Federal Trade Commission v. Borden Company/Dissent Stewart

by Potter Stewart

United States Supreme Court

383 U.S. 637

FEDERAL TRADE COMMISSION, Petitioner,  v.  The BORDEN COMPANY.

 Argued: Jan. 19, 1966. --- Decided: March 23, 1966

Mr. Justice STEWART, with whom Mr. Justice HARLAN joins, dissenting.

I cannot agree that mere physical or chemical identity between premium and private label brands is, without more, a sufficient basis for a finding of 'like grade and quality' within the meaning of § 2(a) of the Robinson-Patman Act. The conclusion that a product that travels at a premium in the marketplace is of 'like grade and quality' with products of inferior commercial value is not required by the language of the Robinson-Patman Act, by its logic, or by its legislative history.

It is undisputed that the physical attributes and chemical constituents of Borden's premium and private label brands of evaporated milk are identical. It is also undisputed that the premium and private label brands are not competitive at the same price, and that if the private label milk is to be sold at all, it must be sold at prices substantially below the price commanded by Borden's premium brand. ^[1] This simple market fact no more than reflects the obvious economic reality that consumer preferences can and do create significant commercial distinctions between otherwise similar products. By pursuing product comparison only so far as the result of laboratory analysis, the Court ignores a most relevant aspect of the inquiry into the question of 'like grade and quality' under § 2(a): Whether the products are different in the eyes of the consumer. ^[2]

There is nothing intrinsic to the concepts of grade and quality that requires exclusion of the commercial attributes of a product from their definition. The product purchased by a consumer includes not only the chemical components that any competent laboratory can itemize, but also a host of commercial intangibles that distinguish the product in the marketplace. ^[3] The premium paid for Borden brand milk reflects the consumer's awareness, promoted through advertising, that these commercial attributes are part and parcel of the premium product he is purchasing. ^[4] The record in the present case indicates that wholesale purchasers of Borden's private label brands continued to purchase the premium brand in undiminished quantities. The record also indicates that retail purchasers who bought the premium brand did so with the specific expectation of acquiring a product of premium quality. ^[5] Contrary to the Courts suggestion ante, p. 644, this consumer expectation cannot accurately be characterized as a misapprehension. Borden took extensive precautions to insure that a flawed product did not reach the consumer. ^[6] None of these precautions was taken for the private brand milk packed by Borden. ^[7] An important ingredient of the premium brand inheres in the consumer's belief, measured by past satisfaction and the market reputation established by Borden for its products, that tomorrow's can will contain the same premium product as that purchased today. To say, as the Court does, that these and other intangibles, which comprise an important part of the commercial value of a product, are not sufficient to confer on Borden's premium brand a 'grade' or 'quality' different from that of private label brands is to ignore the obvious market acceptance of that difference. '(C)ommercially the 'advertised' brands had come in the minds of the public to mean a different grade of milk. The public may have been wrong; * * * it may have been right * * *. But right or wrong, that is what it believed, and its belief was the important thing.' Borden's Farm Products Co. v. Ten Eyck, 11 F.Supp. 599, 601 (D.C.S.D.N.Y.) (opinion of L. Hand, J.). ^[8]

The spare legislative history of the Robinson-Patman Act is in no way inconsistent with a construction of § 2(a) that includes market acceptance in the test of 'like grade and quality.' That history establishes no more than that mere differences in brand or design, unaccompanied by any genuine physical, chemical, or market distinction, are insufficient to negate a finding of 'like grade and quality' under § 2(a). ^[9] Nothing that I have found in the legislative history speaks with precision to the sole issue before us here, the application of § 2(a) to physically or chemically identical products that are in fact differentiated by substantial market factors. ^[10]

Neither the remarks of Representative Patman, ante, p. 643, nor the letter of Mr. Teegarden, ante, p. 641, n. 4, supports the Court's conclusion that Congress intended physical and chemical identity to be the sole touchstone of 'like grade and quality.' Aside from the obviously casual nature of Mr. Patman's reply to the question concerning the effect of the Act on private label brands, ^[11] his remarks go embarrassingly further than the circumspect reading sought to be given them by the Court. On its face, Mr. Patman's statement makes the blanket assertion that all products of the same quality must be sold at the same price. As thus stated, premium brands would have to be sold at the same price as private label brands, regardless of injury to competition, cost justification, or other available defenses under the Act. These undifferentiated remarks are therefore of little assistance in the determination of congressional intent. Far from supporting the Court's interpretation of § 2(a), the final paragraph of the Teegarden letter suggests that Mr. Teegarden considered the bill to have no effect on a premium brand producer's decision to furnish private label brands to purchasers, so long as the private label brands were made available on the same terms to all purchasers. Mr. Teegarden's concern was with the prevention of discrimination between purchasers on the basis of artificial differences in brand. ^[12] That same concern, and no more, is all that may legitimately be read into the rejection by Congress of the proposal to add 'and brands' to the 'like grade and quality' provision in the bill. By rejecting that proposal, it can be inferred only that Congress contemplated 'no blanket exemption * * * for 'like' products which differed only in brand * * *, leaving open the application of the Act to differentiated products reflecting more than a nominal or superficial variation.' Rowe, Price Discrimination Under the Robinson-Patman Act 65 (1962).

The references in the legislative hearings and the House Committee Report to the Commission's decision in Goodyear Tire & Rubber Co., 22 F.T.C. 232, are equally inconclusive on the relevance of commercial acceptance to the determination of 'like grade and quality.' The striking aspect of that case is that Goodyear conceded that the differently branded tires involved in the proceeding were of like grade and quality, 22 F.T.C., at 290. Moreover, the tires purchased by Sears, Roebuck & Co. from Goodyear and sold under Sears' 'All State' label were advertised by Sears as obtained from 'the leading tire manufacturer' and 'the world's foremost tire manufacturer,' so that the market independence of Sears' private brand was compromised. Id., at 295, 297.

The other administrative precedents relied on by the Court also fail to establish any consistently settled interpretation by the Federal Trade Commission that physical identity is the sole touchstone of 'like grade and quality.' Those decisions singularly fail to focus on the significance of consumer preference as a relevant factor in the test of grade and quality. ^[13] Moreover, the Commission has itself explicitly resorted to consumer preference or marketability to resolve the issue of 'like grade and quality' in cases where minor physical variations accompany a difference in product brand. ^[14] The caprice of the Commission's present distinction thus invites Borden to incorporate slight tangible variations in its private label products, in order to bring itself within the Commission's current practice of considering market preferences in such cases.

The Commission's determination of 'like grade and quality' under § 2(a) in this case is seriously inconsistent with the position it has taken under § 2(b) in cases where as seller has presented the defense that he is in good faith meeting the equally low price of a competitor. The Commission decisions are clear that the 'meeting competition' defense is not available to a seller who reduces the price of his premium product to the level of nonpremium products sold by his competitors. The Commission decision under § 2(b) emphasize that market preference must be considered in determining whether a competitor is 'meeting' rather than 'beating' competition. In Standard Oil Co., 49 F.T.C. 923, 952, the Commission put it baldly:

'(I)n the retail distribution of gasoline public acceptance rather than chemical analysis of the product is the important competitive factor.' ^[15]

Could the Commission under § 2(b) now prevent Borden from reducing the price of its premium milk to the level of private label milk? I can see no way that it could, short of maintaining a manifestly unstable equilibrium between § 2(a) and § 2(b). By adopting a keyhole approach to § 2(a), the Court manages to escape resolution of the question, but it does so at the cost of casting grave doubt on what I had regarded as an important bulwark of § 2(b) against a recognized competitive evil.

The Court gives no substantial economic justification for its construction of § 2(a). ^[16] The principal rationale of the restriction of that section to commodities of 'like grade and quality' is simply that it is not feasible to measure discrimination and injury to competition where different products are involved. That rationale is as valid for economic as for physical variation between products. Once a substantial economic difference between products is found, therefore, the inquiry of the Commission should be ended, just as it is ended when a substantial physical difference is found.

In spite of the assertion of the Attorney General's Report quoted by the Court, it is unlikely that economic differences between premium and private label brands can realistically be taken into account by the Commission under the 'injury to competition' and 'cost justification' provisions of § 2(a). ^[17] Even if relevant cost data can be agreed upon, the cost ratio between Borden's premium and private label products is hardly the most significant factor in Borden's pricing decision and market return on those products. Moreover, even if price discrimination is found here, its effect on competition may prove even more difficult to determine than in more conventional cases of price discrimination under § 2(a). Cf. FTC v. Morton Salt Co., 334 U.S. 37, 68 S.Ct. 822, 92 L.Ed. 1196; United Biscuit Co. of America v. FTC, 350 F.2d 615 (C.A.7th Cir.).

The threat presented to primary line competition by Borden's distribution of premium and private label brands is unclear. No allegation was made that Borden has used its dominant position in the premium brand market to subsidize predatory price-cutting campaigns in the private label market. Borden packs its private label brands for national distribution, so that this case is essentially different from those in which geographical price discriminations are involved. Further, Borden's private label brands are aimed in part at a different, more price-conscious class of consumer. Because relevant economic factors differ in the premium and private label markets, conventional notions of price discrimination under the Robinson-Patman Act may not be applicable. ^[18] More important, Borden's extensive distribution of its private label brands has introduced significant low-cost competition for Borden's own premium product. Thus, the large retail chains and cooperative buyer organizations that are Borden's chief private label customers represent a significant source of countervailing power to the oligopoly patten of evaporated milk production. The rise of this sort of competition is well known in other parts of the food industry. ^[19] In these circumstances, the anticompetitive leverage against primary line competition available to Borden through its private label production is sharply curtailed. There is, therefore, no real resemblance in this case to the serious discriminatory practices that the Robinson-Patman Act was enacted to prevent.

The potential economic impact of Borden's distribution of private label brands on secondary line competition is equally ambiguous. It is true that a market test of 'like grade and quality' would enable Borden, so far as § 2(a) is concerned, to make private label brands selectively available to customers of its premium brand. Not all wholesale and retail dealers who carry Borden's premium brand would be able, as of right, to take advantage of Borden's private label production. But the Commission could still apply § 2(a) with full force against discriminations between private label customers. And the Government could still invoke § 2 of the Sherman Act or § 5 of the Federal Trade Commission Act to deal with other forms of price discrimination by Borden against its customers or competitors.

Under the Court's view of § 2(a), Borden must now make private label milk available to all customers of its premium brand. ^[20] But that interpretation of § 2(a) is hardly calculated to speed private label brands to the shelves of retailers. To avoid supplying a private label brand to a premium brand customer, Borden need only forgo further sales of its premium brand to that customer. It is, therefore, not unlikely that the Court's decision will foster a discrimination greater than that which it purports to eliminate, since retailers previously able to obtain the premium Borden brand but not a private label brand, may now find their access to the premium brand foreclosed as well.

In Automatic Canteen Co. of America v. FTC, 346 U.S. 61, 63, 73 S.Ct. 1017, 1019, 97 L.Ed. 1454, this Court cautioned against construction of the Robinson-Patman Act in a manner that might 'give rise to a price uniformity and rigidity in open conflict with the purposes of other antitrust legislation.' Today that warning goes unheeded. In the guise of protecting producers and purchasers from discriminatory price competition, the Court ignores legitimate market preferences and endows the Federal Trade Commission with authority to disrupt price relationships between products whose identity has been measured in the laboratory but rejected in the marketplace. I do not believe that any such power was conferred upon the Commission by Congress, and I would, therefore, affirm the judgment of the Court of Appeals.

Notes

1. For example, one wholesaler, a witness for the Commission, stated:
2. No suggestion is made that any of the private label brands involved in this case show significant commercial differentiation from one another. It is possible, of course, that by extensive promotion private label brands could achieve consumer acceptance equivalent to that of a premium brand. In that situation, the products would still be economically different under the market test of § 2(a) elucidated in this opinion, since the relevant comparison would exclude promotional efforts by persons other than the producer of the premium brand. Thus, promotional activities by customers of Borden in the present case could not affect the determination of 'like grade and quality' with regard to sales by Borden. Cf. Jordan, Robinson-Patman Act Aspects of Dual Distribution by Brand of Consumer Goods, 50 Cornell L.Q. 394, 406 407 (1965).
3. Cf. Chamberlin, The Theory of Monopolistic Competition 56 (8th ed. 1962):
4. For example, a grocer testified in the proceedings before the Commission that:
5. The results of a house-to-house survey conducted for Borden by National Analysts, Inc., indicated that consumers selected Borden's premium brand because of its superior quality. Comparable studies have reached a similar conclusion. Cf. 'Mom Feels Quality, not Ad Cost, Makes Brand Item Costlier, 'Good House' Reports,' Advertising Age, Dec. 7, 1964, p. 30.
6. Borden's Food Products Division maintained a staff of field representatives who inspected code-datings on cans of Borden brand milk in retail stores, in order to insure that older milk was sold first off the retailer's shelves. A witness for Borden testified that the principal dangers of long storage were discoloration of the milk, precipitation of calcium and other minerals, and separation and hardening of fat from the milk. As a further precaution against sales of defective milk, Borden dispatched its milk to wholesalers and retailers under a first-packed, first-shipped rotation plan that occasionally involved high-cost shipments from distant plants or warehouses. In addition, before shipment from a cold storage warehouse, Borden 'tempered' its premium brand milk in order to prevent condensation on the cans, which might have resulted in rust to the cans and damage to the labels.
7. As counsel for the respondent candidly stated on oral argument to the Court, 'The difference as to the private label brand packed by Borden is that, as to that product, the Borden Company washes its hands of it at the factory door.'
8. The Court's suggestion that the commentators are about equally divided upon the issue before us is somewhat misleading. It is true that the members of the Attorney General's National Committee to Study the Antitrust Laws, Report, pp. 156-159 (1955), were sharply divided as to whether significant consumer preferences should be taken into account under the 'like grade and quality' test of § 2(a). However, the very brief discussions of 'like grade and quality' in Austin, Price Discrimination and Related Problems under the Robinson-Patman Act 39 (2d ed. 1959); Patman, Complete Guide to the Robinson-Patman Act 34-35 (1963); and Edwards, The Price Discrimination Law 31, 463-464 (1959), are not addressed to the relevance of significant consumer preferences, and the minimal discussion in Seidman is at best ambiguous, Price Discrimination Cases, reprinted in 2 Hoffmann's Antitrust Law and Techniques 409, 427-428 (1963). Those cursory treatments go no further than the view, with which I wholly agree, that no blanket exemption from § 2(a) is available for private label brands. But that view in no sense disposes of the concrete issue presented in this case. Commentators who have in fact focussed on the significance of consumer preferences uniformly favor inclusion of commercial acceptance in the test of 'like grade and quality.' Rowe, Price Differentials and Product Differentiation: The Issues under the Robinson-Patman Act, 66 Yale L.J. 1 (1956); Rowe, Price Discrimination Under the Robinson-Patman Act 62-76 (1962); Cassady & Grether, The proper Interpretation of 'Like Grade and Quality' within the Meaning of Section 2(a) of the Robinson-Patman Act, 30 So.Cal.L.Rev. 241 (1957); Jordan, Robinson-Patman Act Aspects of Dual Distribution by Brand of Consumer Goods, 50 Cornell L.Q. 394 (1965).
9. The Court's suggestion, ante, p. 644, that a difference in label alone would exclude the reach of § 2(a) if a market test were accepted for 'like grade and quality' is no part of the present case and has never been offered as a serious interpretation of § 2(a). Nor is there any issue raised here as to whether, under a market test of § 2(a), dubious pricing and branding policy adopted by a seller could 'validate itself' and escape the Act by creating precarious distinctions in grade or quality. The price differential between Borden's premium and private label brands is concededly grounded upon a legitimate and stable market preference for the premium product. Moreover, the Commission's willingness to engage in the exhaustive analysis of injury to competition and cost justification under its 'physical identity' test of § 2(a) demonstrates that the Commission's resources would be more than adequate to determine the level of commercial preference sufficient to negate a finding of 'like grade and quality' under a market test of § 2(a).
10. Certain general language in the congressional reports may be taken, however, as supporting the interpretation that market factors are relevant in the construction of § 2(a). The Report of the House Committee on the Judiciary stated that the general object of the bill was 'to amend section 2 of the Clayton Act so as to suppress more effectually discriminations between customers of the same seller not supported by sound economic differences in their business positions * * *.' H.R.Rep. No. 2287, 74th Cong., 2d Sess., p. 7. (Emphasis added.) The Report of the Senate Committee on the Judiciary is phrased in substantially the same language. S.R.Rep. No. 1502, 74th Cong., 2d Sess., p. 3.
11. The remarks of Representative Patman were even more offhand than the opinion of the Court indicates. Prefacing the portion of his remarks quoted by the Court, Mr. Patman said, 'I only have a very short time, and I must finish my statement. I have not time to discuss that feature. * * *'
12. The predominant concern of Congress in enacting the Robinson-Patman amendments to the Clayton Act was to abolish the notorious price discriminations that infected the post-Depression economy, especially the blanket immunity then available for quantity discounts under § 2 of the Clayton Act. An obvious commercial evil at the time was the widespread practice of offering private label brands to favored customers at rates substantially lower than the rates offered to competing purchasers. The abortive attempt, vigorously opposed by Mr. Teegarden, to introduce 'and brands' into the 'like grade and quality' provision would have left that evil completely unremedied. Cf. 80 Cong.Rec. 8234-8236 (rejection of amendments proposing the addition of 'and design' and 'purchased under like conditions' to the 'like grade and quality' clause).
13. In Hansen Inoculator Co., 26 F.T.C. 303, and the two United States Rubber Co. cases, 28 F.T.C. 1489; 46 F.T.C. 998, the finding of 'like grade and quality' was either conceded by the respondent or not challenged. In addition, in Hansen Inoculator, there was significant evidence that the private label product was in fact trading on the reputation of the premium product. Further, in Hansen Inoculator, as in Page Dairy Co., 50 F.T.C. 395, it is doubtful that even the labels on the two products were distinguishable. In Whitaker Cable Corp., 51 F.T.C. 958, the resale prices of both products were identical, so that no commercial preference could have been proved in any event. Finally, in the first United States Rubber case and in Whitaker Cable Corp., there was substantial discrimination by the seller between various purchasers of the private label brands. In setting aside the order of the Commission in the present case, the Court of Appeals for the Fifth Circuit emphasized that in none of these cases was there any showing that the brand names affected the market price of the product sold.
14. Universal-Rundle Corp., CCH Trade Reg. Rep. Transfer Binder, 1963-1965, 16948, at p. 22003-22005 (F.T.C.Dkt. 8070, June 12, 1964) (differences in plumbing fixtures); Quaker Oats Co., CCH Trade Reg. Rep. Transfer Binder, 1963-1965, 17134, at p. 22215 (F.T.C.Dkt. 8112, Nov. 18, 1964) (differences in flour blends). Compare E. Edelmann & Co., 51 F.T.C. 978 (differences in automobile replacement parts); Bruce's Juices, Inc. v. American Can Co., D.C., 87 F.Supp. 895, aff'd 187 F.2d 919 (C.A.5th Cir.) (differences in size of juice cans); Champion Spark Plug Co., 50 F.T.C. 30 (differences in insulator and 'ribs' of spark plugs). Cf. Comment, Like Grade and Quality: Emergence of the Commercial Standard, 26 Ohio State L.J. 294, 296-302 (1965). The Commission appears at one time to have held that brand identity may create a presumption of 'like grade and quality,' regardless of the existence of physical differences between the products. General Foods Corp., 52 F.T.C. 798, 817; Atalanta Trading Corp., 53 F.T.C. 565, 571. In setting aside the Commission's order in Atalanta, the Court of Appeals for the Second Circuit stated that 'The test of products of like grade and quality was evolved to prevent emasculation of the section by a supplier's making artificial distinctions in his product but his does not mean that all distinctions are to be disregarded.' Atalanta Trading Corp. v. FTC, 2 Cir., 258 F.2d 365, 371. In a footnote to that opinion, the Court of Appeals indicated that price differences were among the distinctions to be considered. Id., at 371, n. 5. Cf. Rowe, Price Discrimination Under the Robinson-Patman Act 71-72 (1962).
15. See also Minneapolis-Honeywell Regulator Co., 44 F.T.C. 351, 396-397: 'To accept (the contrary) proposition would mean that any seller of a commodity which generally sells at a premium price may freely discriminate among its customers so long as it does not undercut the prices of competitors'; Anheuser-Busch, Inc., 54 F.T.C. 277, 302: 'It is evident that Budweiser could and did successfully command a premium price in the St. Louis market. * * * The test in such a case is not necessarily a difference in quality but the fact that the public is willing to buy the product at a higher price in a normal market'; Callaway Mills Co., sub nom. Bigelow-Sanford Carpet Co., CCH Trade Reg.Rep. Transfer Binder, 1963-1965, 16,800, at p. 21755 (F.T.C.Dkt. 7634, Feb. 10, 1964): 'Both the courts and the Commission have consistently denied the shelter of the (meeting competition) defense to sellers whose product, because of * * * intense public demand, normally commands a price higher than that usually received by sellers of competitive goods'; Standard Brands, Inc., 46 F.T.C. 1485, 1495; Gerber Products Co. v. Beech-Nut Life Savers, Inc., 160 F.Supp. 916, 920, 921-922 (D.C.S.D.N.Y.). Cf. Porto Rican American Tobacco Co. of Porto Rico v. American Tobacco Co., 30 F.2d 234, 237 (C.A.2d Cir.). In the present case, the Court of Appeals for the Fifth Circuit specifically refused to 'approve of the Commission's construing the Act inconsistently from one case to the next, as appears most advantageous to its position in a particular case.' 339 F.2d 133, at 139. See the comment of Commissioner Mason: 'First the Commission finds you guilty of price discrimination by disregarding popularity of goods, and finds the grade and quality of the commodities in question are the same; then they knock out your meeting of competition defense because your goods are more popular than others, even if the commodities in question are of like grade and quality.' Discriminate in Price between Different Purchasers of Commodities of Like Grade, Quality and Popularity, Proc. Am. Bar Assn. Section of Antitrust Law 82, 91-92 (Aug. 1953). Cf. Eine Kleine Juristische Schlummergeschichte, 79 Harv.L.Rev. 921, 928-929 (1966).
16. The Court's brief discussion of the adverse economic effect of the Fifth Circuit's ruling is concerned primarily with the supposed injury to secondary line competition. The present proceeding arose as the direct result of the primary line injury caused to midwestern packers of private label evaporated milk when Borden expanded its plants in Tennessee and South Carolina to include private label operation, but the opinion of the Court nowhere discusses such competition.
17. It is not clear that the 'injury to competition' and 'cost justification' issues will be reached on the remand. As the opinion of the Court suggests, ante, p. 646, the existence of price discrimination is an issue that remains open in the Court of Appeals. If Borden is able to demonstrate that the price differential between its premium and private label brands is not a price discrimination, the inquiry by the Commission is at an end, and no issue of injury to competition or cost justification under § 2(a) is reached. Nothing in FTC v. Anheuser-Busch, Inc., 363 U.S. 536, 80 S.Ct. 1267, 4 L.Ed.2d 1385, a case concerned only with territorial price discrimination, requires an equation in all circumstances between a price differential and price discrimination. So long as Borden makes private label brands available to all customers of its premium milk, it is unlikely that price discrimination within the meaning of § 2(a) can be made out. Boss Mfg. Co. v. Payne Glove Co., 71 F.2d 768, 770-771 (C.A.8th Cir.); Austin, Price Discrimination and Related Problems under the Robinson-Patman Act 21 (2d ed. 1959); Rowe, Price Discrimination Under the Robinson-Patman Act, supra, at 97-99.
18. Cf. Adelman, Price Discrimination as Treated in the Attorney General's Report, 104 U.Pa.L.Rev. 222, 228-230 (1955).
19. See Staff Report to the Federal Trade Commission, Economic Inquiry into Food Marketing, Part II, The Frozen Fruit, Juice and Vegetable Industry (1962); Jordan, supra, n. 8, at 413 417.
20. The Commission concedes that there is no evidence in the record that Borden refused to sell private label milk to any customer who specifically requested it. Borden's private label business in the period covered by these proceedings was substantial. In 1957, Borden sold 4,300,000 cases of its premium brand evaporated milk and 1,100,000 cases of private label milk (government and export business excluded); net sales of these products were $27,600,000 and $5,700,000 respectively. A major source of Borden's private label business was provided by cooperative associations of wholesalers and retailers, so that, in fact, there was an opportunity for large numbers of small retailers to compete in the sale of private label brands of evaporated milk obtained from Borden. One such group, whose purchases accounted for 11% of Borden's private label volume in 1957, had more than 1,000 retailer members. Not all retailers, however, availed themselves of the opportunity to market private label milk. One wholesaler testified that, a year after his private label brand had been offered to the 600 retail grocers in his service area, only 50 of the grocers had become regular customers.