The question presented involves an important construction and application of § 7 of the Clayton Act,1 38 Stat. 731, as amended, 15 U.S.C. § 18. Consolidated Foods Corp.—which owns food processing plants and a network of wholesale and retail food stores—acquired Gentry, Inc., in 1951. Gentry manufactures principally dehydrated onion and garlic. The Federal Trade Commission held that the acquisition violated § 7 because it gave respondent the advantage of a mixed threat and lure of reciprocal buying in its competition for business and 'the power to foreclose competition from a substantial share of the markets for dehydrated onion and garlic.' It concluded, in other words, that the effect of the acquisition 'may be substantially to lessen competition' within the meaning of § 7, and it ordered divestiture and gave other relief. —- F.T.C. —-, —-. The Court of Appeals, relying mainly on 10 years of post-acquisition experience, held that the Commission had failed to show a probability that the acquisition would substantially lessen competition. 329 F.2d 623. The case is here on certiorari. 379 U.S. 912, 85 S.Ct. 259, 13 L.Ed.2d 183.

We hold at the outset that the 'reciprocity' made possible by such an acquisition is one of the congeries of anticompetitive practices at which the antitrust laws are aimed. The practice results in 'an irrelevant and alien factor,' —- F.T.C., p. —-, intruding into the choice among competing products, creating at the least 'a priority on the business at equal prices.' International Salt Co. v. United States, 332 U.S. 392, 396—397, 68 S.Ct. 12, 15, 92 L.Ed. 20; Northern Pac. R. Co. v. United States, 356 U.S. 1, 3, 6, 12, 78 S.Ct. 514, 517, 2 L.Ed.2d 545. Reciprocal trading may ensue not from bludgeoning or coercion but from more subtle arrangements. A threatened withdrawal of orders if products of an affiliate cease being bought, as well as a conditioning of future purchases on the receipt of orders for products of that affiliate is an anti-competitive practice.2 Section 7 of the Clayton Act is concerned 'with probabilities, not certainties.' Brown Shoe Co. v. United States, 370 U.S. 294, 323, 82 S.Ct. 1502, 8 L.Ed.2d 510; United States v. Philadelphia Nat. Bank, 374 U.S. 321, 362, 83 S.Ct. 1715, 1740, 10 L.Ed.2d 915. Reciprocity in trading as a result of an acquisition violates § 7, if the probability of a lessening of competition is shown. We turn then to that, the principal, aspect of the present case.

Consolidated is a substantial purchaser of the products of food processors who in turn purchase dehydrated onion and garlic for use in preparing and packaging their food. Gentry, which as noted is principally engaged in the manufacture of dehydrated onion and garlic, had in 1950, immediately prior to its acquisition by Consolidated, about 32% of the total sales of the dehydrated garlic and onion industry and, together with its principal competitor, Basic Vegetable Products, Inc., accounted for almost 90% of the total industry sales. The remaining 10% was divided between two other firms. By 1958 the total industry output of both products had doubled. Gentry's share rising to 35% and the combined share of Gentry and Basic remaining at about 90%.3

After the acquisition Consolidated (though later disclaiming adherence to any policy of reciprocity) did undertake to assist Gentry in selling. An official of Consolidated wrote as follows to its distributing divisions:

'Oftentimes, it is a great advantage to know when you are calling on a prospect, whether or not that prospect is a supplier of someone within your own organization. Everyone believes in reciprocity providing all things are equal.

'Attached is a list of prospects for our Gentry products. We would like to have you indicate on the list whether or not you are purchasing any of your supplies from them. If so, indicate whether your purchases are relatively large, small or insignificant. * * *

'Will you please refer the list to the proper party in your organization. * * * If you have any special suggestions, as to how you could be helpful in properly presenting Gentry to any of those listed, it will be appreciated.'

Food processors who sold to Consolidated stated they would give their onion and garlic business to Gentry for reciprocity reasons if it could meet the price and quality of its competitors' products. Typical is a letter from Armour and Co.:

'I can assure you that it is the desire of our people to reciprocate and cooperate with you in any way we can in line with good business practices, and I am sure that if our quality obstacles can be overcome, your quotations will receive favorable consideration. We value our relationship with you very highly and are disappointed that we have been unable lately to reciprocate for your fine cooperation on Armour Pantry Shelf Meats.' Some suppliers responded and gave reciprocal orders. Some who first gave generous orders later reduced them or abandoned the practice. It is impossible to recreate the precise anatomy of the market arrangements following the acquisition, though respondent offers a factual brief seeking to prove that 'reciprocity' either failed or was not a major factor in the post-acquisition history.

The Commission found, however, that 'merely as a result of its connection with Consolidated, and without any action on the latter's part, Gentry would have an unfair advantage over competitors enabling it to make sales that otherwise might not have been made.'

'With two firms accounting for better than 85% of both product lines for eleven successive years, maximum concentration short of monopoly has already been achieved. If it is desirable to prevent a trend toward oligopoly it is a fortiori desirable to remove, so far as possible, obstacles to the creation of genuinely competitive conditions in an oligopolistic industry. Respondent's reciprocal buying power, obtained through acquisition of Gentry, is just such an anticompetitive obstacle.

'This conclusion is buttressed by the peculiar nature of the dehydrated onion and garlic industry. In the first place, the record shows that Gentry's leading competitor, Basic Vegetable Products, Inc., has been the innovator and leader in the field. Gentry has recently made technical strides narrowing, although probably not closing, the gap between them. There is also evidence that the third firm, Puccinelli Packing Co., is not only much smaller—commanding only about 10% of each product market—but is considered by many buyers to offer an inferior product and inferior service.' —- F.T.C., p. —-.

The Court of Appeals, on the other hand, gave post-acquisition evidence almost conclusive weight. It pointed out that, while Gentry's share of the dehydrated onion market increased by some 7%, its share of the dehydrated garlic market decreased 12%. 329 F.2d, p. 626. It also relied on apparently unsuccessful attempts at reciprocal buying. Ibid. The Court of Appeals concluded that 'Probability can best be gauged by what the past has taught.' Id., p. 627.

The Court of Appeals was not in error in considering the post-acquisition evidence in this case. See United States v. E. I. Du Pont De Nemours & Co., 353 U.S. 586, 597, 77 S.Ct. 872, 879, 1 L.Ed.2d 1057, et seq., 602 et seq. But we think it gave too much weight to it. Cf. United States v. Continental Can Co., 378 U.S. 441, 463, 84 S.Ct. 1738, 1750, 12 L.Ed.2d 953. No group acquiring a company with reciprocal buying opportunities is entitled to a 'free trial' period. To give it such would be to distort the scheme of § 7. The 'mere possibility' of the prohibited restraint is not enough. (United States v. E. I. Du Pont De Nemours & Co., supra, 353 U.S. p. 598, 77 S.Ct. p. 880.) Probability of the proscribed evil is required, as we have noted. If the post-acquisition evidence were given conclusive weight or allowed to override all probabilities, then acquisitions would go forward willy-nilly, the parties biding their time until reciprocity was allowed fully to bloom. It is, of course, true that post-acquisition conduct may amount to a violation of § 7 even though there is no evidence to establish probability in limine. See United States v. E. I. Du Pont De Nemours & Co., supra, 353 U.S. pp. 597—598, 77 S.Ct. pp. 879—880. But the force of § 7 is still in probabilities, not in what later transpired. That must necessarily be the case, for once the two companies are united no one knows what the fate of the acquired company and its competitors would have been but for the merger.

Moreover, the post-acquisition evidence here tends to confirm, rather than cast doubt upon, the probable anti-competitive effect which the Commission found the merger would have. The Commission found that Basic's product was superior to Gentry's—as Gentry's president freely and repeatedly admitted. Yet Gentry, in a rapidly expanding market, was able to increase its share of onion sales by 7% and to hold its losses in garlic to a 12% decrease. Thus the Commission was surely on safe ground in reaching the following conclusion:

'If reciprocal buying creates for Gentry a protected market, which others cannot penetrate despite superiority of price, quality, or service, competition is lessened whether or not Gentry can expand its market share. It is for this reason that we reject respondent's argument that the decline in its share of the garlic market proves the ineffectiveness of reciprocity. We do not know that its share would not have fallen still farther, had it not been for the influence of reciprocal buying. This loss of sales fails to refute the likelihood that Consolidated's reciprocity power, which it has shown a willingness to exploit to the full, will not immunize a substantial segment of the garlic market from normal quality, price, and service competition.' —- F.T.C., p. —-.4

But the Court of Appeals ignored the Commission's findings as to the inferiority of Gentry's product; indeed at one point it even supplanted those findings with its own conclusion that Gentry's onions were superior:

'Consolidated's Gentry division in the years following the acquisition, during which time it improved its onion processing equipment to eliminate a problem arising from the presence of wood splinters and achieved a product of higher quality than that of its competitors, increased its share of the rapidly expanding market by only some 7% with respect to dehydrated onion * * *.' 329 F.2d, p. 626. (Emphasis supplied.) But the Commission's contrary conclusion was unquestionably based on substantial evidence, as the following excerpt from the testimony of Gentry's president particularly indicates:

'Q. You mentioned the fact, Dr. Prater, that Gentry had a reputation of being second to Basic in quality. Was one of the factors involved in the quality competition the wood splinter problem?

'A. Yes, the wood splinter problem has been a problem in the dehydration industry for many years. Basic exploited this extensively, and solved it by improvements in production techniques in the use, or by the use of better methods, and by using, instead of wood trays, trays of aluminum plastic glass fibers. We met this competition partially by the improvement of our production techniques and installation of continuous conveyor dehydrators.'

We do not go so far as to say that any acquisition, no matter how small, violates § 7 if there is a probability of reciprocal buying. Some situations may amount only to de minimis. But where, as here, the acquisition is of a company that commands a substantial share of a market, a finding of probability of reciprocal buying by the Commission, whose expertise the Congress trusts, should be honored, if there is substantial evidence to support it.

The evidence is in our view plainly substantial. Reciprocity was tried over and again and it sometimes worked. The industry structure was peculiar, Basic being the leader with Gentry closing the gap. Moreover there is evidence, as the Commission found, 'that many buyers have determined that their source of supply may best be protected by a policy of buying from two suppliers.' When reciprocal buying—or the inducement of it—is added, the Commission observed:

'Buyers are likely to lean toward Basic on the ground of quality, but, in seeking a second, protective supply channel, to purchase from Gentry in the belief that this will further their sales to Consolidated. Not only does Gentry thus obtain sales that might otherwise go to Basic or Puccinelli, but the two-firm oligopoly structure of the industry is strengthened and solidified and new entry by others is discouraged.' —- F.T.C., p. —-.

We conclude that there is substantial evidence to sustain that conclusion and that the order of the Commission should not have been denied enforcement. The judgment of the Court of Appeals is accordingly reversed.

Reversed.

Mr. Justice HARLAN (concurring in the judgment).

Had the Commission's complaint been grounded on § 5 of the Federal Trade Commission Act, it seems manifest to me that no case would have been made out on this record. But given the ambulatory use of § 7 of the Clayton Act sanctioned by the Court in United States v. E. I. Du Pont De Nemours & Co., 353 U.S. 586, 77 S.Ct. 872, 1 L.Ed.2d 1057, I concur in the judgment.

I do so, however, upon the premises stated in the concurring opinion of my Brother STEWART, post, this page, but with one reservation. To the extent that anything in his opinion might be taken as drawing on evidence upon which the Commission indicated no reliance, I could not subscribe to that approach. This Court must review administrative findings as they are made by the agency concerned, and if the evidence will not support the findings and theory upon which the agency acted, an affirmance of the agency's order cannot properly rest upon a reassessment of the record by us. See Securities & Exchange Comm'n v. Chenery Corp., 332 U.S. 194, 196, 67 S.Ct. 1575, 91 L.Ed. 1995; National Labor Relations Board v. Metropolitan Life Ins. Co., 380 U.S., pp. 438, 443—444, 85 S.Ct. pp. 1061, 1064—1065. However, since both sides agree that 'conglomerate' mergers and reciprocal buying are within the purview of § 7, I think the Commission's order is supportable, though barely so, within the confines of the evidence upon which it apparently relied.

Brown Shoe Co. v. United States, 370 U.S. 294, 82 S.Ct. 1502, 8 L.Ed.2d 510, forecloses any contention that the 'market affected' was not substantial enough to bring § 7 into play. In this Court Consolidated has pitched its case on the proposition that it used to the full whatever power it acquired as a result of the merger to bring about reciprocal buying. The Commission found only seven instances of successful efforts by Consolidated to pressure suppliers to buy from Gentry. If in fact these few instances had represented the full measure of Consolidated's ability to induce purchasing from Gentry, they would for me be insufficient to carry the day for the Commission's order, and I would vote to affirm. While I cannot subscribe to the undiscriminating use made in the Court's opinion of the buying statistics, I think there was enough in these seven instances—for example, the Phillips Packing Company, J. J. Gielow & Sons, Illinois Meat Company, and Morgan Packing Company episodes—for the Commission justifiably to find that Consolidated had not used all the reciprocal buying leverage it could muster; the Commission, therefore, could reasonably conclude that the probable effect of the Gentry acquisition would be substantially to lessen competition in the relevant market.

On this basis I concur in the result reached by the Court.

Mr. Justice STEWART (concurring in the judgment).

The Federal Trade Commission, in invalidating a merger between Consolidated Foods and Gentry, Inc., has espoused a novel theory to bring the facts of this case within the scope of § 7 of the Clayton Act. Its resolution of the issue has been much debated and much disputed.1 The Court of Appeals has disagreed with the Commission's appraisal of the facts in this case and with its conclusions concerning the § 7 implications of reciprocity. Other cases are being held awaiting clarification from this Court.2 We must decide the applicability of the Act to the facts of this case, but we should also provide guidance to the Commission and to the courts which will have to grapple in the future with the potentialities of reciprocal buying in § 7 cases. While I agree with the result that the Court has reached, I am persuaded to file this separate statement of my views regarding the issues involved.

Clearly the opportunity for reciprocity is not alone enough to invalidate a merger under § 7. The Clayton Act was not passed to outlaw diversification. Yet large scale diversity of industrial interests almost always presents the possibility of some reciprocal relationships. Often the purpose of diversification is to acquire companies whose present management can benefit from the technical skills and sales acumen of the acquiring corporation. Without more, § 7 of the Clayton Act does not prohibit mergers whose sole effect is to introduce into an arena of 'soft' competition the experience and skills of a more aggressive organization.

It obviously requires more than this kind of bare potential for reciprocal buying to bring a merger within the ban of § 7. Before a merger may be properly outlawed under § 7 on the basis solely of reciprocal buying potentials, the law requires a more closely textured economic analysis. The Court summarizes the 'substantial' evidence before the Commission as follows:

'Reciprocity was tried over and again and it sometimes worked. The industry structure was peculiar, Basic being the leader with Gentry closing the gap.

Moreover there is evidence, as the Commission found, 'that many buyers have determined that their source of supply may best be protected by a policy of buying from two suppliers.' When reciprocal buying—or the inducement of it—is added, the Commission observed: '* * * the two-firm oligopoly structure of the industry is strengthened and solidified and new entry by others is discouraged."

I cannot agree that these elements, singly or together, are sufficient to make unlawful the merger negotiated by Consolidated and Gentry. Certainly the mere effort at reciprocity cannot be the basis for finding the probability of a significant alteration in the market structure. Section 7 does not punish intent. No matter how bent on reciprocity Consolidated might have been, if its activities would not have the requisite probable impact on competition, it cannot be held to have violated this law. And, I think, it is not enough to say that the merger is illegal merely because the reciprocity attempts 'sometimes worked.' If the opportunity for reciprocity itself is not a violation of the Act when the merger occurs, then some standard must be established for determining how effective reciprocity must be before the merger is subject to invalidation. Nor do I think that illegality of this merger can be rested upon the fact that '(t)he industry structure was peculiar, Basic being the leader with Gentry closing the gap.' There is evidence that in the years following 1951, when the merger took place, increased emphasis was placed on solving technical problems which had prevented some processors from relying on dehydrated, rather than raw, onions. The 1950's were a time of flux for the industry. Basic was sometimes the innovator of technological change leading to increased sales; sometimes Gentry had the upper hand. It is possible that this shift to more intensive competition was connected with the merger. Faced with a new competitive situation, Basic may have determined to solve quality control problems which had long been dormant. Indeed, the evidence seems to show that, after the acquisition, the industry reflected the salutary qualities normally associated with free competition. Overall, both Basic and Gentry were furnishing a better product at the end of this period than at the beginning. It is true that the industry had oligopolistic features, but there is no evidence to indicate that barriers to entry were particularly severe.3 And Gentry, while it was 'closing the gap' with regard to dehydrated onions, was falling even farther behind Basic in the sales of dehydrated garlic. Finally, I can attach no significance to the fact that processors, seeking a second source of supply, normally relied on Gentry rather than Puccinelli. That fact can rest on so many alternative hypotheses that it is persuasive as to none.4

The touchstone of § 7 is the probability that competition will be lessened. But before a court takes the drastic step of ordering divestiture, the evidence must be clear that such a probability exists. The Act does not require that there be a certainty of anticompetitive effect. But that does not mean that the courts or the Commission can rely on slipshod information confusingly presented and ambiguous in its implications. The law does not require proof that competition certainly will be lessened by the merger. But the record should be clear and convincing that the requisite probability is present.

To determine that probability, the courts and the Commission should rely on the best information available, whether it is an examination of the market structure before the merger has taken place, or facts concerning the changes in the market after the merger has been consummated. For that reason, I differ with the Court in its assessment of the weight to be accorded post-acquisition evidence. That evidence is the best evidence available to determine whether the merger will distort market forces in the dehydrated onion and garlic industry. The Court of Appeals, in my view, was not wrong because it 'gave too much weight' to the post-acquisition evidence. It erred because of the gloss it placed on the statistics and testimony adduced before the hearing examiner and the Commission.

The Court discounts the value of post-acquisition evidence on the ground that the companies are not entitled to a 'free trial' period after the merger. That characterization, however, misstates the case. No one gives the company a 'free trial' by assessing, in light of what actually happened, what could only be hypotheses at the time the merger occurred. Without post-acquisition evidence, the trier is faced with a blank slate and untested speculation. The merger in this case was achieved in 1951, yet the Commission did not issue a cease-and-desist order until six years later. We may be sure that the Commission relied on post-acquisition factors in issuing its order; there is no reason why we should rely on those factors less in assessing the propriety of the Commission's action. Indeed, if anyone had a 'free trial' period to check the anticompetitive potential of the merger, it was not the respondent but the Commission.

The record in this case is sorely incomplete, and a reviewing court is given little guidance in determining why this merger should be voided, if reciprocity-creating mergers are not per se invalid. Yet our responsibility to the Commission—to respect its findings where there is evidence to support them—requires close scrutiny of the record before its conclusions are upset. I think the record contains just enough to support invalidation of the merger, but because of evidence not referred to in the Court's opinion.

The food processing industry is composed basically of two classes of manufacturers. One class, which includes such processors as Armour and Swift, has built significant brand names commanding consumer acceptance of their products. For such companies, exposure at the retail market is assured. Consolidated Foods, as the wholesaler, is sufficiently dependent on such processors that its economic power over this class is minimal. It cannot readily strong-arm Armour into purchasing dehydrated onions from Gentry at the pain of losing Consolidated's favor. A second class incorporates the smaller processors in the industry. Many of these sell their product to Consolidated in bulk, for packaging under house labels of Consolidated divisions. Many of the products which these processors package under their own labels are not so widely known; they rely on the wholesaler to persuade supermarkets to try them on their counters. These processors are susceptible to the subtle pressures of reciprocity.

My reading of the record persuades me that most of the processors in this second class shifted their buying from Basic to Gentry, though the extent of that shift varied from company to company. It is true that testimony from the purchasing agents of many of these companies attributed the shift to other causes. However, the pattern of movement in this class, when contrasted to the lack of a pattern among the major processors, seems to me sufficient to support the Commission's conclusion that these shifts were in response to the influence of reciprocity, whether express or 'tacitly accommodative.' The pattern is relevant because the independent processors are substantial purchasers in the dehydrated onion and garlic market. Furthermore, this pattern confirms what was assumed by the Commission: that Consolidated has the power to influence the purchases by a substantial segment of its suppliers. Some of the independent processors have failed, and others have merged with large processors leading to greater concentration in the food processing industry. The Commission could, therefore, have fairly concluded that the inhibitory effects of reciprocity in this situation marked this merger with illegality.

For these reasons I concur in the judgment of the Court.