delivered the opinion of the Court.
Petitioner (Texaco) sold gasoline directly to respondents and several other retailers in Spokane, Washington, at its re*547tail tank wagon (RTW) prices while it granted substantial discounts to two distributors. During the period between 1972 and 1981, the stations supplied by the two distributors increased their sales volume dramatically, while respondents’ sales suffered a corresponding decline. Respondents filed an action against Texaco under the Robinson-Patman Act amendment to the Clayton Act (Act), 38 Stat. 730, as amended, 49 Stat. 1526, 15 U. S. C. § 13, alleging that the distributor discounts violated § 2(a) of the Act, 15 U. S. C. § 13(a). Respondents recovered treble damages, and the Court of Appeals for the Ninth Circuit affirmed the judgment. 842 F. 2d 1034 (1988). We granted certiorari, 490 U. S. 1105 (1989), to consider Texaco’s contention that legitimate functional discounts do not violate the Act because a seller is not responsible for its customers’ independent resale pricing decisions. While we agree with the basic thrust of Texaco’s argument, we conclude that in this case it is foreclosed by the facts of record.
I
Given the jury’s general verdict in favor of respondents, disputed questions of fact have been resolved in their favor. There seems, moreover, to be no serious doubt about the character of the market, Texaco’s pricing practices, or the relative importance of Texaco’s direct sales to retailers *548(“throughput” business) and its sales to distributors. The principal disputes at trial related to questions of causation and damages.
Respondents are 12 independent Texaco retailers. They displayed the Texaco trademark, accepted Texaco credit cards, and bought their gasoline products directly from Texaco. Texaco delivered the gasoline to respondents’ stations.
The retail gasoline market in Spokane was highly competitive throughout the damages period, which ran from 1972 to 1981. Stations marketing the nationally advertised Texaco gasoline competed with other major brands as well as with stations featuring independent brands. Moreover, although discounted prices at a nearby Texaco station would have the most obvious impact on a respondent’s trade, the cross-city traffic patterns and relatively small size of Spokane produced a citywide competitive market. See, e. g., App. 244, 283-291. Texaco’s throughput sales in the Spokane market declined from a monthly volume of 569,269 gallons in 1970 to 389,557 gallons in 1975. Id., at 487-488. Texaco’s independent retailers’ share of the market for Texaco gas declined from 76% to 49%.1 Ibid. Seven of the respondents’ stations were out of business by the end of 1978. Id., at 22-23, Record 501.
Respondents tried unsuccessfully to increase their ability to compete with lower priced stations. Some tried converting from full service to self-service stations. See, e. g., App. 55-56. Two of the respondents sought to buy their own tank trucks and haul their gasoline from Texaco’s supply point, but Texaco vetoed that proposal. Id., at 38-41, 59.
*549While the independent retailers struggled, two Spokane gasoline distributors supplied by Texaco prospered. Gull Oil Company (Gull) had its headquarters in Seattle and distributed petroleum products in four Western States under its own name. Id., at 94-95. In Spokane it purchased its gas from Texaco at prices that ranged from 60 to 40 below Texaco’s RTW price. Id., at 31-32. Gull resold that product under its own name; the fact that it was being supplied by Texaco was not known by either the public or the respondents. See, e. g., id., at 256. In Spokane, Gull supplied about 15 stations; some were “consignment stations” and some were “commission stations.” In both situations Gull retained title to the gasoline until it was pumped into a motorist’s tank. In the consignment stations, the station operator set the retail prices, but in the commission stations Gull set the prices and paid the operator a commission. Its policy was to price its gasoline at a penny less than the prevailing price for major brands. Gull employed two truckdrivers in Spokane who picked up product at Texaco’s bulk plant and delivered it to the Gull stations. It also employed one supervisor in Spokane. Apart from its trucks and investment in retail facilities, Gull apparently owned no assets in that market. Id., at 96-109, 504-512. At least with respect to the commission stations, Gull is fairly characterized as a retailer of gasoline throughout the relevant period.
The Dompier Oil Company (Dompier) started business in 1954 selling Quaker State Motor Oil. In 1960 it became a full line distributor of Texaco products, and by the mid-1970’s its sales of gasoline represented over three-quarters of its business. Id., at 114-115. Dompier purchased Texaco gasoline at prices of 3.95¢ to 3.65¢ below the RTW price. Dompier thus paid a higher price than Gull, but Dompier, unlike Gull, resold its gas under the Texaco brand names. Id., at 24, 29-30. It supplied about 8 to 10 Spokane retail stations. In the period prior to October 1974, two of those stations were owned by the president of Dompier but the others were inde*550pendently operated. See, e. g., id., at 119-121, 147-148. In the early 1970’s, Texaco representatives encouraged Dompier to enter the retail business directly, and in 1974 and 1975 it acquired four stations.2 Id., at 114-135, 483-503. Dompier’s president estimated at trial that the share of its total gasoline sales made at retail during the middle 1970's was “[p]robably 84 to 90 percent.” Id., at 115.
Like Gull, Dompier picked up Texaco’s product at the Texaco bulk plant and delivered directly to retail outlets. Unlike Gull, Dompier owned a bulk storage facility, but it was seldom used because its capacity was less than that of many retail stations. Again unlike Gull, Dompier received from Texaco the equivalent of the common carrier rate for delivering the gasoline product to the retail outlets. Thus, in addition to its discount from the RTW price, Dompier made a profit on its hauling function.3 Id., at 123-131, 186-192, 411-413.
The stations supplied by Dompier regularly sold at retail at lower prices than respondents’. Even before Dompier directly entered the retail business in 1974, its customers were *551selling to consumers at prices barely above the RTW price. Id., at 329-338; Record 315, 1250-1251. Dompier’s sales volume increased continuously and substantially throughout the relevant period. Between 1970 and 1975 its monthly sales volume increased from 155,152 gallons to 462,956 gallons; this represented an increase from 20.7% to almost 50% of Texaco’s sales in Spokane. App. 487-488.
There was ample evidence that Texaco executives were well aware of Dompier’s dramatic growth and believed that it was attributable to “the magnitude of the distributor discount and the hauling allowance.”4 See also, e. g., id., at 213-223, 407-413. In response to complaints from individual respondents about Dompier’s aggressive pricing, however, Texaco representatives professed that they “couldn’t understand it.” Record 401-404.
H HH
Respondents filed suit against Texaco in July 1976. After a 4-week trial, the jury awarded damages measured by the difference between the RTW price and the price paid by Dompier. As we subsequently decided in J. Truett Payne Co. v. Chrysler Motors Corp., 451 U. S. 557 (1981), this measure of damages was improper. Accordingly, although it rejected Texaco’s defenses on the issue of liability,5 the Court of Appeals for the Ninth Circuit remanded the case for *552a new trial. Hasbrouck v. Texaco, Inc., 663 F. 2d 930 (1981), cert. denied, 459 U. S. 828 (1982).
At the second trial, Texaco contended that the special prices to Gull and Dompier were justified by cost savings,6 were the product of a good-faith attempt to meet competition,7 and were lawful “functional discounts.” The District Court withheld the cost justification defense from the jury because it was not supported by the evidence and the jury rejected the other defenses. It awarded respondents actual damages of $449,900.8 The jury apparently credited the testimony of respondents’ expert witness who had estimated what the respondents’ profits would have been if they had paid the same prices as the four stations owned by Dompier. See 634 F. Supp. 34, 43 (ED Wash. 1985); 842 F. 2d, at 1043-1044.
In Texaco’s motion for judgment notwithstanding the verdict, it claimed as a matter of law that its functional discounts did not adversely affect competition within the meaning of the Act because any injury to respondents was attributable to decisions made independently by Dompier. The District Court denied the motion. In an opinion supplementing its oral ruling denying Texaco’s motion for a directed verdict, the Court assumed, arguendo, that Dompier was entitled to a *553functional discount, even on the gas that was sold at retail,9 but nevertheless concluded that the “presumed legality of functional discounts” had been rebutted by evidence that the amount of the discounts to Gull and Dompier was not reasonably related to the cost of any function that they performed.10 634 F. Supp., at 37-38, and n. 4.
The Court of Appeals affirmed. It reasoned:
*554“As the Supreme Court long ago made clear, and recently reaffirmed, there may be a Robinson-Patman violation even if the favored and disfavored buyers do not compete, so long as the customers of the favored buyer compete with the disfavored buyer or its customers. Morton Salt, 334 U. S. at 43-44 . . . ; Perkins v. Standard Oil Co., 395 U. S. 642, 646-47 . . . (1969); Falls City Indus., Inc. v. Vanco Beverages, Inc., 460 U. S. 428, 434-35 . . . (1983). Despite the fact that Dompier and Gull, at least in their capacities as wholesalers, did not compete directly with Hasbrouck, a section 2(a) violation may occur if (1) the discount they received was not cost-based and (2) all or a portion of it was passed on by them to customers of theirs who competed with Hasbrouck. Morton Salt, 334 U. S. at 43-44 . . . ; Perkins v. Standard Oil, 395 U. S. at 648-49 . . . ; see 3 E. Kintner & J. Bauer, supra, § 22.14.
“Hasbrouck presented ample evidence to demonstrate that . . . the services performed by Gull and Dompier were insubstantial and did not justify the functional discount.” 842 F. 2d, at 1039.
The Court of Appeals concluded its analysis by observing:
“To hold that price discrimination between a wholesaler and a retailer could never violate the Robinson-Patman Act would leave immune from antitrust scrutiny a discriminatory pricing procedure that can effectively serve to harm competition. We think such a result would be contrary to the objectives of the Robinson-Patman Act.” Id., at 1040 (emphasis in original).
III
It is appropriate to begin our consideration of the legal status of functional discounts11 by examining the language of the Act. Section 2(a) provides in part:
*555“It shall be unlawful for any person engaged in commerce, in the course of such commerce, either directly or indirectly, to discriminate in price between different purchasers of commodities of like grade and quality, where either or any of the purchases involved in such discrimination are in commerce, where such commodities are sold for use, consumption, or resale within the United States or any Territory thereof or the District of Columbia or any insular possession or other place under the jurisdiction of the United States, and where the effect of such discrimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, or to injure, destroy, or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination, or with customers of either of them . . . 15 U. S. C. § 13(a).
The Act contains no express reference to functional discounts.12 It does contain two affirmative defenses that provide protection for two categories of discounts—those that *556are justified by savings in the seller’s cost of manufacture, delivery, or sale,13 and those that represent a good-faith response to the equally low prices of a competitor. Standard Oil Co. v. FTC, 340 U. S. 231, 250 (1951). As the case comes to us, neither of those defenses is available to Texaco.
In order to establish a violation of the Act, respondents had the burden of proving four facts: (1) that Texaco’s sales to Gull and Dompier were made in interstate commerce; (2) that the gasoline sold to them was of the same grade and quality as that sold to respondents; (3) that Texaco discriminated in price as between Gull and Dompier on the one hand and respondents on the other; and (4) that the discrimination had a prohibited effect on competition. 15 U. S. C. § 13(a). Moreover, for each respondent to recover damages, he had the burden of proving the extent of his actual injuries. J. Truett Payne, 451 U. S., at 562.
The first two elements of respondents’ case are not disputed in this Court,14 and we do not understand Texaco to be challenging the sufficiency of respondents’ proof of damages. Texaco does argue, however, that although it charged different prices, it did not “discriminate in price” within the meaning of the Act, and that, at least to the extent that Gull and Dompier acted as wholesalers, the price differentials did not injure competition. We consider the two arguments separately.
IV
Texaco’s first argument would create a blanket exemption for all functional discounts. Indeed, carried to its logical conclusion, it would exempt all price differentials except those given to competing purchasers. The primary basis for *557Texaco’s argument is the following comment by Congressman Utterback, an active sponsor of the Act:
“In its meaning as simple English, a discrimination is more than a mere difference. Underlying the meaning of the word is the idea that some relationship exists between the parties to the discrimination which entitles them to equal treatment, whereby the difference granted to one casts some burden or disadvantage upon the other. If the two are competing in the resale of the goods concerned, that relationship exists. Where, also, the price to one is so low as to involve a sacrifice of some part of the seller’s necessary costs and profit as applied to that business, it leaves that deficit inevitably to be made up in higher prices to his other customers; and there, too, a relationship may exist upon which to base the charge of discrimination. But where no such relationship exists, where the goods are sold in different markets and the conditions affecting those markets set different price levels for them, the sale to different customers at those different prices would not constitute a discrimination within the meaning of this bill.” 80 Cong. Rec. 9416 (1936).
We have previously considered this excerpt from the legislative history and have refused to draw from it the conclusion which Texaco proposes. FTC v. Anheuser-Busch, Inc., 363 U. S. 536, 547-551 (1960). Although the excerpt does support Texaco’s argument, we remain persuaded that the argument is foreclosed by the text of the Act itself. In the context of a statute that plainly reveals a concern with competitive consequences at different levels of distribution, and carefully defines specific affirmative defenses, it would be anomalous to assume that the Congress intended the term “discriminate” to have such a limited meaning. In Anheuser-Busch we rejected an argument identical to Texaco’s in the context of a claim that a seller’s price differential had injured *558its own competitors—a so-called “primary line” claim.15 The reasons we gave for our decision in Anheuser-Busch apply here as well. After quoting Congressman Utterback’s statement in full, we wrote:
“The trouble with respondent’s arguments is not that they are necessarily irrelevant in a § 2(a) proceeding, but that they are misdirected when the issue under consideration is solely whether there has been a price discrimination. We are convinced that, whatever may be said with respect to the rest of §§2(a) and 2(b)—and we say nothing here—there are no overtones of business buccaneering in the §2(a) phrase ‘discriminate in price.’ Rather, a price discrimination within the meaning of that provision is merely a price difference.” Id., at 549.
After noting that this view was consistent with our precedents, we added:
“[T]he statute itself spells out the conditions which make a price difference illegal or legal, and we would derange this integrated statutory scheme were we to read other conditions into the law by means of the nondirective phrase, ‘discriminate in price.’ Not only would such action be contrary to what we conceive to be the meaning of the statute, but, perhaps because of this, it would be thoroughly undesirable. As one commentator has succinctly put it, ‘Inevitably every legal controversy over any price difference would shift from the detailed governing provisions —“injury,” cost justification, “meeting competition,” etc.—over into the “discrimination” concept for ad hoc resolution divorced from specifically pertinent statutory text.’ Rowe, Price Differen*559tials and Product Differentiation: The Issues Under the Robinson-Patman Act, 66 Yale L. J. 1, 38.” Id., at 650-561.
Since we have already decided that a price discrimination within the meaning of § 2(a) “is merely a price difference,” we must reject Texaco’s first argument.
V
In FTC v. Morton Salt Co., 334 U. S. 37, 46-47 (1948), we held that an injury to competition may be inferred from evidence that some purchasers had to pay their supplier “substantially more for their goods than their competitors had to pay.” See also Falls City Industries, Inc. v. Vanco Beverage, Inc., 460 U. S. 428, 435-436 (1983). Texaco, supported by the United States and the Federal Trade Commission as amici curiae (the Government), argues that this presumption should not apply to differences between prices charged to wholesalers and those charged to retailers. Moreover, they argue that it would be inconsistent with fundamental antitrust policies to construe the Act as requiring a seller to control his customers’ resale prices. The seller should not be held liable for the independent pricing decisions of his customers. As the Government correctly notes, Brief for United States et. al. as Amici Curiae 21-22 (filed Aug. 3, 1989), this argument endorses the position advocated 35 years ago in the Report of the Attorney General’s National Committee to Study the Antitrust Laws (1955).
After observing that suppliers ought not to be held liable for the independent pricing decisions of their buyers,16 and *560that without functional discounts distributors might go uncompensated for services they performed,17 the Committee wrote:
“The Committee recommends, therefore, that suppliers granting functional discounts either to single-function or to integrated buyers should not be held responsible for any consequences of their customers’ pricing tactics. Price cutting at the resale level is not in fact, and should not be held in law, ‘the effect of’ a differential that merely accords due recognition and reimbursement for actual marketing functions. The price cutting of a customer who receives this type of differential results from his own independent decision to lower price and operate at a lower profit margin per unit. The legality or illegality of this price cutting must be judged by the usual legal tests. In any event, consequent injury or lack of injury should not be the supplier’s legal concern.
“On the other hand, the law should tolerate no subterfuge. For instance, where a wholesaler-retailer buys only part of his goods as a wholesaler, he must not claim a functional discount on all. Only to the extent that a buyer actually performs certain functions, assuming all the risk, investment, and costs involved, should he le*561gally qualify for a functional discount. Hence a distributor should be eligible for a discount corresponding to any part of the function he actually performs on that part of the goods for which he performs it.” Id., at 208.
We generally agree with this description of the legal status of functional discounts. A supplier need not satisfy the rigorous requirements of the cost justification defense in order to prove that a particular functional discount is reasonable and accordingly did not cause any substantial lessening of competition between a wholesaler’s customers and the supplier’s direct customers.18 The record in this case, however, adequately supports the finding that Texaco violated the Act.
*562The hypothetical predicate for the Committee’s entire discussion of functional discounts is a price differential “that merely accords due recognition and reimbursement for actual marketing functions.” Such a discount is not illegal. In this case, however, both the District Court and the Court of Appeals concluded that even without viewing the evidence in the light most favorable to respondents, there was no substantial evidence indicating that the discounts to Gull and Dompier constituted a reasonable reimbursement for the value to Texaco of their actual marketing functions. 842 F. 2d, at 1039; 634 F. Supp., at 37, 38. Indeed, Dompier was separately compensated for its hauling function, and neither Gull nor Dompier maintained any significant storage facilities.
Despite this extraordinary absence of evidence to connect the discount to any savings enjoyed by Texaco, Texaco contends that the decision of the Court of Appeals cannot be affirmed without departing “from established precedent, from practicality, and from Congressional intent.” Brief for Petitioner 14.19 This argument assumes that holding suppliers liable for a gratuitous functional discount is somehow a novel practice. That assumption is flawed.
As we have already observed, the “due recognition and reimbursement” concept endorsed in the Attorney General’s *563Committee’s study would not countenance a functional discount completely untethered to either the supplier’s savings or the wholesaler’s costs. The longstanding principle that functional discounts provide no safe harbor from the Act is likewise evident from the practice of the Federal Trade Commission, which has, while permitting legitimate functional discounts, proceeded against those discounts which appeared to be subterfuges to avoid the Act’s restrictions. See, e. g., In re Sherwin Williams Co., 36 F. T. C. 25, 70-71 (1943) (finding a violation of the Act by paint manufacturers who granted “functional or special discounts to some of their dealer-distributors on the purchases of such dealer-distributors which are resold by such dealer-distributors directly to the consumer through their retail departments or branch stores wholly owned by them”); In re Ruberoid Co., 46 F. T. C. 379, 386, ¶ 5 (1950) (liability appropriate when functional designations do not always indicate accurately “the functions actually performed by such purchasers”), aff’d, 189 F. 2d 893 (CA2 1951), rev’d on rehearing, 191 F. 2d 294, aff’d, 343 U. S. 470 (1952).20 See also, e. g., In re Doubleday & *564Co., 52 F. T. C. 169, 209 (1955) (“[T]he Commission should tolerate no subterfuge. Only to the extent that a buyer actually performs certain functions, assuming all the risks and costs involved, should he qualify for a compensating discount. The amount of the discount should be reasonably related to the expenses assumed by the buyer”); In re General Foods Corp., 52 F. T. C. 798, 824-825 (1956) (“A seller is not forbidden to sell at different prices to buyers in different functional classes and orders have been issued permitting lower prices to one functional class as against another, provided that injury to commerce as contemplated in the law does not result,” but “[t]o hold that the rendering of special services ipso facto [creates] a separate functional classification would be to read Section 2(d) out of the Act”); In re Boise Cascade Corp., 107 F. T. C. 76, 212, 214-215 (1986) (regardless of whether the FTC has judged functional discounts by reference to the supplier’s savings or the buyer’s costs, the FTC has recognized that “functional discounts may usually be granted to customers who operate at different levels of trade, and thus do not compete with each other, without risk of secondary line competitive injury under the Act”), rev’d on other grounds, 267 U. S. App. D. C. 124, 837 F. 2d 1127 (1988).21 *565Cf. FLM Collision Parts, Inc. v. Ford Motor Co., 543 F. 2d 1019, 1027 (CA2 1976) (“We do not suggest or imply that, if a manufacturer grants a price discount or allowance to its wholesalers (whether or not labelled ‘incentive’), which has the purpose or effect of defeating the objectives of the Act, § 2(a)’s language may not be construed to defeat it”); C. Edwards, Price Discrimination Law 286-348 (1959) (analyzing cases).22
Most of these cases involved discounts made questionable because offered to “complex types of distributors” whose “functions became scrambled.” Doubleday & Co., 52 F. T. C., at 208. This fact is predictable: Manufacturers will more likely be able to effectuate tertiary line price discrimination through functional discounts to a secondary line buyer when *566the favored distributor is vertically integrated. Nevertheless, this general tendency does not preclude the possibility that a seller may pursue a price discrimination strategy despite the absence of any discrete mechanism for allocating the favorable price discrepancy between secondary and tertiary line recipients.23
Indeed, far from constituting a novel basis for liability under the Act, the fact pattern here reflects conduct similar to that which gave rise to Perkins v. Standard Oil Co. of Cal., 395 U. S. 642 (1969). Perkins purchased gas from Standard, and was both a distributor and a retailer. He asserted that his retail business had been damaged through two violations of the Act by Standard: First, Standard had sold directly to its own retailers at a price below that charged to Perkins; and, second, Standard had sold to another distributor, Signal, which sold gas to Western Hyway, which in turn *567sold gas to Regal, a retailer in competition with Perkins.24 The question presented was whether the Act—which refers to discriminators, purchasers, and their customers—covered injuries to competition between purchasers and the customers of customers of purchasers. Id., at 646-647. We held that a limitation excluding such “fourth level” competition would be “wholly an artificial one.” Id., at 647. We reasoned that from “Perkins’ point of view, the competitive harm done him by Standard is certainly no less because of the presence of an additional link in this particular distribution chain from the producer to the retailer.”25 The same may justly be said in this case. The additional link in the distribution chain does not insulate Texaco from liability if Texaco’s excessive discount otherwise violated the Act.26
*568Nor should any reader of the commentary on functional discounts be much surprised by today’s result. Commentators have disagreed about the extent to which functional discounts are generally or presumptively allowable under the Robinson-Patman Act. They nevertheless tend to agree that in exceptional cases what is nominally a functional discount may be an unjustifiable price discrimination entirely within the coverage of the Act.27 Others, like Frederick *569Rowe, have asserted the legitimacy of functional discounts in more sweeping terms,28 but even Rowe concedes the existence of an “exception to the general rule.” Rowe 174, n. 7; id., at 195-205.29
We conclude that the commentators’ analysis, like the reasoning in Perkins and like the Federal Trade Commission’s practice, renders implausible Texaco’s contention that holding it liable here involves some departure from established understandings. Perhaps respondents’ case against Texaco *570rests more squarely than do most functional discount cases upon direct evidence of the seller’s intent to pass a price advantage through an intermediary. This difference, however, hardly cuts in Texaco’s favor. In any event, the evidence produced by respondents also shows the scrambled functions which have more frequently signaled the illegitimacy under the Act of what is alleged to be a permissible functional discount. Both Gull and Dompier received the full discount on all their purchases even though most of their volume was resold directly to consumers. The extra margin on those sales obviously enabled them to price aggressively in both their retail and their wholesale marketing. To the extent that Dompier and Gull competed with respondents in the retail market, the presumption of adverse effect on competition recognized in the Morton Salt case becomes all the more appropriate. Their competitive advantage in that market also constitutes evidence tending to rebut any presumption of legality that would otherwise apply to their wholesale sales.
The evidence indicates, moreover, that Texaco affirmatively encouraged Dompier to expand its retail business and that Texaco was fully informed about the persistent and marketwide consequences of its own pricing policies. Indeed, its own executives recognized that the dramatic impact on the market was almost entirely attributable to the magnitude of the distributor discount and the hauling allowance. Yet at the same time that Texaco was encouraging Dompier to integrate downward, and supplying Dompier with a generous discount useful to such integration, Texaco was inhibiting upward integration by the respondents: Two of the respondents sought permission from Texaco to haul their own fuel using their own tank wagons, but Texaco refused. The special facts of this case thus make it peculiarly difficult for Texaco to claim that it is being held liable for the independent pricing decisions of Gull or Dompier.
*571As we recognized in Falls City Industries, “the competitive injury component of a Robinson-Patman Act violation is not limited to the injury to competition between the favored and the disfavored purchaser; it also encompasses the injury to competition between their customers.” 460 U. S., at 436. This conclusion is compelled by the statutory language, which specifically encompasses not only the adverse effect of price discrimination on persons who either grant or knowingly receive the benefit of such discrimination, but also on “customers of either of them.” Such indirect competitive effects surely may not be presumed automatically in every functional discount setting, and, indeed, one would expect that most functional discounts will be legitimate discounts which do not cause harm to competition. At the least, a functional discount that constitutes a reasonable reimbursement for the purchasers’ actual marketing functions will not violate the Act. When a functional discount is legitimate, the inference of injury to competition recognized in the Morton Salt case will simply not arise. Yet it is also true that not every functional discount is entitled to a judgment of legitimacy, and that it will sometimes be possible to produce evidence showing that a particular functional discount caused a price discrimination of the sort the Act prohibits. When such anti-competitive effects are proved—as we believe they were in this case—they are covered by the Act.30
VI
At the trial respondents introduced evidence describing the diversion of their customers to specific stations supplied by Dompier. Respondents’ expert testimony on damages also focused on the diversion of trade to specific Dompier-supplied stations. The expert testimony analyzed the entire *572damages period, which ran from 1972 and 1981 and included a period prior to 1974 when Dompier did not own any retail stations (although the jury might reasonably have found that Dompier controlled the Red Carpet stations owned by its president from the outset of the damages period). Moreover, respondents offered no direct testimony of any diversion to Gull and testified that they did not even know that Gull was being supplied by Texaco. Texaco contends that by basing the damages award upon an extrapolation from data applicable to Dompier-supplied stations, respondents necessarily based the award upon the consequences of pricing decisions made by independent customers of Dompier. Texaco argues that the damages award must therefore be judged excessive as a matter of law.
Even if we were to agree with Texaco that Dompier was not a retailer throughout the damages period, we could not accept Texaco’s argument. Texaco’s theory improperly blurs the distinction between the liability and the damages issues. The proof established that Texaco’s lower prices to Gull and Dompier were discriminatory throughout the entire 9-year period; that at least Gull, and apparently Dompier as well, was selling at retail during that entire period; that the discounts substantially affected competition throughout the entire market; and that they injured each of the respondents. There is no doubt that respondents’ proof of a continuing violation of the Act throughout the 9-year period was sufficient. Proof of the specific amount of their damages was necessarily less precise. Even if some portion of some of respondents’ injuries may be attributable to the conduct of independent retailers, the expert testimony nevertheless provided a sufficient basis for an acceptable estimate of the amount of damages. We have held that a plaintiff may not recover damages merely by showing a violation of the Act; rather, the plaintiff must also “make some showing of actual injury attributable to something the antitrust laws were designed to prevent. Perkins v. Standard Oil Co., 395 U. S. *573642, 648 (1969) (plaintiff ‘must, of course, be able to show a causal connection between the price discrimination in violation of the Act and the injury suffered’).” J. Truett Payne Co. v. Chrysler Motors Corp., 451 U. S., at 562. At the same time, however, we reaffirmed our “traditional rule excusing antitrust plaintiffs from an unduly rigorous standard of proving antitrust injury.” Id., at 565. See also Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U. S. 100, 123-124 (1969); Bigelow v. RKO Radio Pictures, Inc., 327 U. S. 251, 264-265 (1946).31 Moreover, as we have noted, Texaco did not object to the instructions to the jury on the damages issue. A possible flaw in the jury’s calculation of the amount of damages would not be an appropriate basis for granting Texaco’s motion for a judgment notwithstanding the verdict.
The judgment is affirmed.
It is so ordered.
“Q. Did you have any conversations with Texaco during this period of time encouraging you to—Dompier Oil Company to change its emphasis and to move into the retail business? A. Yes, we did.
“Q. Would you tell the jury about that? [A.] Well, at various times Texaco encouraged us to begin supplying retail service stations. In the early Seventies they did that, and then as time went on, they encouraged us to own the stations that we were supplying; in other words, to try to control our own retail business. And beginning about 1974—we did purchase a station in ’74 and some more in ’75 and we began operating those as company operations with salaried company employees.” Id., at 116-117.
“Q. That would have been a rate—that if you had hired a common carrier to haul the product for you, you would have paid them to haul it? A. That’s right.
“Q. And do you understand—to your understanding does that common carrier rate have a built-in-profit? A. I am sure that it does.
“Q. Did you find it to be an advantage to you to be hauling your own product? A. Yes.” Id., at 126.
At trial one of Texaco’s defenses was based on its obligation to comply with certain federal regulations during periods of shortage. In one of its communications to the Federal Government, a Texaco vice president wrote, in part:
“We believe that the dramatic shift in gasoline sales from the independent retailer classes of purchaser to the independent distributor classes of purchaser can be explained almost entirely by the magnitude of the distributor discount and the hauling allowance.” Id., at 413.
Texaco had argued that its pricing practices were mandated by federal regulations and that its sales in the Spokane market were not “in commerce” within the meaning of the Act.
Section 2(a) of the Act provides in part:
“That nothing herein contained shall prevent differentials which make only-due allowance for differences in the cost of manufacture, sale, or delivery-resulting from the differing methods or quantities in which such commodities are to such purchasers sold or delivered.” 15 U. S. C. § 13(a).
Section 2(b) of the Act provides in part:
“Provided, however, That nothing herein contained shall prevent a seller rebutting the prima-facie case thus made by showing that his lower price or the furnishing of services or facilities to any purchaser or purchasers was made in good faith to meet an equally low price of a competitor, or the services or facilities furnished by a competitor.” 15 U. S. C. § 13(b).
The award to each particular respondent of course differed. The awards represented an average of $5,486.59 per year for each of the respondents.
“While there is a serious question as to whether Dompier was entitled to a ‘functional discount’ on the gas it resold at retail, compare Mueller Co., 60 F. T. C. 120 (1962), aff’d, 323 F. 2d 44 (7th Cir. 1963), cert. denied, 377 U. S. 923 . . . (1964) (entitlement to functional discount based on resale level) with Doubleday and Co., 52 F. T. C. 169 (1955) (entitlement to functional discount based on level of purchase), the court assumes, arguendo, that the mere fact that Dompier retailed the gas does not preclude a ‘functional discount.’” 634 F. Supp. 34, 37, n. 4 (ED Wash. 1985) (emphasis in original).
“Secondly, the functional discounts negatively affected competition because they were, in part, reflected in the favored purchasers’ (or their customers’) retail prices. In other words, the discount was not consumed or absorbed at the level of the favored buyers; rather, the amount of the discount (or a significant portion) appeared in the favored purchasers’ retail price, or in the favored purchasers’ price to their customers and in their customers’ retail prices. Under such circumstances, the otherwise innocuous nature and presumed legality of functional discounts is rebutted, for it is universally recognized that a functional discount remains legal only to the extent it acts as compensation for the functions performed by the favored buyer. See 3 Kintner & Bauer, Federal Antitrust Law 309-10 (1983); Rill, Availability and Functional Discounts Justifying Discriminatory Pricing, 53 Antitrust L. J. 929, 939-41 (1985). The discount must ‘be reasonably related to the expenses assumed by the [favored] buyer’ and the discount ‘should not exceed the cost of . . . the function [the favored buyer] actually performs . . . ’ Doubleday and Company, 52 F. T. C. at 209, cited in Boise Cascade Corp., Docket No. 9133, slip op. at 117 (Feb. 14, 1984) (initial decision). If the discount exceeds such costs, it cannot be justified as a functional discount, particularly where, as here, the excess has a negative effect on competition.
“In this case Texaco made no serious attempt to quantitatively justify its functional discounts. While a precise accounting of the value of the performed functions is not mandated, merely identifying some of the functions is not sufficient. There is no substantial evidence to support Texaco’s position that the discounts were justified.” Id., at 38 (footnote omitted).
In their brief filed as amici curiae, the United States and the Federal Trade Commission suggest the following definition of “functional discount,” *555which is adequate for our discussion: “A functional discount is one given to a purchaser based on its role in the supplier’s distributive system, reflecting, at least in a generalized sense, the services performed by the purchaser for the supplier.” Brief for United States et al. as Amici Curiae 10 (filed Aug. 3, 1989).
The legislative history indicates that earlier drafts of the Act did include such a proviso. See, e. g., Shniderman, “The Tyranny of Labels”—A Study of Functional Discounts Under the Robinson-Patman Act, 60 Harv. L. Rev. 571, 583-586, and nn. 40-57 (1947). The deletion of this exception for functional discounts has ambiguous significance. It may be, as one commentator has suggested, that the circumstances of the Act’s passage “must have conveyed to the congressional mind the realization that the judiciary and the FTC would view what had occurred as a narrowing of the gates through which the functional classification plan of a seller had to pass to come within the law.” Id., at 588. In any event, the deletion in no way detracts from the blunt direction of the statutory text, which indicates that any price discrimination substantially lessening competition will expose the discriminator to liability, regardless of whether the discriminator attempts to characterize the pricing scheme as a functional discount.
See n. 6, supra.
Texaco has not contested here the proposition that branded gas and unbranded gas are of like grade and quality. See FTC v. Borden Co., 383 U. S. 637, 645-646 (1966) (“‘[T]he economic factors inherent in brand names and national advertising should not be considered in the jurisdictional inquiry under the statutory “like grade and quality” test’ ”).
It has proved useful in Robinson-Patman Act cases to distinguish among “the probable impact of the [price] discrimination on competitors of the seller (primary-line injury), on the favored and disfavored buyers (second-line injury), or on the customers of either of them (third-line injury).” See 3 E. Kintner & J. Bauer, Federal Antitrust Law § 20.9, p. 127 (1983).
“In the Committee’s view, imposing on any dual supplier a legal responsibility for the resale policies and prices of his independent distributors contradicts basic antitrust policies. Resale-price fixing is incompatible with the tenets of a free and competitive economy. What is more, the arrangements necessary for policing, detecting, and reporting price cutters may be illegal even apart from the resale-price agreement itself. And even short of such arrangements, a conscious adherence in a supplier’s sales to retail customers to the price quotations by independent competing *560distributors is hardly feasible as a matter of business operation, or safe as a matter of law.” Report of the Attorney General’s National Committee to Study the Antitrust Laws 206-207 (1955) (footnotes omitted).
“In our view, to relate discounts or prices solely to the purchaser’s resale activities without recognition of his buying functions thwarts competition and efficiency in marketing. It compels affirmative discrimination against a substantial class of distributors, and hence serves as a penalty on integration. If a businessman actually fulfills the wholesale function by relieving his suppliers of risk, storage, transportation, administration, etc., his performance, his capital investment, and the saving to his suppliers, are unaffected by whether he also performs the retailing function, or any number of other functions. A legal rule disqualifying him from discounts recognizing wholesaling functions actually performed compels him to render these functions free of charge.” Id., at 207.
In theory, a supplier could try to defend a functional discount by invoking the Act’s cost justification defense, but the burden of proof with respect to the defense is upon the supplier, and interposing the defense “has proven difficult, expensive, and often unsuccessful.” 3 E. Kintner & J. Bauer, Federal Antitrust Law § 23.19, pp. 366-367 (1983). Moreover, to establish the defense a “seller must show that the price reductions given did not exceed the actual cost savings,” id., § 23.10, p. 345, and this requirement of exactitude is ill suited to the defense of discounts set by reference to legitimate, but less precisely measured, market factors. Cf. Calvani, Functional Discounts Under the Robinson-Patman Act, 17 B. C. Ind. & Com. L. Rev. 543, 546, n. 16 (1976) (distinguishing functional discounts from cost justified price differences); Report of the Attorney General’s National Commmittee on the Antitrust Laws, at 171 (“[T]he cost defense has proved largely illusory in practice”).
Discounters will therefore likely find it more useful to defend against claims under the Act by negating the causation element in the case against them: A legitimate functional discount will not cause any substantial lessening of competition. The concept of substantiality permits the causation inquiry to accommodate a notion of economic reasonableness with respect to the pass-through effects of functional discounts, and so provides a latitude denied by the cost justification defense. Cf. Shniderman, 60 Harv. L. Rev., at 603-604 (substantiality defense in functional discount cases). We thus find ourselves in substantial agreement with the view that:
“Conceived as a vehicle for allowing differential pricing to reward distributive efficiencies among customers operating at the same level, the cost justification defense focuses on narrowly defined savings to the seller derived from the different method or quantities in which goods are sold or delivered to different buyers . . . . Moreover, the burden of proof as to the cost *562justification defense is on the seller charged with violating the Act, whereas the burden of proof remains with the enforcement agency or plaintiff in circumstances involving functional discounts since functional pricing negates the probability of competitive injury, an element of a prima facie case of violation.” Rill, Availability and Functional Discounts Justifying Discriminatory Pricing, 53 Antitrust L. J. 929, 935 (1985) (footnotes omitted).
Texaco continues the argument by summoning a parade of horribles whose march Texaco believes is at issue in this case: According to Texaco, the Court of Appeals’ rule “would multiply distribution costs, rigidify and increase consumer prices, encourage resale price maintenance in violation of the Sherman Act. . . , and jeopardize the businesses of wholesalers.” Brief for Petitioner 14.
See also, e. g., In re Whiting, 26 F. T. C. 312, 316, ¶ 3 (1938) (functional classification of customers involved unlawful price discrimination because of functional overlap); In re Standard Oil Co., 41 F. T. C. 263 (1945), modified and aff’d, 173 F. 2d 210, 217 (CA7 1949) (“The petitioner should be liable if it sells to a wholesaler it knows or ought to have known . . . is using or intends to use [the wholesaler’s] price advantage to undersell the petitioner in its prices made to its retailers”), rev’d and remanded on other grounds, 340 U. S. 231 (1951).
In the Standard Oil case, the FTC itself on remand dropped the part of its order prohibiting Standard Oil from giving functional discounts. See C. Edwards, Price Discrimination Law 309 (1959). The FTC’s pre-remand theory in the Standard Oil case has of course been the subject of harsh criticism. See, e. g., Report of the Attorney General’s National Committee to Study the Antitrust Laws, at 206. Much, if not all, of this criticism rests upon the view that, under the FTC’s Standard Oil ruling, a “supplier is charged with legal responsibility for the middlemen’s pricing tactics, and hence must control their resale prices lest they undercut him to the unlawful detriment of his directly purchasing retailers. Alternatively, *564the seller may forego his operational freedom by matching his quotations to retailers with theirs.” Ibid. Nothing in our opinion today should be read to condone or approve such a result.
See also In re Mueller Co., 60 F. T. C. 120, 127-128 (1962) (refusing to make allowance for functional discounts in any way that would “add a defense to a prima facie violation of Section 2(a) which is not included in either Section 2(a) or Section 2(b)”), aff’d, 323 F. 2d 44 (CA7 1963), cert. denied, 377 U. S. 923 (1964). The FTC in Mueller expressly disavowed dicta from Doubleday suggesting that functional discounts are per se legal if justified by the buyer’s costs. Mueller held that the discounts were controlled instead by the reasoning propounded in General Foods, which refers to the value of the services to the supplier giving the discount. 60 F. T. C., at 127-128.
We need not address the relative merits of Mueller and Doubleday in order to resolve the case before us. We do, however, reject the require*565ment of exactitude which might be inferred from Doubleday’s dictum that a functional discount offered to a buyer “should not exceed the cost of that part of the function he actually performs on that part of the goods for which he performs it.” 52 F. T. C., at 209. As already noted, a causation defense in a functional discount case does not demand the rigorous accounting associated with a cost justification defense.
The Government’s position in this case does not contradict this course of decision. The Government’s amicus brief on Texaco’s behalf criticizes the Court of Appeals opinion on the theory that it “would require a supplier to show that a functional discount is justified by the wholesaler’s costs,” and that it imposed “liability for downstream competitive effects of legitimate functional discounts. ” Brief for United States et al. as Amici Curiae 6 (filed Aug. 3, 1989). Cf. Boise Cascade Corp. v. FTC, 267 U. S. App. D. C. 124, 138-140, 837 F. 2d 1127, 1141-1143 (1988) (summarizing debate about relevance of buyer’s costs to defense of functional discounts). If the Court of Appeals were indeed to have endorsed either of these rules, it would have departed perceptibly from the mainstream of the FTC’s reading of the Act. We need not decide whether the Government’s interpretation of the Court of Appeals opinion is correct, for we affirm its judgment for reasons that do not entail the principles criticized by the Government. Indeed, the Government itself opposed the petition for certiorari in this case on the ground that “we do not think that this case on its facts presents the broad issue that petitioner discusses (whether a supplier must show that its discounts to wholesalers relative to retailers are cost based).” Brief for United States as Amicus Curiae 12 (filed May 16, 1989).
The seller may be willing to accept any division of the price difference so long as some significant part is passed on to the distributor’s customers. Although respondents here did not need to show any benefit to Texaco from the price discrimination scheme in order to establish a violation of the Act, one possibility is indicated by the brief filed amicus curiae by the Service Station Dealers of America (SSDA), an organization representing both stations supplied by independent jobbers and stations supplied directly by sellers. See Brief for SSDA as Amicus Curiae 1-2. SSDA suggests that an indirect price discount to competitors may be used to force directly supplied franchisees out of the market, and so to circumvent federal restrictions upon the termination of franchise agreements. See 92 Stat. 324-332, 15 U. S. C. §§ 2801-2806.
One would expect that—absent a safe harbor rule making functional discounts a useful means to engage in otherwise unlawful price discrimination-excessive functional discounts, of the sort in evidence here would be rare. As the Government correctly observes, “[t]his case appears to reflect rather anomalous behavior on the part of the supplier.” Brief for United States et al. as Amici Curiae 17, n. 15 (filed Aug. 3, 1989). See also Brief for United States as Amicus Curiae 15 (filed May 16, 1989) (“[M]arket forces should tend to discourage a supplier from offering independent wholesalers discounts that would allow them to undercut the supplier’s own retail customers”).
Much of Perkins’ case parallels that of respondents. “There was evidence that Signal received a lower price from Standard than did Perkins, that this price advantage was passed on, at least in part, to Regal, and that Regal was thereby able to undercut Perkins’ price on gasoline. Furthermore there was evidence that Perkins repeatedly complained to Standard officials that the discriminatory price advantage given Signal was being passed down to Regal and evidence that Standard officials were aware that Perkins’ business was in danger of being destroyed by Standard’s discriminatory practices. This evidence is sufficient to sustain the jury’s award of damages under the Robinson-Patman Act.” 395 U. S., at 649.
We added: “Here Standard discriminated in price between Perkins and Signal, and there was evidence from which the jury could conclude that Perkins was harmed competitively when Signal’s price advantage was passed on to Perkins’ retail competitor Regal. These facts are sufficient to give rise to recoverable damages under the Robinson-Patman Act.” Id., at 648.
In fact, the principle applied in Perkins—that we will not construe the Robinson-Patman Act in a way that “would allow price discriminators to avoid the sanctions of the Act by the simple expedient of adding an additional link to the distribution chain,” id., at 647—seems capable of governing this case as well. It might be possible to view Perkins as standing for a narrower proposition, either because Signal apparently exercised majority control over the intermediary, Western Hyway, and its retailer, Regal, see id., at 651 (Marshall, J., concurring in part and dissenting in part), *568or because Standard did not assert that its price to Signal reflected a “functional discount.” However, as the Perkins dissent pointed out, ibid., the Perkins majority did not put any such limits on the principle it declared.
See, e. g., Celnicker & Seaman, Functional Discounts, Trade Discounts, Economic Price Discrimination and the Robinson-Patman Act, 1989 Utah L. Rev. 813, 857 (1989) (concluding that “[t]rade discounts often are manifestations of economic price discrimination. . . . If a trade discount violates the normal competitive disadvantage criteria used under the Act, no special devices should be employed to protect it”); Rill, 53 Antitrust L. J., at 940-941 (“Although it is entirely appropriate for the FTC and the courts to insist that some substantial services be performed in order for a buyer to earn a functional discount, a requirement of precise mathematical equiva lency makes no sense”); 3 E. Kintner & J. Bauer, Federal Antitrust Law 318-320, and n. 305 (1983) (“Functional discounts . . . are usually deemed lawful,” but this usual rule is subject to exception in cases, “arising in unusual circumstances,” when the seller’s “discrimination caused” the tertiary line injury); Calvani, 17 B. C. Ind. & Com. L. Rev., at 549, and n. 26 (discounts to wholesalers are generally held not to injure competition, but this rule is subject to qualifications, and “[p]erhaps the most important caveat focuses on the situation where the seller sells to both resellers and consumers and the resellers pass on to their consumers all or part of the wholesaling functional discount”); C. Edwards, Price Discrimination Law 312-313 (1959) (“It is not surprising that from time to time the Commission has been unable to avoid finding injurious discrimination between direct and indirect customers nor to avoid corrective orders that sought to define the gap between prices at successive levels of distribution”); Kelley, Functional Discounts Under the Robinson-Patman Act, 40 Calif. L. Rev. 526, 556 (1952) (concluding that the “characterization of a price differential between two purchasers as a functional or trade discount accords it no cloak of immunity from the prohibitions of the Robinson-Patman Act”); Shniderman, 60 Harv. L. Rev., at 599-600 (Commission’s approach to functional discounts “may have been influenced by the possibility of subtle price *569discriminating techniques through the employment of wholesalers receiving more than ample discount differentials”).
Professor Edwards, among others, describes the status of functional discounts under the Robinson-Patman Act with clear dissatisfaction. He complains that “[t]he failure of the Congress to cope with the problem . . . has left the Commission an impossible job in this type of case.” Price Discrimination Law, at 313. He adds that the Commission’s “occasional proceedings” have been attributed to the “Commission’s wrong-headedness.” Id., at 312. Professor Edwards’ observations about the merits of the statute and about prosecutorial discretion are obviously irrelevant to our own inquiry. Unlike scholarly commentators, we have a duty to be faithful to congressional intent when interpreting statutes and are not free to consider whether, or how, the statute should be rewritten.
“In practice, the competitive effects requirement permits a supplier to quote different prices between different distributor classes—so long as those who are higher up (nearer the supplier) on the distribution ladder pay less than those who are further down (nearer the consumer).” F. Rowe, Price Discrimination Under the Robinson-Patman Act 174 (1962) (footnote omitted) (hereinafter Rowe); see also id., at 178.
Rowe, writing prior to this Court’s Perkins decision, describes the exception, which he identifies with the Standard Oil cases, as “of dubious validity today.” Rowe 196. Rowe’s analysis is flawed because he assumes that seller liability for tertiary line implications of wholesaler discounts must follow the logic of the Standard Oil complaint, and likewise assumes that this logic exposes to liability any seller who fails to monitor the resale prices of its wholesaler. Rowe 204. Indeed, Rowe’s own discussion suggests one defect in his argument: Legitimate wholesaler discounts will usually be insulated from liability by an absence of evidence on the causation issue. Id., at 203-204. In any event, nothing in our opinion today endorses a theory of liability under the Robinson-Patman Act for functional discounts so broad as the theory Rowe draws from Standard Oil.
The parties do not raise, and we therefore need not address, the question whether the inference of injury to competition might also be negated by evidence that disfavored buyers could make purchases at a reasonable discount from favored buyers.
In J. Truett Payne, 451 U. S., at 565-566, we quoted with approval the following passage:
“[D]amage issues in these cases are rarely susceptible of the kind of concrete, detailed proof of injury which is available in other contexts. The Court has repeatedly held that in the absence of more precise proof, the factfinder may 'conclude as a matter of just and reasonable inference from the proof of defendants’ wrongful acts and their tendency to injure plaintiffs’ business, and from the evidence of the decline in prices, profits and values, not shown to be attributable to other causes, that defendants’ wrongful acts had caused damage to the plaintiffs.’ Bigelow v. RKO Pictures, Inc., [327 U. S.], at 264. See also Eastman Kodak Co. v. Southern Photo Materials Co., 273 U. S. 359, 377-379 (1927); Story Parchment Co. v. Paterson Parchment Paper Co., 282 U. S. 555, 561-566 (1931).” Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U. S., at 123-124.