We consider for the second time a petition by the Borden Company to review and set aside a cease-and-desist order of the Federal Trade Commission based on its decision that Borden violated Section 2(a) of the Clayton Act as amended by the Robinson-Patman Act, 15 U.S.C. Sec. 13(a).1 The order directs Borden to cease and desist from discriminating in price between purchasers of its Borden brand evaporated milk and purchasers of its private label evaporated milk, which is packaged under a brand owned by the purchaser. The only distinguishing feature in the manufacture of the milk is the label placed upon the can: otherwise, the milk is of the same chemical composition and is packed in the identical way. But because Borden is a nationally advertised brand which commands a consumer preference, reflected in the willingness to pay a premium price, the milk bearing the Borden label is sold by Borden, and at all levels of distribution, at a substantially higher price than the lesser known private label milk.2 It is this price difference initiated by Borden and its effect which the Commission held violated Sec. 2(a).

The threshold inquiry concerning a violation of Sec. 2(a) is whether the goods are 'of like grade and quality.' In our previous decision,3 we held that the marked consumer preference for the Borden brand was sufficient to differentiate the products and to place the price difference beyond the reach of Sec. 2(a). The Supreme Court reversed, holding that the economic factors inherent in brand names should not be considered in the jurisdictional inquiry under the 'like grade and quality' test but rather under the more flexible 'injury' and 'cost justification' provisions of the statute. For that purpose, the case was remanded for a resolution of the remaining issues which had been raised before us by Borden. 383 U.S. 637, 86 S. Ct. 1092, 16 L. Ed. 2d 153 (1966). These included challenges to (a) the Commission's finding of injury to competition, (b) the Commission's rejection of the cost-justification defense, and (c) the scope of the order.

Borden has been producing and packaging evaporated milk under both the Borden brand and private brands since 1938. The Borden brand milk is sold nationally at a uniform delivered price. On the other hand, the private brand milk is sold f.o.b. at several of the nine Borden plants, the price being figured pursuant to a cost-plus formula.

This controversy originated in 1956 and 1957 when Borden expanded its packaging of milk under private labels to plants situated in the South, notably in Tennessee and South Carolina, which theretofore had marketed milk only under the Borden label. As a result, some private label business previously held by certain competitors of Borden, relatively small canners located in the Midwest, was diverted to those southern Borden plants which had begun to package private label milk. It was this diversion of private label sales which precipitated the complaint against Borden.

The complaint charged that from January 1956 through March 1958, Borden had 'discriminated in price between different purchasers of its evaporated milk of like grade and quality by selling it to some of its purchasers at substantially lower prices than to other of its purchasers.' This discrimination in price was alleged to have dealt injury to the primary line of competition comprised of sellers and also to the secondary line of competition comprised of customers.

The hearing examiner concluded that the goods were of like grade and quality but that See. 2(a) had not been violated because of the failure to prove the requisite injury to either line of competition, and, in any event, that the price difference was cost justified. The Commission rejected both of the latter conclusions, vacated and set aside the examiner's order, and issued the cease-and-desist order.4

It is not disputed that a price discrimination, within the meaning of the statute, is present. However, we should point out that no overtones of business buccaneering are intended in the phrase 'discriminate in price'. In the context of Sec. 2(a), price discrimination means only a price difference, not an invidious price structure. Once the fact of a price difference is established, other provisions of the statute must be applied to determine whether the price difference is legal or illegal. FTC v. Anheuser-Busch, Inc., 363 U.S. 536, 549-550, 80 S. Ct. 1267, 4 L. Ed. 2d 1385 (1960). This distinction is important to the instant case because although there is a price difference, Borden is not in any sense guilty of predatory behavior similar to that which may accompany territorial price wars.5 Borden did not subsidize belowcost or unrealistically low prices on its private label milk with profits received from sales of the Borden brand. Nor does this case involve any device similar to the conventional volume discount.6 All of Borden's customers, large and small alike, paid the same prices for the Borden brand milk or for the private label milk. And although the Commission in its complaint may have implied that Borden favored some customers over others by a reference to private label customers as 'selected' customers, it was conceded by the Commission both before the Supreme Court, 383 U.S. at 661 n. 20, 86 S. Ct. 1092, and to us that there is no evidence in the record that Borden refused to sell private label milk to any customer who specifically requested it. The complaint was issued solely because Borden marketed milk of like grade and quality under private labels at prices lower than under the Borden brand.

We deal with the question whether the Commission's finding of competitive injury, a requisite element of a Sec. 2(a) violation, is supported by substantial evidence. Section 2(a) proscribes price discrimination whose effect 'may be substantially to lessen competition * * * in any line of commerce, or to injury * * * competition with any person who either grants or knowingly received the benefit of such discrimination, or with customers of either of them.' The statute seeks to protect against two types of injuries which may result from a seller's use of price differentials. First, it affords protection from a general injury to competition, usually with respect to the primary line. In this context, injury to a particular seller is to be distinguished from injury to the vigor of competition since here it is the latter to which the statute refers.7 American Oil Co. v. FTC, 325 F.2d 101, 104-105 (7th Cir. 1963), cert. denied, 377 U.S. 954, 84 S. Ct. 1631, 12 L. Ed. 2d 498 (1964); Anheuser-Busch, Inc. v. FTC, 289 F.2d 835, 840 (7th Cir. 1961). Second, by prohibiting price differentials whose effect may be substantially to injury competition with a customer, the statute does safeguard individual customers of a seller, as distinguished from competition generally, from a more narrow type of injury. See Foremost Dairies, Inc. v. FTC, 348 F.2d 674, 678 (5th Cir.), cert. denied, 382 U.S. 959, 86 S. Ct. 435, 15 L. Ed. 2d 362 (1965).8

The statute does not require a finding that the discriminations in price have in fact had the prescribed effect on competition, 'but only that there is a reasonable possibility that they 'may' have such an effect. Corn Products Refining Co. v. Federal Trade Comm., 324 U.S. 726, 742, 65 S. Ct. 961, 969, 89 L. Ed. 1320.' FTC v. Morton Salt Co., supra, 334 U.S. at 46, 68 S. Ct. 822, 828.

Seven Midwestern canners, competitors of Borden, testified in support of the complaint with respect to competitive injury to the primary line. Marketing only private label milk, they sell on a local basis without advertising. Their private label milk was sold on a delivered price basis, in contrast to Borden's method of selling its private label milk f.o.b. plant. The competitors complained that portions of their sales had been lost to Borden during the period in question. The record discloses that sales constituting about 7% Of their production, or roughly 240,000 cases, had been diverted to Borden. In essence, this business was attracted to Borden because it was selling private label milk cheaper than were these competitors. There is disagreement as to the reason for Borden's lower prices. The examiner was of the opinion that the primary reason for Borden's lower prices was that by using an f.o.b. plant method, Borden was able to take advantage of its obviously more favorable plant locations. Simply put, Borden's plants in the South which recently had begun to package private label milk were located closer to the buyers in the South and East than were the competitors' plants, none of which were located in the southern or eastern areas of the United States.9 The examiner reasoned that the buyers could obtain private label milk cheaper from Borden, by purchasing it f.o.b. and paying the transportation costs themselves, then they could purchase it at delivered prices from the competitors' plants located farther away. The examiner concluded that 86% Of the sales were diverted to Borden because of its advantageous locations and consequent freight advantage.

The Commission disagreed that freight advantage was the foremost reason for the diversion of business, declaring that the lower prices of Borden's private label milk, as compared to the Borden brand milk, was another important element; it also stressed Borden's size in relation to the smaller competitors. However, freight advantage clearly was one of the relevant factors, if not the most significant factor, causing the shift of business. Indeed, there is evidence that Borden itself had lost private label sales in the Northeast to other competitors (Pet and Carnation) whose more favorable plant locations in that area had given them a similar advantage in freight over Borden.10

Although the testifying competitors did lose a significant number of sales to Borden, it is also true that they gained from other sources enough new sales (525,000 cases) to achieve an overall increase in absolute sales. Borden argues that this fact demonstrates that the competitors did not, and will not, suffer the requisite competitive injury. The Commission rejected this contention on the ground that part of the new sales had been obtained from buyers who formerly were customers of 10 milk manufacturers which had gone out of business since 1950. Thus the Commission deemed the new sales to be mere windfalls which could not be expected to recur on a regular basis. The Commission implies that Borden's practices contributed to the decline of these companies, but the record contains no evidence as to the reason for their ceased production. Although the Commission noted that the evaporated milk industry had suffered a decline in sales, it apparently did not consider the possibility that the closing of these producers may have been due to a declining demand for evaporated milk.11

In any event, it is plain that Borden's practices did not lead to a substantial increase in Borden's market position at the expense of these competitors. Borden's share of the market increased only from 9.9% In 1955 to 10.7% In 1957. During the same period, the total market share of the competitors increased in about the same proportion, from 6.3% To 6.8%.

We conclude for two reasons that the record does not contain substantial evidence to support a finding that there may be a substantial injury to competition at the seller's level. The first is that we think it significant that the testifying competitors have experienced an increase in absolute sales volume12 and have bettered their market position in approximately the same proportion as has Borden. See Minneapolis-Honeywell Regulator Co. v. FTC, 191 F.2d 786, 790 (7th Cir. 1951), cert. dism'd, 344 U.S. 206, 73 S. Ct. 245, 97 L. Ed. 245 (1952).

The second reason is the absence of the necessary causal relationship between the difference in prices and the alleged competitive injury.13 The Commission's position is that the competing sellers may be hurt because Borden sells its private label milk cheaper than its Borden brand milk. But none of the evidence adduced by the testifying competitors relates to the price difference between the milks marketed by Borden; instead it relates to the price difference between their own private label milk and Borden's private label milk. The competitors actually assert only that Borden was able to sell private label milk for a lower price than they could, and regarding that assertion, the price of Borden brand milk is immaterial in this case. In short, the evidence simply does not support the precise price discrimination alleged in the complaint. Therefore, injury proved in the primary line, if any, is not the effect of the price difference in issue. Perhaps the Commission's difficulty in matching its evidence of injury with the alleged price discrimination impliedly illustrates our conclusion below that competitive injury is not caused by a price differential which measures only the economic value of a premium label as compared to a private label.

Evidence on the issue of injury in the secondary line was supplied by whole salers and retailers from North Carolina and South Carolina. They had purchased from Borden the premium brand milk but not the private brand milk. Their testimony as to competitive injury was meager. In substance, it was that they would have been interested in buying the private brand milk in addition to the premium brand milk. The record shows, however, that wholesalers who did try to sell private label milk in the Carolinas met with minimal success. Many retailers never inquired about private label milk, and others who received specific offers from wholesalers either declined the offer or often waited as long as a year before accepting. And it is to be remembered that no customer who requested private brand milk from Borden was denied the right to purchase it at the same price being charged to other customers.14

It is easily understook why the private label milk is sold at all levels of distribution for substantially less than Borden brand milk. By increased advertising and promotional efforts over the years, Borden has created a decided consumer preference for milk bearing a Borden label. The Label has come to represent a value in itself.15 Thus it was not surprising that the testifying wholesalers and retailers admitted that the private label milk was interesting to them only at a price $1.50 to $2.00 less per case than the Borden brand milk.16 This position reflects their knowledge that they would have had to sell the private label milk for a correspondingly lower price.17

There has been no doubt that the economic factors associated with a premium brand would receive recognition under Sec. 2(a), the only question being the appropriate provision. Upon reversing this case, the Supreme Court made it clear that these factors should be taken into consideration under the injury and cost provisions.

The Commission apparently ignored completely the bearing which the value attached to the Boarden label might have upon competitive injury. Relying upon FTC v. Morton Salt Co., supra, the Commission declared that in order to justify a finding of the prescribed harmful competitive effect in the secondary line, it was sufficient that some merchants paid more for Borden brand than their competitors paid for like goods packaged under a private label. Aside from the total failure to consider, or even mention, the commercial significance of the consumer appeal of the Boarden label, Morton Salt used a standard volume discount to create a price differential, whose effects are so different from those involved here as to make that decision inapposite to the one before us.

We are of the firm view that where a price differential between a premium and nonpremium brand reflects no more than a consumer preference for the premium brand, the price difference creates no competitive advantage to the recipient of the cheaper private brand product on which injury could be predicated. 'Rather it represents merely a rough equivalent of the benefit by way of the seller's national advertising and promotion which the purchaser of the more expensive branded product enjoys.'18 The record discloses no evidence tending to show that Borden's price differential exceeds the recognized consumer appeal of the Borden label. Nor has it been suggested that the prices are unreasonably high for Borden brand milk on the one hand, or unrealistically low for the private label milk on the other.

We conclude that there is not substantial evidence to support a finding that Borden has violated Sec. 2(a). The price difference does not create a competitive advantage by which competition could be injured, and, furthermore, no customer has been favored over another.

Our holding on the injury issue again will make it unnecessary for us to reach other issues concerning cost justification and the scope of the order, as Borden has noted in its brief. In view of the fact that this is the second occasion we have not reached these issues, normally we would be concerned by the possibility that the ultimate disposition of the case might be even further delayed in the event our decision on the injury issue is reversed, necessitating our further consideration. However, we resist the temptation to dispose of the remaining issues at this time because we feel that little delay would result in such an event since the issues which we do not decide have been briefed so extensively by both parties that they could be determined without delay on the basis of the present briefs.

Petition to set aside the cease-and-desist order is

Granted.