Theory of the Growth of the Firm

by Edith Penrose

  52 At this point I should like to thank W. W. Cooper for the suggestions and comments in his ‘Discussion’ on the paper I presented at the December 1954 meeting of the American Economic Association (American Economic Review, Papers and Proceedings, Vol. XLV, No. 2, May, 1955, pp. 559–65). His remarks were primarily directed to the organizational significance of my analysis but his criticism of my use of the ‘traditional definitions and categories of cost’ and his approval of my introducing what he called ‘hierarchical arrays of planning agents’ (which, I suppose, is one way of describing the type of organization mentioned above), made me re-think and revise some of my former propositions. On the other hand, I must reject, for reasons which will become clear presently, his criticism of my treatment of ‘unused’ productive services.

  53 This possibility was suggested to me by Ralph Turvey. It has significance primarily for the argument to be presented later that unused managerial and other productive services provide an incentive for the expansion of the firm.

  54 Essentially the same point is made in the following statement: ‘. .. while authority may be divided many ways, responsibility can be only partially delegated. An executive can delegate the responsibility for doing a job, but he still retains the responsibility for seeing that the job is done. It is this complication, more than any other, that makes delegation so difficult. For it means that the top executive is fully responsible for checking up on everybody below him, and on everything that might affect the company’s success.’ Perrin Stryker ‘The Subtleties of Delegation’, Fortune, Vol. LI, No. 3 (March 1955), p. 97. See also P. W. S. Andrews, Manufacturing Business (London: Macmillan, 1949), pp. 128 ff.

  55 Francis J. Curtis, vice president of Monsanto Chemical Company, is reported to have used the concept of ‘value added by manufacture’ in discussing the increasing value of an executive as he gained experience in a firm and rose through the ranks: he is quoted as saying ‘that a man 47 years old who had started his business career at 22 at a salary of $3,600 and had advanced to the second echelon from the top management post and a salary of $40,000 was worth to the company, at the very least, $425,000’. Quoted in New York Times, March 29th, 1954.

  56 P. W. S. Andrews has emphasized the same point: ‘It has also to be recognized that a progressive business of any size will have another driving force making for this sort of expansion in the young up-and-coming men, pressing for promotion and looking for ways in which to achieve it. If they see a job which would be likely to be profitable and which they could add to their department, they will press for the consequent extension. This will make management more complex....’ P. W. S. Andrews, op. cit., p. 282.

  57 Frank H. Knight, Risk, Uncertainty and Profit (Boston and New York: Houghton Mifflin, 1921), p. 77.

  58 This fact, of course, in no way interferes with the usefulness for price and output analysis of the ‘static’ assumptions. As we have seen in Chapter II, the ‘size’ of the firm—that is, the output of a given commodity—in the so-called ‘theory of the firm’ need not be materially affected by the process described here.

  59 In this discussion we shall not be concerned with how the ‘optimum’ output, or the ‘maximum-profit’ output, should be calculated by the rational entrepreneur under conditions of uncertainty. This is the usual issue under discussion in the economic literature on risk and uncertainty, but is designed to explore problems rather different from the ones raised here, as will become clear presently.

  60 See especially M. Kalecki, ‘The Principle of Increasing Risk,’ Economica, New Series, Vol. IV (Nov. 1937), pp. 440–7.

  61 Jacob Schmookler has suggested to me that the progressive development of highly systematic and precise planning techniques may have led to a substantial improvement in firms’ notions of what is likely to happen in the future, and thus to a reduction in the incidence of mistakes, especially on the part of the large modern corporations.

  62 The fact that managerial research undertaken to reduce uncertainty involves expenditures to improve the prospects of successful growth, is of some relevance for the economic effects of a profits tax. This has nothing to do with the present discussion, but it has frequently been argued that the imposition of a progressive profits tax will tend to reduce the willingness of a firm to undertake the more risky investments. If, in this context, ‘risk’ includes both risk and uncertainty, as is reasonable, then the effect of such a tax may be substantially to increase expenditures designed to reduce uncertainty, both because the marginal profit may become of less interest to the firm than before the imposition of the tax, and because expenditures reduce taxable profits. If one of the purposes of a firm in attempting to increase its profits is to provide funds for reinvestment in the firm, as we have argued in Chapter II, there is clearly an increased incentive to disguise such reinvestment as expenditure, if by doing so tax payments can be reduced.

  63 In discussing the problems of expansion with the responsible executives of business firms I made a special point of trying to discover what they felt was the ultimate limit on the expansion they would undertake. In a fair number of cases, the firms voluntarily limited their access to capital, either refusing to borrow at all or borrowing only for limited purposes. In almost all other cases, businessmen agreed that managerial capacities set the limit to expansion. If I was told that expansion was not warranted in existing lines for a variety of reasons and I asked why they did not go into new fields, I was almost invariably told, in effect, ‘We are—slowly. We don’t know enough to go outside our field too fast. We can hire experts, but we have to know a bit about these things ourselves.’ As we shall see later (Chapter VIII), even expansion through acquisition is generally related in some way to the existing activities of the firm, although one advantage of acquisition as a method for expanding into new fields is that a firm can acquire not only assets but an experienced managerial group as well. Furthermore, mergers are often preceded by considerable managerial planning of the selection of companies to be acquired and in the execution of the merger.

  64 Short-run ‘flexible’ planning is apparently one of the more common ways in which firms deal with the problems of change. A. G. Hart has a good theoretical discussion of this type of action in Chapter IV of his Anticipations, Uncertainty, and Dynamic Planning (New York: Kelley, 1951). And one empirical investigation of the investment plans of a sample of firms in a large industrial area found that: ‘Firm capital investment plans are typically limited to a few months. Quick and unpredicted changes in such plans are the order of the day’. Walter W. Heller, ‘The Anatomy of Investment Decisions’, Harvard Business Review, Vol. 29, No. 2 (March 1951), p. 99.

  Incidentally, the importance of management as a brake on expansion was clearly brought out in the same study by Heller. In addition to the curbs on capital investment created by the ‘fetish of prosperous companies for internal financing and the inability of limping companies to get outside financing’, he found that ‘One of the unforseen—and most interesting—investment barriers encountered was the bottleneck in top engineering and management talent. In more than half of the companies studied, it was flatly stated that either (1) the postwar pace of capital expansion had been too fast for top management and engineering staffs to handle efficiently and digest thoroughly, and a pause for digestion was now in order, or (2) the rate of capital expenditure had been, and still was being, held down to what the very scarce factor of ‘brains’—engineering and managerial—could handle. One executive put this point particularly effectively:

  ‘Perfecting a layout involves a minimum amount of managerial and technical work that you can’t escape. You have to handle expansion projects in a series, because you simply don’t have the necessary number of men of the required caliber around to keep up a doubled-up pace,’ Ibid., p. 102.

  65 Even those raw materials which are in principle finely divisible must usually be acquired in minimum-sized bundles because to acquire less than the ‘standard unit’ is usually disproportionately expensi
ve. However, this type of indivisibility is probably not of much practical importance.

  66 In 1832 Charles Babbage pointed out: ‘the extent of a factory. . . ought to consist of such a number of machines as shall occupy the whole time of one workman in keeping them in order, and in making any casual repairs; if it is extended beyond this, the same principle of economy would point out the necessity of doubling or tripling the number of machines, in order to employ the whole time of two or three skilful workmen’. Charles Babbage, On the Economy of Machinery and Manufactures (London: Charles Knight, 1832), p. 175. So far as I know, E. A. G. Robinson was the first to use the term ‘balance of processes’. The Structure of Competitive Industry (New York: Harcourt Brace, 1932), pp. 31–5. P. Sargant Florence uses the term ‘principle of multiples’, The Logic of British and American Industry (London: Routledge and Kegan Paul, 1953), p. 51.

  67 See, for example, E. A. G. Robinson, op. cit., p. 33.

  68 For example, an industrial engineer in charge of product development in a firm is quoted as having stated: ‘Every time we make something, we have something left over, and have to find something to do with that. And when we find something to do with it we usually find that leaves us with something else. It is an endless process.’ A. D. H. Kaplan, Big Enterprise in a Competitive System (Washington D.C.: Brookings Institution, 1954), footnote, p. 191.

  69 For a good discussion see P. Sargant Florence, op. cit., pp. 74 ff. He summarizes his argument as follows: ‘The economy of integration due to common costs boils down to this, that if a manufacturer has a certain unused capacity in equipment or in research or in finance (or brains of himself and staff), it may pay to “balance up” by taking on as a side line new processes or products using that idle capacity. As a summary to the American survey puts it under the heading of utilization of resources, “diversification may result from an attempt to make full use of managerial or manufacturing capacity”, op. cit., p. 76. The ‘American survey’ referred to is that of the Temporary National Economic Committee, The Structure of Industry, Monograph 27 (Washington, D.C., 1941).

  70 Not all ‘idle’ or ‘free’ services, however, provide genuinely profitable opportunities. The problems of expanding on such a basis—sometimes called ‘burden absorption’ or ‘creep’ are discussed in Chapter VII.

  71 See, for example, Allyn Young, ‘Increasing Returns and Economic Progress’, Economic Journal, Vol. XXXVIII, No. 152 (Dec., 1928), pp. 527–42. The increasing scope for the division of labour led to a disintegration of industry in the sense that different processes became concentrated in separate specialized firms. There is no technological or organizational necessity for division of labour to take this form; it could just as well have taken the form of a division of function between different establishments of the same firm or different parts of the same establishment, and if there were no restrictions on the rate of growth of firms I should expect this to have occurred. One reason for the limit to the rate of expansion of firms was the difficulty of obtaining capital for expansion by the individual firm, particularly in the days before the corporation or joint-stock company was the normal form of industrial enterprise and equity markets little developed. But in addition the fact that the rate of internal expansion of firms is limited under any circumstances encourages specialization of firms in periods of rapidly growing demand. This limit on expansion was even more restrictive in the 19th century than it is today, for acquisition and merger were not common means of expansion before the ‘corporate age’.

  72 There are many discussions of the relation between specialization and the scale of operations of firms and there is no need to give an extensive review of the subject here. Among the more comprehensive discussions is that of P. Sargant Florence in the works already cited.

  73 P. Sargant Florence, op. cit., p. 52.

  74 As a matter of fact, no consistent principle other than practical convenience has been adopted. In some circumstances the service itself can be expressed in homogeneous units (BTU’s for coal, haulage-miles for trucks) and these units may be used; in others a unit of the resource itself has to be used (bales of cotton, pounds of sulphuric acid). The chief problem is to obtain a classification related to the nature of the resource within which the required degree of homogeneity exists. When a choice of units is feasible, measures that reflect most directly the actual services rendered tend to be preferred.

  75 The subdivision cannot go so far that each input is defined as a separate resource, however. The only purpose of devising a ‘unit’ of resources or services is to enable us to measure the number of units within a given category. If this number is always one, no purpose is served by the classification.

  76 The argument that the patent system stimulates invention implies that increased knowledge is a function of prospective profits. The economics of the argument have never been developed with any rigour but it remains one of the few economic ‘models’ in which knowledge becomes a function of an economic variable. See E. T. Penrose, Economics of the International Patent System (Baltimore: Johns Hopkins Press, 1951), pp. 34 ff.

  77 Frank H. Knight, Risk, Uncertainty, and Profit (Boston: Houghton Mifflin, 1921), p. 61.

  78 Ibid., p. 97.

  79 A study of the significance of the fact that new technology becomes embodied physically in capital equipment has been made by W. E. G. Salter in a monograph to be published by the Cambridge Institute of Applied Economics.

  80 For example, Alderson and Sessions, marketing and management counsellors, have pointed out that ‘. . . it is essential to distinguish between what the economist has called the elasticity of demand and the more fundamental factor of plasticity. The intended difference is suggested by the common meaning of the words. “Elastic” refers to something that can be stretched, and “plastic” to something that can be molded. Economics long ago pointed out that demand can be stretched to include more units of a product by the simple expedient of reducing the price. Much less attention has been devoted to the fact that demand can often be remolded into quite different forms. The investigation of plasticity of demand has generally been left to the market analyst rather than to the economist. The re-molding of demand to make a place for new products has proceeded to a spectacular extent in the United States. To make use of the innate plasticity of demand means to find ways of changing the habits and attitudes of consumers. Changing a buying habit means, among other things, making it as convenient as possible for consumers to buy the new product. Changing buying attitudes means supplying consumers with reasons for preferring the new product’. Cost and Profit Outlook, Vol. 5, No. 8 (Aug. 1952). Italics mine.

  81 In fact, when one considers the almost appalling efficiency with which market specialists, using psychological and sociological information and techniques developed in universities and elsewhere, are learning to manipulate consumers (in the guise, of course, of meeting their unexpressed needs and desires), one almost wonders if in time the whole of the economist’s theory of the market will not have to be completely reversed and no place at all left for the notion of independently formed consumer demand. Here, eventually, may lie the real origin of the ‘Brave New World’ to come.

  82 Cf. the comment of Alfred Sherrard in ‘Advertising, Product Variation, and the Limits of Economics’, Journal of Political Economy, Vol. LIX, No. 2 (April 1951), p. 134.

  83 The point can of course, be expressed in terms of cost: those products for the production of which a firm is thoroughly unsuited can only be produced at prohibitive or ‘infinite’ cost. This is not a convenient way of looking at the matter, however, partly because businessmen don’t think in these terms about products they don’t think about at all.

  84 We return to this point in Chapter IX.

  85 There have been a number of such studies. See, for example, Robert Eisner, ‘Determinants of Capital Expenditures: An Interview Study’, Studies in Business Expectations and Planning, No. 2 (Urbana: University of Illinois, 1956). Unfortunately the more extensive study of inf
luences on the investment decisions of firms by John R. Meyer and Edwin Kuh, The Investment Decision: An Empirical Study (Cambridge: Harvard University Press, 1957) could not well take into account anticipated demand for a firm’s products because of the fact that such anticipations could not easily be related statistically to any of the ‘observed’ variables with which the study was concerned. It is, therefore, of no help to us here.

  86 Joseph Schumpeter has stated ‘That new commodities or new qualities or new quantities of commodities are forced upon the public by the initiative of entrepreneurs. . . is a fact of common experience. . . ’. ‘The Instability of Capitalism’, Economic Journal, Vol. 38 (1928), p. 379. G. H. Evans, Jr., has taken exception to this view, arguing that ‘...the entrepreneur has been essentially an economic opportunist. . . . This hypothesis is in sharp contrast to the assumption that change in consumers’ taste is incident to, and brought about by producers’ action.’ ‘A Theory of Entrepreneurship’, Journal of Economic History, Vol. II (1942), p. 142. One may, however, at the same time hold that the entrepreneur is an ‘opportunist’ and that he moulds the tastes of consumers if the ‘opportunities’ to which he responds arise, not from the market, but from some other source.