87 The early struggles of the aluminium industry provide one of the many examples of this. Great difficulties were experienced by the new firm in getting its product recognized and accepted. It was early brought up against ‘ . . . the greatest truth in all business: that no matter what you have to offer, it takes a vast amount of study and ingenuity to fit it into the world’s needs and an unbelievable amount of persistent argument and shoe-leather to sell it’. An official of ALCOA quoted by Robert R. Updegraff in ‘Aluminum Tells Its Story’, The Magazine of Business, Vol. I, 56. (Aug. 1929), pp. 123–5.
88 Several writers concerned with the study of the business firm have, of course. made this same point. For example: ‘The clue to successful formulation of objectives [for the firm] is to think in terms of what the company can accomplish through use of its resources rather than in terms of what products it may happen to find.’ Thomas A. Staudt, ‘Program for Product Diversification,’ Harvard Business Review, Vol. 32, No. 6 (Nov.–;Dec. 1954), p. 124. And A. D. H. Kaplan speaks of the firm as ‘a pool of resources, the profitable use of which requires that products and processes be revised and machines be improved if they are to win and retain market acceptance.’ Op. cit., p. 157.
89 One businessman has stated frankly that attempts ‘to determine what is wanted by the consumer and developing value to meet this want...can be based only on opinion. Very often this opinion is that of a few individuals with strong, and perhaps strongly subjective, reactions and rarely is it based on factual data.’ In fact he considered this approach ‘uneconomical and wasteful of time and money’, preferring free to develop new products from research, and then to sell them to consumers by developing markets using ‘adequate techniques’. Quoted in Research Operations in Industry. (David B. Hertz, editor). Papers delivered at the Third Annual Conference on Industrial Research. June, 1952. (New York: King’s Crown Press, 1953), pp. 232–3.Even the opposite approach, the ‘evaluation of sales intelligence’, is often not really a question of discovering what consumers actually demand, but one of what they ought to demand. ‘. . . sales intelligence is the nexus of communication between buyer and seller, guiding the seller toward those areas where ingenuity and sweat can most profitably be expended....The buyer may not even realize that they are problems. But many an industry has been founded on a seller’s alert perception of an unnoticed need—that is, his skill in gathering and evaluating his sales intelligence’. Editors of Fortune, Why Do People Buy? (New York: McGraw-Hill, 1953), p. 168.
90 An analysis of this question will be made in Chapter X.
91 Nearly every economist concerned with the operations of industry has at some point discussed this subject from Babbage in 1832, through Marshall, both in the Principles and in his Industry and Trade, up to the modern discussions of E. A. G. Robinson in The Structure of Competitive Industry (New York: Harcourt Brace, 1932), and P. Sargant Florence in The Logic of Industrial Organization (London: Kegan Paul, 1933) and The Logic of British and American Industry (London: Routledge and Kegan Paul, 1953). Abba Lerner in Economics of Control (New York: Macmillan, 1944) and Fritz Machlup in Economics of Sellers’ Competition (Baltimore: Johns Hopkins Press, 1952) have useful theoretical discussions. A recent review of the empirical work on the subject as well as a penetrating analysis of the theoretical and conceptual problems can be found in Caleb Smith, ‘Survey of the Empirical Evidence on Economies of Scale’, in Business Concentration and Price Policy (New York: Princeton University Press for the National Bureau of Economic Research, 1955), pp. 213 ff. Many competent industry studies, too numerous to list, are also available.
92 P. Sargant Florence, for example, discusses the economies of large-scale operation in a very generalized form, recognizing that the same fundamental principles lie behind all kinds of economies regardless of which type of organization they apply to. He analyzes three basic principles: the ‘principle of bulk transactions’; the ‘principle of massed (or pooled) reserves’; and the ‘principle of multiples’. The Logic of British and American Industry, op. cit., pp. 50–1.
93 Where this is not true, the mere fact that a variety of productive activities are carried on in close geographical proximity does not provide an economic reason for calling the collection of activities a single plant.
94 It should also be noted that changes in the price of the final product change the opportunity cost of resources used in its production and therefore their value to the firm. This has relevance not only for the optimum size of plant in an industry, but also for the scale on which each of the particular products of a multi-product firm will be produced. In addition, even if an important resource used is specific to a single product and has no opportunity cost to the firm, its value will change with the price of the product and rent should be imputed to the resource and added to the cost of production. This, too, will change the optimum size of plant. For a full discussion of this point see Fritz Machlup, op. cit., pp. 288–99.
95 This is not a fanciful illustration. Executives of a large firm will often tell you of smaller firms—sometimes very small firms—that can produce one of the large firm’s products at the same, or lower cost merely because the small firm does not have the managerial overhead of the larger.
96 W. Baldamus, in a very interesting article, has attempted to explain the trend in plant size in a variety of industries in Britain in terms of the relative influence of ‘mechanization’ and ‘utilization’—the former being an increase in output by expansion of plant, the latter, by more intensive utilization of existing plant. He found that in the ‘newer’ industries, mechanization, i.e., ‘technical progress’, tended to be largely responsible for rapid increases in the size of plants. ‘Then there comes a point when utilization takes over as the dominant principle controlling expansion, because it is no longer possible or profitable to carry through radical technological innovations.’ W. Baldamus, ‘Mechanization, Utilization and Size of Plant’, Economic Journal, Vol. LXII (March 1953), p. 68.
97 The reader will notice the similarity between the discussion of the economies of size and our earlier analysis of the significance of unused productive services. Since the two discussions are merely different ways of analyzing the same problem, some repetition is inevitable but will help to bring out the relation between the economies of size and those of growth.
98 This principle has been particularly stressed by P. Sargant Florence under the name of ‘massed (or pooled) reserves’ He points out that it appears in ‘ . . . many apparently unrelated branches of economic life . . . in schemes for the decasualization of labour at the docks and underlies all forms of insurance and banking; the reserves that are economized may in fact be labour, liquid monetary resources, stocks of goods and materials or any other factors in production, when the demands upon these factors are somewhat uncertain in their incidence.’ It illustrates ‘the statistical theory of large numbers, based on probable error, that the greater the number of similar items involved the more likely are deviations to cancel out and to leave the actual results nearer to the expected results. The probable deviation in orders for similar items that a reserve guards against is thus proportionately less when orders are many, and the cost of reserves per unit of output falls correspondingly’. P. Sargant Florence, op. cit., pp. 50–1.
99 See, for example, E. A. G. Robinson, op. cit., pp. 39–40; and P. Sargant Florence, op. cit., p. 52.
100 I am not suggesting that the persistent removal of the fruit would not reduce the future supply of fruit; I am only concerned to make clear the distinction between the two types of economies of size. Incidentally, that there may be economies in expansion without enduring economies of size often helps to explain the voluntary reduction in the size of a firm when it sells one of its ‘businesses’ to another firm. See the discussion below in Chapter VIII of the ‘Purchase and Sale of Businesses That are Not Firms’.
101 The significance of this is discussed more fully in the next chapter, which deals with the economics of diversification.
102 It should be clear, once again, that in rejecting the notion of an optimum size of firm in this context we are not quarrelling with the concept of the optimum size of firm as it appears in the ‘theory of the firm’, since the ‘optimum firm’ in that context is merely the optimum output of a given product. Furthermore, the above discussion refers only to the firm, and not to the plant—there probably is in any particular economy an optimum size of plant for many industries.
103 Essentially the same point is made by E. A. G. Robinson with respect to the costs of selling: ‘The cost of selling is only in part, and in certain conditions, a cost of production. At other times and in other conditions, it is a cost not of producing but of growing. For once the market has been won, it can be retained at a lower selling cost than is necessary to secure it initially. We have, then, two quite distinct questions to which we must give an answer. First, is a larger firm more efficient than a smaller firm? Second, will it pay to grow from being smaller to being larger? The high cost of selling may be, paradoxically, at the same time a source of economy, making the already large firm more efficient than the smaller firm, and a cost of growth which makes it unprofitable for the small firm to grow up to its most efficient size.’ E. A. G. Robinson, op. cit., p. 67.
104 For example, the executives of one prominent United States firm that I studied felt strongly that increased size brought nothing but administrative headaches; at the same time they knew they could not afford to pass up promising opportunities for expansion.
105 On the other hand, many large firms insist on the social advantages of their large size, largely, I suspect, because they feel they must justify their existing state. Even if size is no advantage, no firm wants to be broken up by outside action.
106 Sometimes the ‘evolutionary process’ takes place in such a way that the new activities are no longer desired by the firm. For example, in 1920 General Motors Corporation organized a consolidation of several manufacturers of plate glass as part of the Fisher Body Corporation, a subsidiary of General Motors, because at that time Fisher Body was having trouble obtaining adequate supplies of plate glass. In 1931 General Motors reported: ‘Through evolution the situation has changed and the point was reached at which it became essential for the Corporation to enter into the manufacture and general distribution of plate glass and allied products, or to retire from the field and turn its interest over to others. For that reason it was decided to sell the productive properties of the National Plate Glass Company to the Libby-Owens-Ford Glass Company.’ 1931 Annual Report, p. 11.
107 Nicholas Kaldor, for example, concludes that “spreading of production” is always attended with some cost; i.e. the physical productivity of a given quantity of resources calculated in terms of any of the products will always be less, the greater the number of separate commodities they are required simultaneously to produce.’ He gives the following as ‘evidence’ for his proposition: ‘That this is the case for a large proportion of jointly produced commodities is shown by the fact that the development of an “industry” is always attended by “specialization” or “disintegration”, i.e. the reduction of the number of commodities produced by single firms.’ Nicholas Kaldor, ‘Market Imperfection and Excess Capacity’, Economica, Vol. II (New Series), 1935, p. 48.
And P. Sargant Florence writes: ‘Integration within a plant or a firm must be suspected therefore of small-scale inefficient production till it is proved innocent.’ The Logic of British and American Industry (London: Routledge and Kegan Paul, 1953), p. 74.
108 The process finds great favour in the eyes of those who see it as an important means of sustaining investment in a highly developed capitalist economy.
109 Temporary National Economic Committee, The Structure of Industry, Monograph No. 27 (Washington, D.C., 1941) Part VI, ‘The Product Structures of Large Corporations’.
110 Report of the Federal Trade Commission on Industrial Concentration and Product Diversification in the 1,000 Largest Manufacturing Companies, 1950 (Washington D.C., 1957).
111 The report gives a brief discussion of the limitations of product-class enumerations as a measure of the diversity of a company’s manufacturing activities, pointing out the uneven refinement of the classifications, the conceptual difficulties in the meaning of ‘diversity of activity’, and the different significance of different products from the point of view of the manufacturing process. Ibid., pp. 25–6.
112 Nor can the difficulty be evaded by trying to distinguish diversification into different products (or product-classes) from diversification into different industries, because the same problems plague the attempt to define an industry. If, for example, the elasticity of substitution in consumption is the criterion for defining an industry, then women’s and children’s shoes belong to entirely different industries, for surely the elasticity of substitution is close to zero. If raw materials are the criterion, then canvas and leather shoes are different industries. Clearly we are no better off than we were before.
113 The same physical product from a production point of view may be a different product from the selling point of view, and a firm, by entering new markets with old products, may achieve results similar to those achieved by diversification. In order to avoid overloading the concept of diversification and thereby confusing the reader we have not included this kind of ‘market diversification’ in our definition, but probably most of the considerations influencing diversification from the same production base are relevant to the entry of a firm into new markets with old products.
114 For a perceptive discussion of research and innovation as an ‘ordinary’ business activity see Carolyn S. Solo, ‘Innovation in the Capitalist Process’, Quarterly Journal of Economics, Vol. LXV, No. 3 (Aug. 1951), pp. 417–28.
115 I am not here concerned with the growth taking place during the process of acquiring such monopoly positions through extensive merger but with the growth that takes place after such a position has been obtained.
116 This is, of course, precisely what Schumpeter maintained: ‘But in capitalist reality as distinguished from its textbook picture, [the]. . . kind of competition which counts [is] the competition from the new commodity, the new technology, the new source of supply, the new type of organization. . . it is hardly necessary to point out that competition of the kind we now have in mind acts not only when in being but also when it is merelya never-present threat. It disciplines before it attacks.’ Joseph Schumpeter, Capitalism, Socialism and Democracy (New York: Harper, 2nd ed. 1947), pp. 84–5.
117 In a recent survey of industrial research in the United States in which officials of firms were asked about the factors inducing their firms to undertake research, the role of competition was prominent: ‘Research officials in virtually all industries cited competition as an important consideration—often an overriding one—in appraisals of company research needs.’ A common refrain was ‘we have to do some research to stay in business’; and the importance of research for ‘growth’ and for ‘diversification objectives’ was stressed. National Science Foundation, Science and Engineering in American Industry: Final Report on a 1953–1954 Survey. (Washington D.C.: U.S. Government Printing Office, 1956), pp. 43–4.
118 For example, in a recent survey of 110 companies rated as ‘excellently managed’ by the American Institute of Management, almost nine-tenths of the respondents to a questionnaire expressed the opinion that ‘staying abreast or out ahead in the innovative race is more important to their long-range business success than a “defensive” policy of basing prices closely on costs.’ James S. Earley, ‘Marginal Policies of “Excellently Managed” Companies’, American Economic Review, Vol. XLVI, No.1 (March 1956), p. 59.
119 The research laboratory is itself one of the most important sources of new ideas for further research. In one survey of 121 outstanding industrial research laboratories, questions were asked about the frequency with which research projects originate from various sources. It was found that 50 per cent came from existi
ng research activities, including commercial research. David B. Hertz (ed.), Research Operations in Industry (New York: King’s Crown Press, 1953), p. 122.
120 For example, the chief products of Schenley Distillers Corporation required a knowledge of grains, fermentation processes, and moulds. When antibiotics began to be developed, the firm found that its knowledge and research gave it a ‘better understanding of the fundamental nature of the complex fermentation processes involved and hence an important advantage in this new and rapidly growing industry. (Annual Report, 1948).
Theory of the Growth of the Firm Page 43