ity, or whether it is achieved by removing an independent firm through acquisition makes a great deal of difference.

Mr. COULTER. Well, yes, surely. Now, would you not concede that there are cases where vertical integration through merger is desirable? I can conceive of a variety of cases where this would be the case.

Dr. MARTIN. I would hate to have to sit down and write out a list of all the circumstances surrounding cases which I would say are desirable and cases which are not desirable. I think that one question I would certainly ask is can the advantages to be achieved through the vertical merger or acquisition be achieved as well through some less drastic way of handling the problem. I would consider less drastic various contractual arrangements between independent firms that remain independent as well as the going into a line of business by an existing firm without removing another firm.

Mr. COULTER. Yes. Well, do you not think though there might be some cases where these less far reaching measures would not be adequate?

Dr. MARTIN. Yes; there might be.

Mr. COULTER. I have one other question. I am a little struck with your concluding thoughts and your discussion of what you consider to be the need for going into action aimed at the dissolution of some of the large firms which have been large for many years. You say on the last page:

By the very nature of the problem there are not really very many such firms that need to be dissolved.

Would you care to hazard a guess as to how many?

Dr. MARTIN. No, I do not think I would like to pick a number. I would simply say that there are a number of firms in existence today operating apparently legally that clearly would not have been created if the law at the time they were created had been interpreted the way it is now interpreted.

To a certain extent this gives us a double standard. It makes it impossible for small firms today to combine together to achieve the sort of power that other existing firms have, while we sit by and do nothing about the existing firms. I can see that there possibly is a question of fairness and equity here in the restrictions that are being placed on the firm.

Mr. COULTER. I am just curious whether you think there would be say 10 or 100 or 1,000 such firms.

Dr. MARTIN. I would say it would be closer to 100 than to 10 or 1,000. Mr. COULTER. That should be dissolved?

Dr. MARTIN. Yes. I do not think it should be done without careful study and evaluation.

Mr. CHUMBRIS. Mr. Martin, I was not going to get into this area of discussion because I was going to save it for another occasion_but it seems to me what you are recommending is similar to what has been recommended by several of our other witnesses who try to lay a foundation for what will result in some type of legislation that would cause a divestiture of many corporations that we have in the country today. Walter Adams brought it out in his testimony when he referred to a bill that he prepared for Congressman Celler back in 1949 which was

based on the theory of the Public Utilities Holding Act of 1935, and Professor Kaysen referred to it briefly in his testimony several weeks ago. But he did not come out with the same version of Professor Adams' bill. Are you really serious about a bill that would try to do what they suggest, and do you suggest that Congress consider such a bill?

Dr. MARTIN. No, quite the contrary. I think that the Sherman Act was enacted by Congress in 1890 for this very purpose, to prevent the continued existence of trusts and the development of new ones, and that the form of trusts was changed to evade the Sherman Act through the creation of State-chartered corporations with the same powers that the trusts had and that the Court only partially counteracted that with its interpretation of the law in 1920.

I think this was based on poor economic reasoning in 1920-the conclusion that adequate competition existed simply because United States Steel did only about half the business of an industry. I do not feel that legislation should be pushed in Congress to achieve this goal, not because I disagree with Adams, Edwards, and Kaysen on the goal. It is a very old goal in our society, and I am heartily in concurrence with it. But I think that the basis exists in the Sherman Act to achieve what is needed, and until the Court indicates otherwise, I would say the proper way to proceed would be in the courts.

Mr. CHUMBRIS. Then your point is that since we have operated under a system that has been created for over 50 years, Congress should not pass a law requiring that a corporation that assumes say 35 percent of the market or 25 percent of the market to divest itself of whatever subsidiaries or divisions that it may have, or requiring an industry to have 40 corporations having a market now held by 10 corporations.

You would not be supporting such a proposition as that, would you?

Dr. MARTIN. I would reserve judgment on the advisability of such legislation until I find that it is not possible to achieve what is needed through the existing legislation, and I think that the Court just a year ago in the Lexington Bank case gave an interpretation of section 1 of the Sherman Act as it relates to mergers for the first time since the Columbia Steel case in 1949. It distinguished the Columbia Steel case on the basis of some special facts and laid open, it seems to me, a very real possibility of using the Sherman Act to achieve what is needed in the way of reorganization of the highly concentrated areas of the economy.

Mr. CHUMBRIS. This matter came up again in 1958 when we were discussing the spin-off theory. In 1963 we had Dr. Adams come before us when we had hearings. First it was a compendium where each presented their papers and then we had I think 6 to 8 days of hearings in which these experts came and appeared personally. Now we have had it in this round. I just thought that perhaps we ought to bring this thing out in the open.

We have a suggestion here and there, and I thought probably we ought to bring it out in the open and find out exactly if these people are serious in introducing or suggesting legislation that will try to break up our way of doing business and say "All right, instead of

having the Big-4 and the Big-8 we are going to break this up. We are going to have 50 equally big sized corporations to do what the 8 are doing today," that type of thing. That is part of the testimony that has been brought out.

Dr. MARTIN. I have read some of this testimony, and in general I am in agreement with Corwin Edwards, Kaysen, and Adams on this question, except that I do not think when they were testifying they had fully taken into account the effect of the Lexington Bank case. Mr. COULTER. What that amounts to, then you believe we can and should do through legal action what Adams recommends doing through legislative action.

Dr. MARTIN. I think before resorting to legislative action, I would like to see a court test. Whenever you pass new legislation, you create a long period of uncertainty about the state of the law. You pass legislation to clarify the law, but the law is not immediately clarified by the legislation. The Celler-Kefauver Act was enacted in 1950 and it was not until 1962 that the Supreme Court gave its first interpretation of the substance of that law. I think it would be quicker to get the policy clearly stated, to take existing statutes and try to get a case to the Supreme Court that would test the issue.

Mr. CHUMBRIS. I have no further questions.

Senator HART. Thank you very much, Professor.

The second witness today is Prof. Norman R. Collins* of the Institute of Business and Economic Research of the University of California. Professor Collins also has contributed to the field that we are studying, and is the coauthor of an article which I think is already in our record, the "Size Structure of the Largest Industrial Firms." Professor, we welcome you. Your statement will be printed in full in the record as though given in full, and you are free to summarize or read it.

STATEMENT OF DR. NORMAN R. COLLINS, DEPARTMENT OF AGRICULTURAL ECONOMICS AND SCHOOL OF BUSINESS ADMINISTRATION, UNIVERSITY OF CALIFORNIA, BERKELEY, CALIF.

Dr. COLLINS. Thank you. May I just make a slight correction on my affiliation. I am a member of the faculty of the Department of Agricultural Economics and the School of Business Administration at the University of California.

Senator HART. The record will show that precise description. Dr. COLLINS. Thank you very much, Senator Hart.

I would first like to express my appreciation for the invitation to participate in these hearings on the subject of industrial concentration. The material that I will present today was prepared in cooperation with my colleague, Prof. Lee Preston who appeared before you last year discussing some other aspects of our work. He also worked

Ph. D., Harvard University (1956); professor of agricultural economics and business administration, University of California (Berkeley). Coauthor (with L. E. Preston). "The Size Structure of the Largest Industrial Firms, 1909-58" (American Economic Review, December 1961) and "The Structure of Food Processing Industries, 1935-55" (Journal of Industrial Economics, July 1961); "Changing Role of Price in Agricultural Marketing" (Journal of Farm Economics, August 1959); coauthor (with W. F. Mueller): "GrowerProcessor Integration in Fruit and Vegetable Marketing" (Journal of Farm Economics, December 1957).

with me in the preparation of this material that I will discuss with you today.

Senator HART. The committee remembers the testimony of Professor Preston.

Dr. COLLINS. Both economic theory and industrial experience suggest that the structural features of an industry strongly influence the competitive behavior of its member firms and the performance outcomes-prices, profits, output, et cetera-in its markets. Only under narrowly specified and rather extreme theoretical conditions, however, has it been possible to deduce market performance entirely from structural factors. Beyond the limiting cases of complete monopoly and perfect competition, theory is at best a guide to the identification of potentially significant variables and to the development of hypotheses. So we have in the discussions during these hearings data given on the levels of concentration, whether concentration has increased or decreased; and yet we are interested, in addition to this, in the effect of changes or of varying levels of concentration, and it is to this point that I would like to talk today.

Without underestimating the importance of several recent attempts to increase the relevance of economic theory to the analysis of these relationships, it is generally agreed that new and more precise generalizations as to the relation of industry structure to conduct and performance will depend heavily upon continued empirical research. În an early article, Professor Bain pointed out the need for

detailed empirical studies which would formulate specific hypotheses on the relations of market structure to market performance and would then test such hypotheses with available evidence."

Such studies would produce results not only of academic interest but also of fundamental importance for the development of relevant and effective public policies for the promotion of market competition.* The importance of this type of work has been recognized and the conduct of such work substantially assisted by this committee in the development of information through its investigations and in its support of the collection and publication of relevant data.

The comments that we present today are based upon an empirical study of certain structural and performance characteristics of an important segment of American industry, food manufacturing. Broadly speaking, the study on which we report confirms the importance of structural variables in an explanation of the differences in market performance among industries. In particular, the structural variables examined, including the level of concentration, are shown to contribute importantly to the explanation of interindustry differences in price-cost margins.

2 For example see:

G. J. Stigler, "A Theory of Oligopoly," Journal of Political Economy, vol. LXXII, No. 1 (February 1964), pp. 44-61.

M. Shubik. "Strategy and Market Structure: Competition, Oligopoly and the Theory of Games" (New York: John Wiley & Sons, 1958), especially chs. 10 and 11.

R. L. Bishop. "Duopoly: Collusion or Warfare?" American Economic Review, vol. L, No 5 (December 1960), pp. 933-961.

W. J. Baumol, Business Behavior, Value and Growth (New York: Macmillan, 1959), pt. I, chs. 2-8.

3 J. S. Bain, "Relation of Profit Rate to Industry Concentration: American Manufac turing, 1936-40," Quarterly Journal of Economics, vol. LXV, No. 3 (August 1951), p. 293. E. S. Mason in preface to Carl Kaysen and D. F. Turner, "Antitrust Policy: An Economic and Legal Analysis" (Cambridge: Harvard University Press, 1959), p. xx.

FOCUS OF STUDY

The aspect of market performance to which we direct our attention is the relationship between prices and costs. Economic theory suggests that

when oligopolists sell at prices above marginal cost and oligopsonists by at prices below marginal value, relative prices become unreliable as indexes of relative scarcities and relative demands; and the producers, whose policies are guided by market prices, may make socially undesirable decisions. In particular, too little will be produced and too few resources utilized in industries with high margins; and too much will be produced and too many resources utilized in industries with low margins."

It is our purpose here to examine the relationship between industry concentration and certain other structural factors and price-cost margins in food manufacturing industries. We begin with one of the most familiar propositions of economic theory that, other things being equal, prices are higher and price-cost margins wider under conditions of monopoly than under conditions of competition. The "monopoly" and competition" referred to here are theoretical, not emprical, concepts, and other things of course are never entirely "equal.' The theoretical proposition can, however, be converted into an emprically relevant and testable statement by the following series of arguments:

1. The industrial evironment presents a spectrum of structures; these structures may be ordered according to their relative closeness to the theoretical conditions of monoply and competition.

2. The prices of different products in different markets cannot be compared directly; however, the relationship between prices and costs provides a basis for comparison of market performance among quite different industries and products.

3. Since capital requirements differ among industries and these differences are not fully reflected in current costs, it is necessary to take them specifically into account in any analysis of price-cost margins."

This line of reasoning yields the following proposition that we wish to examine empirically: The closer an industry is to a monopolistic structure, the higher will be the price-cost margins, after account is taken of differences in capital requirements.

We define the term price-cost margin, as used here, to be the difference between average price and average costs, expressed as a percentage of average price; and utilize the four-firm concentration ratio computed on the basis of value of industry shipments as the index of closeness to monopolistic structure. There are, of course, other measures of firm size inequality and positions of market power. Professor Kaysen has stated:

Many other features of the market are relevant to this "power relation." At least the following are of equal importance with the number and size dis

Tibor Scitovsky, "Economic Theory and the Measurement of Concentration," Business Concentration and Price Policy (Princeton: Princeton University Press, 1955), p. 104. It may be argued that differences in capital requirements are not relevant to the identification of an association between industry structure and price-cost margins in the shortrun. Such margins are wide because buyers are willing to pay particular prices rather than go without the product. High capital requirements do not, in the shortrun, cause high profit margins; costs will not be covered by revenues simply because they have been incurred. However, the theoretical proposition under examination here deals with equilibrium adjustments, not unstable short-run situations. We therefore must consider the possibility that differences in the current price-cost margins in various industries are requisite for the attainment of equal profits on the capital employed over the long run.