’s proposal of average variable costs).

Ideally, an inquiry into whether a monopolist had sold his product below cost would look at the true marginal cost — we would attempt to discover the precise cost to the firm of producing the extra product that it is alleged to have sold below cost. But because the true marginal costs of production are difficult to generate, this Court attempts to estimate them by using average variable costs. See id. at 724. In this analysis, we attempt to distinguish between costs that are fixed — at least over the short term — and costs that vary with the amount produced. See Adjusters, 735 F.2d at 889. Thus salaried labor costs, rent or depreciation on real estate, and certain capital expenses are considered fixed. But inputs like hourly labor, the cost of materials, transport, and electrical consumption at a plant will vary, and are relevant to a predation inquiry.

This Court has found that judgment as a matter of law is appropriate when a plaintiff fails to adequately specify how the challenged pricing undercuts the defendant’s variable costs. “Plaintiffs did not offer any evidence respecting [Defendant’s] variable and fixed costs of operation. Rather, plaintiffs interpreted [Defendant’s] admission that it had suffered ‘a net loss from operations’ to be effectively an admission of predatory pricing. This was a costly error.” Adjusters, 735 F.2d at 891 (affirming J.M.L. for defendants). Here, Stearns has similarly erred. It has largely rested its allegation on evidence that FMC may have bid at a “negative margin” without exploring the relationship between variable costs, fixed costs, and profits.

Stearns has barely attempted to sort out what these costs may have been on the projects in question. Its expert’s opinion for the most part completely ignored the legal standard embraced by this Court and instead opted to engage in a comparison of what FMC bid for the Washington airport project *533 when it was proposing a sole source contract and what it charged when the project went to competitive bid against Stearns. 14 That the cost of the competitive bid was lower is evidence that the airport authority was wise to reject the sole-sourcing proposal. It is not evidence of a bid below variable cost.

For the bulk of the challenged bids, Steams’ only evidence was FMC’s risk memorandum on the challenged projects. The record indicates, and Stearns was eager to point out when they were used against it, that these documents are of little utility in estimating the true costs of a project. Nevertheless, they are the only evidence Stearns could produce that even suggested the costs FMC incurred on the projects. On one of these documents, Stearns claims it has found its smoking gun — a note on the bottom that part C of the project would run at a negative operating margin of 3.7%. 15 But this allegation is undermined by the fact that the table from which the margin is drawn includes a section for general and administrative expenses (“G&A”). FMC contends that G&A is one of several categories in which profits are allocated. If this is correct, then the negative margin referenced in the document indicates only that a portion of the project failed to meet a benchmark profit target, not a bid below cost — variable or otherwise. Resolution of the G&A question is also critical for the other challenged projects, for which the sole basis for Steams’ charges of below-cost pricing is the fact that the G&A percentage was reduced from its customary eighteen percent to ten percent.

FMC has produced evidence indicating that G&A is not a variable cost. The risk memorandum contained allocation categories for G&A, markup, and margin. FMC produced affidavits that stated all three of these categories were expected profit on a deal, and in other sections of the Washington risk memorandum G&A and margin are combined and collectively referred to as margin. An examination of the risk memoranda supports the conclusion that these categories were at the least not a variable cost. 16 Material, manufacturing, and engineering were separate cost sections in the memorandum, and these categories describe the bulk of the variable costs one would expect on a project like this. Both the material and manufacturing section contained percentage increases for overhead. Thus after calculating the labor costs for a project, the memorandum added a separate charge of 265% of labor costs for manufacturing overhead. These overhead projections would seem to be a logical place to find the full variable costs of the project — including things like sales expenses.

*534 Stearns nevertheless claims that G&A represents a variable cost, not a profit margin or an internal allocation of fixed costs. The difficulty with this assertion is that nothing supports it. We have no attempt by Stearns to refute the credible evidence FMC has put forth that removes G&A from the realm of variable costs. As noted earlier, although Stearns’ expert mentioned the concept of variable cost in passing, he erroneously concluded that it was unnecessary to address it. He thus spent most of his argument on a tangent that is irrelevant to our central inquiry. At no point did the expert explain what G&A represented or state that it was a variable cost. In its briefs, Stearns could only restate its contention that G&A did not constitute profits. And, Stearns has not offered a coherent explanation of what G&A represents if it does not represent profits, let alone evidence that (or to what extent) it is a variable cost to FMC.

Slightly more concrete evidence in favor of Stearns comes from the testimony of Richard Pell. Mr. Pell, a former employee of FMC, testified in his deposition that FMC’s marketing department and its accountants maintained separate books. As an executive in engineering, he reported that several times his department’s estimate of its costs on projects were lowered by the marketing department prior to a bid. The net result was that his department experienced frequent cost overruns on projects trying to meet the artificially low projection enshrined in the bids. It appears from the record that these costs might be viewed as variable costs.

However, we are not concerned here with reviewing the interoffice politics or internal cost allocations of FMC. Stearns failed to develop Mr. Pell’s testimony or explain how manipulation of the engineering variable cost could have rendered an entire project — or even a discrete portion of one — below variable cost. While one of Mr. Pell’s objections was that this kind of behavior could lead to losing money on a project, he could not and did not opine that any of the projects he worked on were on the whole below variable cost. Stearns did not provide us with any evidence that the understating of cost on the engineering projects in question was severe enough to cancel out the ten to eighteen percent G&A profits they generated. The risk memorandum that we have indicates that engineering costs were a relatively minor cost compared to materials and manufacturing.

The sections of Stearns’ expert testimony that even hinted at FMC’s costs suffer from a similar defect. Dr. Eads testified that on the Washington project FMC eliminated its inflation adjustment, thus incurring the risk that the company’s costs would be higher than anticipated for the sections of the project occurring in the years following the initial bid. Assuming arguendo that the risk that inflation will exceed the return FMC receives on its capital over the life of the project is a variable cost, Dr. Eads was silent as to the amount of the “cost” of this risk. He also claimed that FMC’s bid on the maintenance section of the contract was suspect because it was much lower than the analogous bid by Stearns. Of course, the mere fact that a producer can or does charge less than a competitor does not indicate below-cost pricing. The opinion was silent as to what variable costs FMC could expect to incur providing maintenance and how the bid was below them.

Because Stearns has failed to raise a genuine issue of material fact regarding both its exclusionary conduct and predatory pricing claims under the Sherman Act, summary judgment on these claims must be affirmed. This result also mandates affirmation of the summary judgment on Stearns’ Robinson-Patman and state law claims, which are urged on appeal only derivatively of the Sherman Act claims.

IV. Denial of Discovery motion.

Stearns contends that the district court erred in denying its Rule 56(f) motion to suspend summary judgment pending the completion of discovery. We review the denial of a Rule 56(f) motion for abuse of discretion. See, e.g., Fontenot v. Upjohn Company, 780 F.2d 1190, 1193 (5th Cir.1986). Such motions are generally favored, and should be liberally granted. See International Shortstop, Inc. v. Rally’s Inc., 939 F.2d 1257, 1267 (5th Cir.1991). However, to *535 justify a continuance, the Rule 56(f) motion must demonstrate 1) why the movant needs additional discovery and 2) how the additional discovery will likely create a genuine issue of material fact. Krim v. BancTexas Group, Inc., 989 F.2d 1435, 1442 (5th Cir.1993). On appeal, we will not consider justifications for granting a continuance that were not presented with the original motion. See Solo Serve Corp. v. Westowne Associates, 929 F.2d 160, 167 (5th Cir.1991).

We begin our analysis by pointing out that this case had been pending for over fifteen months prior to the district court’s entry of final judgment in favor of FMC. The section 1 claim, on which the district court had entered summary judgment earlier, required almost identical factual support as the claims at issue here. The record indicates that Stearns had reviewed over half a million FMC documents and had also subpoenaed documents from twenty municipal airport authorities. It had also conducted several depositions of key FMC employees. While the district court’s scheduling order had indicated that the final deadline for discovery was April 30, 1997, the order clearly contemplated summary judgment prior to that date, since its cutoff for the filing of such motions was almost two months prior to the discovery cut-off. Stearns also delayed filing its Rule 56(f) motion until the time its response to FMC’s summary judgment motion was due, a deadline that the district court had already extended at Stearns’ request.

The district court denied Stearns’ motion because it lacked specificity in identifying the needed discovery. On review of the record, we cannot say that this ruling constituted an abuse of discretion. Stearns’ motion specifically requested a stay pending the deposition of several FMC executives. The motion explains that all of the proposed deponents were in a position of authority and were connected to several documents relied on by Stearns. Stearns argued that deposing these parties was necessary because “Stearns expects that the depositions will provide evidence on a number of topics, including FMC’s predatory and exclusionary conduct, FMC’s strategies to avoid competitive bidding and price competition, the relevant product and geographic markets in this case and the barriers to enter this market.” If these depositions proved fruitful, Stearns put the court on notice that it might pursue additional depositions of bridge customers which would develop “testimony on topics including the market, FMC’s predatory conduct, and the injuries to Stearns and competition caused by FMC’s predatory conduct.”

While Stearns’ motion indicated how the desired discovery was in a quite general sense relevant to the case, the district court did not abuse its discretion in finding that the motion lacked needed specificity. The movant must be able to demonstrate how postponement and additional discovery will allow him to defeat summary judgment; it is not enough to “rely on vague assertions that discovery will produce needed, but unspecified, facts.” Washington v. Allstate Insurance Co., 901 F.2d 1281, 1285 (5th Cir.1990). See also Krim, 989 F.2d at 1441 (must demonstrate how discovery will lead to genuine issue of fact). Here, Stearns’ exclusionary conduct claims required either 1) a showing that the conduct at issue could not be justified by FMC without reference to its effect on its competitors or 2) a showing that the employees and agents of consumers of bridges and those with influence on them had somehow had the independence and integrity of their judgment clouded by external forces so that we may not assume that their decisions in FMC’s favor were intended by them to be in the best interests of their employers. The motion fails to identify how further discovery could prove either of these points. It does not claim, for example, that further depositions will reveal that the airlines and airport staff were bribed or otherwise driven by anything other than their perceptions of the merits of the product when they recommended FMC bridges or specifications to the municipal authorities.

The motion is also unhelpful in detailing how the predatory pricing claims could be saved from summary judgment. The crucial issue of FMC’s variable costs is not even hinted at in the motion. With regards to recoupment, while the motion does mention barriers to entry, the predatory pricing claim required not only a showing of such barriers, *536 but also a demonstration that the extent and duration of the alleged below-variable cost pricing was sufficient to drive Stearns from the market. There is nothing specific in the motion suggesting that evidence of more below-variable cost pricing would have been revealed by additional discovery.

Stearns claims that the district court’s ruling was an abuse of discretion because concurrent with the denial of its motion, the court allowed FMC to exceed the ten deposition limit imposed by the Federal Rules. These are separate issues. FMC indicated that the requested depositions were necessary to preserve the testimony of parties who would not be available to testify at trial. We cannot say that the granting of an unrelated request transforms the district court’s proper exercise of judgment into an abuse of discretion.

V. Taxation of Costs

The district court awarded FMC costs under 28 U.S.C. § 1920 and Rule 54(d)(1). On appeal, Stearns does not challenge the award itself, but rather attacks the inclusion of certain deposition and photocopying costs. Costs related to the taking of depositions and the copying of documents are allowed if the materials were necessarily obtained for use in the case. This Court reviews a lower court’s allowance of costs for clear abuse of discretion, granting the lower court “great latitude in this determination.” Fogleman v. ARAMCO, 920 F.2d 278, 285-86 (5th Cir.1991).

The record indicates that the lower court exercised oversight over FMC’s claimed costs, striking from its bill of costs items such as mini-transcripts and computer copies. Stearns’ challenge to the deposition costs is grounded in the fact that certain depositions were not used in FMC’s summary judgment filings. It thus claims that they were merely for general discovery and not necessary to the case. But we have indicated that it is not required that a deposition actually be introduced in evidence for it to be necessary for a case — as long as there is a reasonable expectation that the deposition may be used for trial preparation, it may be included in costs. Id. at 285. We are satisfied that the district court did not abuse its discretion in finding the depositions in question could be expected to be used at trial.

Stearns’ challenge to FMC’s photocopying charges must also fail. While we have indicated that multiple copies of relevant documents may not be charged to an opponent, we have never held that a district court may not award a litigant the cost of preparing a single set of the documents in a case. See id. at 286. The district court did not abuse its discretion in approving these costs.

Conclusion

We find that summary judgment in favor of FMC on the Sherman Act claims was warranted. Accordingly, we also affirm the summary judgment on the derivative Robinson-Patman and state law claims. We do not find that the district court abused its discretion in denying Stearns’ Rule 56(f) continuance motion and in its determination of costs.

For the reasons stated above, the judgment of the district court is

AFFIRMED.

Notes

. While this category is the most ominous sounding, it was functionally identical to the specification effort. In both instances, FMC attempted to get certain features it possessed and Stearns did not — such as electromechanical design, certification from an outside body, or a direct legal responsibility for the product — incorporated in the specifications.

. Stearns attempts to distinguish SuperTurf and Triple M because these contracts had an "or equal” clause. This is a distinction without a difference. In these cases, the specifications were for a specific brand. The result of such a specification and the addition of an “or equal” clause is essentially the same as that generated by the general specification here. A contract might say “FMC Bridge or equal” or it might say "electromechanical bridge.” In either case, FMC would meet the specification but Steams could also quality by demonstrating it could produce an electromechanical bridge equal to FMC’s.

. Stearns relies heavily on a case in this Circuit without recognizing that it was properly vacated pursuant to a settlement agreement and thus carries no precedential weight. Great Western Directories v. Southwestern Bell Telephone, 63 F.3d 1378, 1386 (5th Cir.1995), modified, 74 F.3d 613, vacated pursuant to settlement agreement (August 21, 1996), cert dismissed, 518 U.S. 1048, 117 S.Ct. 26, 135 L.Ed.2d 1120 (1996). However, Great Western also involved conduct that harmed the monopolist and could only be understood when one recognized that competitors suffered more severe harm. In Great Western the defendant asked an affiliated supplier to raise prices across the board, raising defendant’s costs but inflicting more pain on its cash-starved competitor. Id. at 1386 (quoting expert economic testimony that such behavior was irrational except as an attempt to cripple competitors).

. Stearns continually refers back to FMC documents which indicate that FMC was aware that a certain specification would prevent Stearns from bidding, and was pleased by this fact. Writing a memo saying that you are winning a competition on the merits does not change the fact that it is a competition on the merits. If Stearns reached the bidding stage, calculated its bid, and believed that it could maintain an adequate margin at a price FMC could not match, it would not violate section 2 by expressing its satisfaction with the expected result.

. Apparently, a handful of negotiations between FMC and municipalities may have led to technical violations of the relevant public contracting statutes. Steams was either silent or half-heart-ed in complaining to the relevant authorities when these violations occurred. It now attempts to claim that violation of these municipal bidding statutes constitutes a per se antitrust violation. But a major purpose of these state statutes is protection of the municipal taxpayer from corruption — which we have no evidence of (there is nothing to indicate that in any of these instances the municipal authorities were acting in other than what they thought was best for the municipality in respect to the particular purchase being made). The Sherman Act, in contrast, protects the consumer from anticompetitive forces. We decline Steam’s invitation. “Even a direct contract for the Guilbert system, without any pretense of putting the job out for bid (and thus a clear violation of the competitive bid statute), would not in itself have constituted a restraint of trade under the Sherman Act if the selection of Guilbert had been made in an atmosphere free from anticompetitive restraints.” Security Fire Door, 484 F.2dat 1031.

. Stearns fails to articulate how under its theory of the case a municipality could ever make a decision to favor quality over price when the higher quality producer has a strong market share. The logic of Stearns’ argument would not only make it impossible for an informed municipal consumer to pick FMC's smart bridge over Steams' cheaper bridge, it would also bar a consumer from purchasing aircraft boarding bridges in the first place. Steams’ expert noted that other methods of boarding passengers on airplanes exist. — notably stairs — but bridges are "so superior to these other methods that it has largely displaced them.’’ This superiority is reflected in the specifications that airports now draw up — all of the contracts at issue specified bridges. Thus even if a stair assembly was cheaper, its manufacturer has been excluded from the final bidding process. But no one would argue that a municipality is forbidden from making this choice, even if Steams and FMC collectively have a lock on the passenger boarding device market. Nor could it credibly be maintained — without evidence that the bridge manufacturers had corrupted the judgment of the consumer’s agents — that this exclusion was not on the merits.

.But Steams complains just as vociferously about FMC’s attempts to include quality as a weighting factor in determining bids — when it clearly could compete, albeit with some recognition of the difference in technologies — as it does about bids in which it was barred by a particular specification. And it should be noted that nothing is stopping Steams from developing smart-bridges and electromechanical technologies that would match the specifications it complains of. While the record indicates that the time and cost involved in retooling make this impractical in regards to a particular bid after its specifications have been announced, in terms of future bids we have no evidence to sustain a finding that *526 Stearns could not begin this, process tomorrow and thus expand its ability to compete on such projects.

. And, at oral argument it was asserted that the "inducement” had inevitably led to the officers ending up “in cahoots.”

. Stearns attempts to avoid the application of this standard by claiming that here, unlike in Matsushita and Brooke Group, a single defendant with overwhelming market share is involved. While the Court noted that the multi-party nature of the claimed predatory pricing conspiracies (between Japanese electronics manufacturers and large tobacco companies, respectively) increased the irrationality of the claimed conduct, on both occasions the Court indicated that the reasoning of the opinions applies to claims against a single firm. See Brooke Group, 113 S.Ct. at 2590 (schemes are "even more improbable” when multiple firms involved); Matsushita, 106 S.Ct. at 1357 (“These observations apply even to predatory pricing by a single firm seeking monopoly power.”). The basic insight of these cases — that predation as a strategy is so unlikely to reap rewards that it should not be inferred easily, and that such an inference should be especially shunned since it is so hard to disentangle from the type of vigorous price competition that the antitrust laws seek to promote — is not altered when only one defendant is involved. See American Academic Suppliers v. Beckley-Cardy, Inc., 922 F.2d 1317, 1319-21 (7th Cir.1991) (affirming summary judgment for single defendant on section 1 predatory pricing claims by noting lack of evidence that defendant could have hoped to recoup losses incurred during alleged scheme). The size of defendant's market share may of course be relevant in determining the ease with which he may drive out a competitor through his scheme — but it does not, standing alone, allow a presumption that this can occur. Nor does market share tell us anything about the problem of new entrants preventing the sustained charging of supra-competitive prices necessary for recoupment.

. ‘‘[I]f transportation costs were the only problem, firms from overseas could set up manufacturing facilities here in North America.”

. Richard Pell, a former employee of FMC whose testimony is critical for all of Stearns’ predatory pricing claims, was quite emphatic that FMC could not rely on barriers to entry to protect it from competition. "There will always be a competitor to Jetway. There may be short periods when there aren’t; but if they manage to put Stearns out of business, for example, within two years, there will be another bridge manufacturer competing with Jetway.”

. Summary judgment is appropriate when an ill-reasoned expert opinion suggests the court adopt an irrational inference, or rests on an error of fact or law. See Matsushita, 106 S.Ct, at 1360 n. 19 (expert opinion on predation has little probative value in light of economic factors that indicate expert’s scenario is irrational); Bell, 847 F.2d at 1184 (affirming decision of district court on the question of market power, since expert imprecisely defined the market, which led to a legal error in his conclusion). Here, Stearns’ expert rested his conclusion on an error of law— he assumed FMC’s efforts to sell its products violated section 2 of the Sherman Act.

.Stearns places great reliance on its claim that when Stearns does not bid on a project, FMC's bids are significantly higher. It directs the court in particular to two instances involving very small projects in which FMC came to the bid armed with two proposals. When Steams failed *532 to bid, it used its higher bid and pocketed the other proposal. Insofar as we are asked to infer from this that the lower, unused bid was predatory, we note that there is no evidence showing the unused bid was below cost. If these isolated episodes are relied on to show FMC's intent and ability to charge supracompetitive prices, we note that the antitrust laws cannot protect consumers from the inadequacies of a competitor. A new entrant to the market may not be as cavalier about letting business opportunities pass as Steams was in these instances. Of course, documentation of these incidents would seem to provide Stearns an excellent — albeit apparently unused — tool to persuade municipalities that ov-erreliance on FMC is not in their best interest.

. It must be noted that the expert relied on an erroneous interpretation of the law regarding predatory pricing. The opinion clearly indicated that the expert believed the law of this Circuit allowed a finding of predation when prices are above a firm’s variable costs but below a "short-run profit maximizing price." As we explained above, this position is no longer tenable in the wake of Brooke Group. This error may explain, but does not excuse, the expert’s failure to address the question of variable cost. In affirming summary judgment, we may disregard the conclusions of an expert opinion grounded in an error of law. See Bell, 847 F.2d at 1184.

. A threshold problem with this allegation is that even if part C was bid below-cost. Steams has not alleged that the project as a whole was unprofitable. In an analogous case, we rejected an argument that price cuts in the original equipment market could be examined in isolation when the evidence indicated that the replacement market and original equipment market were inseparable. Stitt Spark Plug Co. v. Champion Spark Plug Co., 840 F.2d 1253, 1256 (5th Cir.1988) (”[A]ny meaningful comparison of price and cost must encompass Champion’s sales to both markets. Stitt's evidence did not demonstrate that Champion’s practices were ‘predatory’ across both markets.”). Here, the fact that FMC may have chosen for internal reasons or salesmanship purposes to shift costs in this manner is not objectionable without a showing that the project as a whole was not priced above its variable cost. When a company has a "buy one, get one free” promotion, it would be incorrect to look at the nominal price of the "free” product— zero — and infer predation from this fact.

.Even if we were to simply disregard FMC’s affidavit, there is nothing inherent in the term "general and administrative” that automatically allows an inference that this category was a variable cost. Administrative costs may involve the work of salaried workers not subject to overtime — which over the short term is a fixed cost. And as a catch-all category G&A would seem a natural place to allocate a percentage of long term fixed costs — like rent and the salary of the CEO — for internal accounting purposes.