The courts have persistently interpreted section 1 of the Sherman law differently from the way they have interpreted section 2 of the Sherman Act. The courts have interpreted section 1 of the Sherman Act in its application to loose combinations differently from the way in which they have interpreted it in its application to so-called tight combinations.

In a number of cases, including the Trenton Potters Association case-a decision handed down in about 1927, as I recall-and again in the Socony-Vacuum case in the 1930's, and indeed, throughout the whole history of the interpretation of the Sherman Act, the courts have held that an agreement among competitors, independent competitors, engaged in interstate commerce, with respect to prices and market sharing and the like, however reasonable the arrangement may be, is contrary to, violates the antitrust statutes.

On the other hand, the Supreme Court has held, in the United States Steel Corp. case, 1920, that although the steel corporation was large, it didn't violate either section 1 or section 2 of the act. The Court called attention to the fact that the corporation had not abused such power as it had, and concluded that the public interest would not be served by dissolving the corporation. But if the independent steel companies that went into the corporation had remained independent, and had entered into a price-fixing agreement, or a market-sharing agreement among themselves, the Court unquestionably would have held that such an arrangement violated the law.

Now, the tight sort of combination which you get when the United States Steel Corp. absorbs directly 12 and indirectly 132 firms, obviously is much more effective in insuring common action with respect to pricing than a loose arrangement among those concerns would have been.

It seems to me that the antitrust laws ought to be modified in such way as to serve as a better guide to the courts with respect to the sort of combinations Congress is approving, and the sort of combinations it isn't approving.

It seems to me that the Congress ought to lay down a policy which makes it perfectly clear that the objectives of section 1 and section 2 of the Antitrust Act are to encourage effective competition.

Mr. DENTON. I think the evidence we heard was that the steel corporation was formed to control about 40 percent of the market, and now it controls only about 30 percent. With these other companies in the steel business do you think prices can be controlled by United States Steel alone, or would there have to be an agreement of some kind, express or implied, between it and the other steel companies?

Mr. STOCKING. I have already discussed that. I think whenever you get few sellers in the market, they can act with respect to prices in such a way as to give them a noncompetitive stability without overt agreement.

I also think that it is much easier for them to enter into an arrangement which may achieve their objective without their violating the law. For those reasons I am skeptical of markets with only a few sellers in them.

The CHAIRMAN. You mean we ought to address ourselves to the power?

Mr. STOCKING. Yes.

May I read something which I have written? I have no pride of authorship in this, but it expresses the idea which I have in mind.

I think, for example, that a preamble to the Sherman Antitrust Act of this sort might be useful [reading]:

It is the policy of the Congress of the United States to encourage an effectively competitive economy. This policy is designed to insure as many sellers in interstate markets as is consistent with the economies of mass production. It is not intended to prevent growth through efficiency, but to prevent the accumulation of market power particularly through the merger of business rivals, the preemption of the supply of limited natural resources, the abuse of patent privileges, and to prevent any and all restrictive agreements among business rivals. It is aimed at power over the market, not economic efficiency.

I have one other suggestion. I have seen a proposal to modify the Clayton Act which reads as follows and it seems to me that this proposal has considerable merit [reading]:

Any corporation whose size and power are such as to substantially lessen competition or tend to create a monopoly in any line of commerce in any section of the country shall be dissolved into a number of independent enterprises sufficient to restore competition in such line of commerce, provided that no action under this section shall be taken if the corporation proceeded against can demonstrate that the proposed action would materially lessen efficiency in any line of commerce.

If such a policy is adopted I believe that it would clarify existing policy, and I think then it ought to be applied on a case-by-case basis by competent administrative agencies and by the courts.

The CHAIRMAN. You would think both suggestions should be considered?

Mr. STOCKING. I think so.

The CHAIRMAN. That is a preamble to the Sherman Act, and an amendment to the Clayton Act?

Mr. STOCKING. Yes, sir.

That concludes my statement, except to add that, obviously, some of the statements I have made represent an informed opinion, as I previously have said.

I don't believe that all economists would agree with everything I have said.

I think it is the obligation of a public body to get different points of view with respect to these matters, because I believe in competition of ideas just as I believe in competition in the market place, and I think that wisdom is apt to be the distilled product from such a procedure.

Mr. WILSON. Doctor, you don't believe that size necessarily means monopoly, then?

Mr. STOCKING. No. Certainly small size wouldn't mean monopoly. Mr. WILSON. Do you believe that any certain sized business has a monopoly on efficiency by reason of its size either small- or middlesized or large?

Mr. STOCKING. The testimony with regard to the steel corporation would indicate a decided "No" to that question.

The CHAIRMAN. Doctor, you have made a very full and most effective contribution, which was the result of painstaking effort, I am sure. We are very grateful to you.

The committee will now adjourn until 2 o'clock, when we will hear Dr. George Stigler, professor of economics, Columbia University,

and unless the members of the committee wish to hear other witnesses, that will conclude our hearings with reference to the steel industry. (Whereupon, at 12:30 p. m., a recess was taken, to reconvene at 2 p. m. of the same day.)

AFTERNOON SESSION

The CHAIRMAN. The committee will come to order.

Unless there is a desire on the part of the members of the committee to hear additional witnesses, the testimony we will take this afternoon from Mr. George J. Stigler, professor of economics, Columbia University, will be the last in this inquiry concerning the steel industry.

At this time the Chair wishes to express its gratitude to its colleagues on the committee for their intense interest, their patience and their regular attendance at the hearings, at great sacrifice to themselves.

I hope their work here will yield the Congress most worthy results. The sessions have been long and difficult-at times, burdensomebut it is refreshing to note and to receive the members' reactions which have been, always, most cooperative. I want to be sure that the record clearly indicates that I am very grateful to the members of this committee.

After the session this afternoon, the committee will recess for about a month during which time the staff will prepare for the inquiry into newsprint, and it is hoped that we may reassemble after that preparation a bit after the middle of June.

Dr. Stigler, we will be very glad to hear from you.

STATEMENT OF GEORGE J. STIGLER, PROFESSOR OF ECONOMICS, COLUMBIA UNIVERSITY

Mr. STIGLER. I wish to begin by stating my understanding of the task of this committee, and the role of the steel industry in the present hearings. The committee seeks to discover whether our antitrust policy is sufficient to deal with the present problem of monopoly in the United States. The steel industry is an important example of the class of industries that are not old-fashioned monopolies, and yet which need seldom engage in formal conspiracies of the cartel type in order to display important monopolistic tendencies. I assume that the steel industry has not been chosen for study by this committee to demonstrate the existence of monopolistic practices with which the existing law can deal-indeed, if the monopolistic practices can be dealt with effectively under existing law, there is no need to change the law. Rather, steel is simply a case study in an important. form of the monopoly problem.

It would be pointless, therefore, for me to discuss the steel industry only in terms of the evidence presented at these hearings, informative as that evidence is. The steel industry has been the object of disapproving study by disinterested economists for many decades, and it would be peculiar to exclude this information from the committee's deliberations. Moreover, it seems to me proper and even necessary to introduce views drawn from observation of other industries, in order to reach generally valid conclusions.

With these understandings, I wish to comment particularly on the deficiency in our present antitrust policy, the issue of bigness, and the nature of the policy which we should adopt.

1. Oligopoly and the antitrust laws: The traditional antitrust policy was designed to deal with two important problems: Conspiracies among the firms in an industry; and "trusts" or monopolies that control most or all of an industry. When the policy has been pursued vigorously it has been successful in dealing with both problems, and even in periods and areas where it has not been applied vigorously, it has had beneficial effects.

Partly because of the very effectiveness of the Sherman law in dealing with these striking forms of monopolization, a form of organization known to economists by the ungainly name of oligopoly has developed in certain important industries. Oligopoly means "few sellers," and it is used to describe both industries with few sellers (three in aluminum, for example) and industries with many sellers if a few firms are of dominant size. The steel industry is oligopolistic in this latter sense.

Oligopolies pose the crucial unsolved problem of antitrust policy. Because the individual companies have control over only 10, 20, 30, or 40 percent of the industry, they are apparently not subject to dissolution under the Sherman Act. On the other hand, because they are few, the companies cannot fail to take account of their effects on one another, and gradually establish rapport on mutually profitable monopolistic policies. Moreover, they are so few that they need not engage in frequent and visible collusion in any formal sense. One firm can assume the task of announcing the price changes, the others following obediently without direct communication. It would be wrong to say that the Sherman Act condones this form of tacit collusion, for in the Tobacco and Cement cases the court was willing to infer collusion from concerted behavior. But it would also be wrong to say that the Sherman Act, as now construed, is capable of dealing with the basic problem; a conviction often leads to a court order to the oligopolists to forget that they are oligopolists.

Suppose an industry consists of three firms. In the course of time they find that cooperation (monopoly) is more profitable than competition, and they establish the principle of buying equal shares of the important raw materials or all selling at a price one of the companies announces. In due, or perhaps after an undue, time they are convicted for violating the antitrust laws, because their equal shares of raw material or uniform price maneuvers reveal the tacit conspiracy. (So far I have merely described in general terms the kind of behavior that was found in the Tobacco case.) Now let us consider the probable effect of the conviction.

No one of the three firms can possibly disregard its rivals; for example, no firm can assume that if it reduces its price, the rival companies will not reduce their prices sufficiently rapidly to eliminate important gains from price competition. It will also be impossible for the companies to forget that monopolization is more profitable than competition. So perhaps they meet once and divide up the country, each company thereafter selling in 16 States. In due or undue time they will again be convicted. They may then hire an economist, and if he is not a particularly good economist he may

advise each company to sell say 4 percent of its product in rivals' States. They will then be convicted again. Eventually they may find a businessman, lawyer, or economist who is smarter than the staff of the Antitrust Division, devise an acceptable plan of cooperation, and be left undisturbed. Whether and until they reach this happy state, lawyers' fees are less expensive than competition.

The important point this example illustrates is this: In oligopolistic industries, competitive behavior can be achieved only by a careful and continuous supervision of every business practice and every price and investment decision that the industry makes. (Even this requires perfect supervision, and the kinds of supervision actually displayed by regulatory bodies suggest that competitive results would not be achieved-and would not be sought.) The orthodox antitrust procedures can eliminate some of the worst monopolistic practices, but they cannot compel an industry to display satisfactorily competitive behavior.

Let us illustrate this problem of oligopoly in the steel industry. The representatives of this industry are quite right, by the way, in stating that it is by no means the most concentrated industry in the country. On the other hand, it is a good choice: If economists were asked to name the most important industry in America in which there are important monopolistic elements, steel would be very high in the composite list. Moreover, the conventional figures on concentration underestimate the relative amount of monopoly in steel as compared with other industries. There are few good substitutes for steel in most of its uses, and it is an industry which cannot be entered quickly by new concerns. I find it somewhat misleading to be told that steel ingots were less concentrated in 1937 (when the four largest firms produced 53 percent of ingots by weight) than the industry making two-piece men's suits with extra knickers (78 percent being produced by the four largest firms).

(1) Consider the international steel cartels. To meet the letter of the unfortunate Webb-Pomerene Act, the leading American firms explicitly reserved the right of foreign competitors to sell in the United States. What does such an agreement mean? Even a verbatim transcript does not record winks and grins. It is utterly improbable that an amicable division of foreign markets and the maintenance of cartel prices would endure if foreign producers actively invaded our market, and this would be self-evident to every member of the cartel. If the committee were to examine the records of imports from foreign members of cartels before and during the cartels, they will be able to verify this general awareness on the part of oligopolists of the economic realities of collusion.

How would a Sherman law conviction change this? The American firms could arrange for token imports, or the foreign firms could offer noncompetitive bids, or extra quotas could be given to American producers abroad in exchange for import quotas to foreign producers, or a dozen other devices could be contrived-all without sacrificing the substance of removal of foreign competition from the American market.

(2) Consider the competition of integrated companies with nonintegrated fabricators. This is a complicated set of problems, and I shall discuss only one simple aspect. In recent years the basic steel