with whom Mr. Justice Harlan and Mr. Justice Stewart join, dissenting.
I agree with the Court’s holding that there existed an agreement among the defendants to exchange price information whenever requested. However, I cannot agree that that agreement should be condemned, either as illegal per se, or as having had the purpose or effect of restricting price competition in the corrugated container industry in the Southeastern United States.
Under the antitrust laws, numerous practices have been held to be illegal per se without regard to their precise purpose or harm. As this Court said in Northern Pacific R. Co. v. United States, 356 U. S. 1, 5 (1958), “there are certain agreements or practices which because of their pernicious effect on competition and lack of any redeeming virtue are conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm they have caused or the business excuse for their use.” Among these practices are price-*341fixing, United States v. Socony-Vacuum Oil Co., 310 U. S. 150, 223 (1940); division of markets, United States v. Addyston Pipe & Steel Co., 85 F. 271 (C. A. 6th Cir. 1898), aff’d, 175 U. S. 211 (1899); group boycotts, Fashion Originators’ Guild v. FTC, 312 U. S. 457 (1941); and tying arrangements, International Salt Co. v. United States, 332 U. S. 392 (1947). We have recently added to this list certain sales-commission systems for the marketing of tires, batteries, and accessories by service stations affiliated with major oil companies. FTC v. Texaco Inc., ante, p. 223 (1968). This Court has refused to apply a per se rule to exchanges of price and market information in the past. See American Column & Lumber Co. v. United States, 257 U. S. 377 (1921); United States v. American Linseed Oil Co., 262 U. S. 371 (1923); Maple Flooring Mfrs. Assn. v. United States, 268 U. S. 563 (1925); Cement Mfrs. Protective Assn. v. United States, 268 U. S. 588 (1925). I believe we should follow the same course in the present case.
Per se rules always contain a degree of arbitrariness. They are justified on the assumption that the gains from imposition of the rule will far outweigh the losses and that significant administrative advantages will result. In other words, the potential competitive harm plus the administrative costs of determining in what particular situations the practice may be harmful must far outweigh the benefits that may result. If the potential benefits in the aggregate are outweighed to this degree, then they are simply not worth identifying in individual cases.
I do not believe that the agreement in the present case is so devoid of potential benefit or so inherently harmful that we are justified in condemning it without proof that it was entered into for the purpose of restraining price competition or that it actually had that effect. The agreement in this case was to supply, when re*342quested, price data for identified customers. Each defendant supplied the necessary information on the expectation that the favor would be returned. The nature of the exchanged information varied from case to case. In most cases, the price obtained was the price of the last sale to the particular customer; in some cases, the price was a current quotation to the customer. In all cases, the information obtained was sufficient to inform the defendants of the price they would have to beat in order to obtain a particular sale.
Complete market knowledge is certainly not an evil in perfectly competitive markets. This is not, however, such a market, and there is admittedly some danger 'that price information will be used for anticompetitive purposes, particularly the maintenance of prices at a high level. If the danger that price information will be so used is particularly high in a given situation, then perhaps exchange of information should be condemned.
I do not think the danger is sufficiently high in the present case. Defendants are only 18 of the 51 producers of corrugated containers in the Southeastern United States. Together, they do make up 90% of the market and the six largest defendants do control 60% of the market. But entry is easy; an investment of $50,000 to $75,000 is ordinarily all that is necessary. In fact, the number of sellers has increased from 30 to the present 51 in the eight-year period covered by the complaint. The size of the market has almost doubled because of increased demand for corrugated containers. Nevertheless, some excess capacity is present. The products produced by defendants are undifferentiated. Industry demand is inelastic, so that price changes will not, up to a certain point, affect the total amount purchased. The only effect of price changes will be to reallocate market shares among sellers.
*343In a competitive situation, each seller will cut his price in order to increase his share of the market, and prices will ultimately stabilize at a competitive level— i. e., price will equal cost, including a reasonable return on capital. Obviously, it would be to a seller’s benefit to avoid such price competition and maintain prices at a higher level, with a corresponding increase in profit. In a market with very few sellers, and detailed knowledge of each other’s price, such action is possible. However, I do not think it can be concluded that this particular market is sufficiently oligopolistic, especially in light of the ease of entry, to justify the inference that price information will necessarily be used to stabilize prices. Nor do I think that the danger of such a result is sufficiently high to justify imposing a per se rule without actual proof.
In this market, we have a few sellers presently controlling a substantial share of the market. We have a large number competing for the remainder of the market, also quite substantial. And total demand is increasing. In such a case, I think it just as logical to assume that the sellers, especially the smaller and newer ones,1 will desire to capture a larger market share by cutting prices as it is that they will acquiesce in oligopolistic behavior. The likelihood that prices will be cut and that those lower prices will have to be met acts as a deterrent to setting prices at an artificially high level in the first place. Given the uncertainty about the probable effect of an exchange of price information in this context, I would require that the Government prove that the exchange was entered into for the purpose of, or that it had the effect of, restraining price competition.
*3441 do not find the inference that the exchange of price information has had an anticompetitive effect as “irresistible” as does the Court. Like my Brother Fortas, I would prefer that a finding of anticompetitive effect be supported by “evidence in the record.” I cannot agree that the evidence in this case was sufficient to prove such an effect. The Government has simply not proved its case.
The Court does not hold that the agreement in the present case was a deliberate attempt to stabilize prices. The evidence in the case, largely the result of stipulation, would not support such a holding. The Government points to a few isolated statements found in the depositions of industry witnesses, but I find these few fragmentary references totally insufficient. The weight of the evidence in the present case indicates that the price information was employed by each defendant on an individual basis, and was used by that defendant to set its prices for a specific customer; ultimately each seller wanted to obtain all or part of that customer’s business at the expense of a competitor. The District Court found that there was no explicit agreement among defendants to stabilize prices and I do not believe that the desire of a few industry witnesses to use the information to minimize price cuts supports the conclusion that such an agreement was implicit. On the contrary, the evidence establishes that the information was used by defendants as each pleased and was actually employed for the purpose of engaging in active price competition.
Nor do I believe that the Government has proved that the exchange of price information has in this case had the necessary effect of restraining price competition.2 In *345its brief before this Court, the Government relies very largely on one finding of the District Court and upon economic theory. The Government has presented a convincing argument in theoretical terms. However, the evidence simply does not square with that theory. And, this is not a case in which it would be unduly difficult to demonstrate anticompetitive effects.
The record indicates that defendants have offered voluminous evidence concerning price trends and competitive behavior in the corrugated container market. Their exhibits indicate a downward trend in prices, with substantial price variations among defendants and among their different plants. There was also a great deal of shifting of accounts. The District Court specifically found that the corrugated container market was highly competitive and that each defendant engaged in active price competition. The Government would have us ignore this evidence and these findings, and assume that because we are dealing with an industry with overcapacity and yet continued entry, the new entrants must have been attracted by high profits. The Government then argues that high profits can only result from stabilization of prices at an unduly high level. Yet, the Government did not introduce any evidence about the level of profits in this industry, and no evidence about price levels. Not one customer was called, although the Government surely had ample access to defendants’ customers. The Government admits that the price trend was down, but asks the Court to assume that the trend would have been accelerated with less informed, and hence more vigorous, price competition.3 In the absence of any proof what*346soever, I cannot make such an assumption. It is just as likely that price competition was furthered by the exchange as it is that it was depressed.
Finally, the Government focuses on the finding of the District Court that in a majority of instances a defendant, when it received what it considered reliable price information, would quote or charge substantially the same price.4 The Court and my Brother Fortas also focus on this finding. Such an approach ignores, however, the remainder of the District Court’s findings. The trial judge found that price decisions were individual decisions, and that defendants frequently did cut prices in order to obtain a particular order.5 And, the absence of any price parallelism or price uniformity and the downward trend in the industry undercut the conclusion that price information was used to stabilize prices.6
The Government is ultimately forced to fall back on the theoretical argument that prices would have been more unstable and would have fallen faster without price information. As I said earlier, I cannot make this assumption on the basis of the evidence in this record. The findings of the Court below simply do not indicate that the exchange of information had a significant anti-*347competitive effect; if we rely on these findings, at worst all we can assume is that the exchange was a neutral factor in the market.7 As this Court said in Maple Flooring, supra, at 585: “We realize that such information, gathered and disseminated among the members of a trade or business, may be the basis of agreement or concerted action to . . . raise prices beyond the levels . . . which would prevail if no such agreement or concerted action ensued and those engaged in commerce were left free to base individual initiative on full information of the essential elements of their business.” However, here, as in Maple Flooring, the Government has not proved that the information was so used. Rather, the record indicates that, while each defendant occasionally received price information from a competitor, that information was used in the same manner as other reliable market information — i. e., to reach an individual price decision based upon all available information. The District Court’s findings that this was a competitive industry, lacking any price parallelism or uniformity, effectively refute the Government’s assertion that the result of those decisions was to maintain or tend to maintain prices at other than a competitive level. Accordingly, I would affirm the decision of the court below.
The record does not indicate whether all manufacturers engaged in exchange of price information, or whether the practice was limited to defendants. There is no indication that other manufacturers would not have been given price information had they requested it.
Here it is relevant to note again that the evidence was largely the result of stipulation, with the Government admittedly introducing very little evidence on the actual effect of the allegedly illegal practice.
There was no effort to demonstrate that the price behavior of those manufacturers who did not exchange price information, if any, varied significantly from the price behavior of those who did. In fact, several of the District Court’s findings indicate that when cer*346tain defendants stopped exchanging price information, their price behavior remained essentially the same, and, in some cases, prices actually increased.
It should be noted that, in most cases, this information was obtained from a .customer rather than a competitor, a practice the Government does not condemn.
Immediately following the particular sentence emphasized by the Government, there appears the finding that “[i]n many instances, however, depending upon particular circumstances, each defendant quoted lower or higher prices, and in all instances the determination as to the price to be charged or quoted was its individual decision.” Other findings of fact are to the same effect.
As mentioned above, no evidence was introduced that would indicate that more than minimal price cuts were economically feasible.
See n. 3, supra.