Theory of the Growth of the Firm

by Edith Penrose

  212 Richard B. Tennant, in his extensive study of the American cigarette industry, provides an example of this. He writes: ... if technological methods had remained unchanged, it seems unlikely that cigarette firms as large as those we know today could have grown up. ... With the introduction of machinery, however, the problem of supervising hand labour was removed, and the principal obstacle to continued industrial concentration disappeared. It was possible to increase the output of a firm simply by adding more machines, and the firm was free to expand to the extent that its sales would allow. There was no obvious limit to the number of machines which a firm might operate without decline in efficiency. In this purely negative sense, the adoption of machinery removed potential barriers to the scale of enterprise’. Richard B. Tennant, The American Cigarette Industry (New Haven: Yale University Press, 1950), p. 20.

  213 This implies, of course, that so long as large firms are able to maintain their rate of growth through the acquisition of other firms there will be an increasing degree of concentration of assets in the hands of the large firms which must, by definition, eventually bring the process to an end. As we shall see in our analysis of concentration, it is only under the most peculiar set of assumptions that it is possible to imagine a situation where the rate of growth of individual firms remains unchanged throughout an indefinite lifetime while at the same time no change takes place in the concentration of industrial activity.

  214 Furthermore from the above analysis we cannot as yet say anything about the relative rates of growth of large firms as a group and small firms as a group where the large firms are in a position to block the expansion of small firms. It must be remembered that we have been considering the individual growing firm with an opportunity enabling it eventually to reach the ‘large’ category; the other question is dealt with in Chapter XI.

  215 That acquisition is quantitatively less significant for large than for small firms in the United States in recent times has been clearly demonstrated by J. K. Butters, J. Lintner, and W. L. Cary, The Effect of Taxation on Corporate Mergers (Boston: Harvard, 1951), Ch. IX.

  216 Furthermore, calculations of the statistical rate of growth of a firm will vary depending on the measure of size adopted, whether total sales, assets of one kind or another, employment, or something else. It should be noted that if total employment is used as a measure of size and growth, the above analysis of the substitution of capital for labour as a means of maintaining the rate of growth will not be applicable.

  217 See, for example, the analysis of the pattern of growth of twelve major steel companies in the United States between 1900 and 1950 in Gertrude C. Schroeder, The Growth of Major Steel Companies, 1900–1950 (Baltimore: Johns Hopkins Press, 1952), especially, pp. 206–7. Although average rates of growth from one short period to the next indicate, as would be expected, substantial spurts and lags, the trend or ‘growth curve’ is clear enough. It is astonishing how little careful statistical work has been done on the growth pattern of individual firms. This study and an unpublished dissertation by Edgar O. Edwards, Studies on the Growth of the Individual Firm (Baltimore: Johns Hopkins University 1951) are the only works of which I know dealing with the rates of growth of a number of industrial firms over a considerable period.

  218 See, for example, P. Sargant Florence, The Logic of British and American Industry (London: Routledge and Kegan Paul, 1953), pp. 63–6; J. Steindl, Small and Big Business (Oxford: Basil Blackwell, 1945), pp. 59–61; and A. D. H. Kaplan, Small Business (New York: McGraw-Hill, 1948).

  219 Where even such skills are lacking there is still the possibility of starting a small commercial shop. Here we are concerned only with manufacturing firms.

  220 Incidentally, it might be noted that if, in fact, it is the type of entrepreneurial and managerial ability as much as the lack of capital, which causes small firms to cluster in fields where opportunities are relatively poor under any circumstances, measures to ease their task of obtaining capital may not have a significant effect on the mortality of small firms.

  221 It may often happen, of course, that a field already developed by a small firm will attract the interest of a larger firm which thereupon enters and destroys the small firm’s opportunity, either by driving it out of business or purchasing it outright. See the discussion in the next chapter of the effect on the interstices of acquisition and merger.

  222 The bearing of the present discussion on the process of industrial concentration is discussed in the next chapter. I remind the reader, incidentally, that all of the analysis in this book is intended to be applicable only to economies where the industrial corporation is the dominant form of business organization; hence not to industrial development occurring before the last quarter of the last century.

  223 I should perhaps once more repeat that none of this discussion has any bearing on the question whether or not large firms have higher profits or lower costs of production than small firms. No consensus on either has yet emerged from the various attempts to investigate comparative profits and costs, the results being conflicting and to a large extent non-comparable because different investigators use different data, different techniques, applied to different types of firms at different periods of time. I am not prepared to take any position on this controversial subject. Useful recent surveys, evaluations, and citations of the literature dealing with attempts to compare profits or costs statistically will be found in P. Sargant Florence, op. cit.; Richard C. Osborn, ‘Efficiency and Profitability in Relation to Size’, Harvard Business Review, Vol. 29, No. 2 (March 1951), pp. 82-94; and Hans Staub, Le Profit des Grandes Entreprises Americaines (Paris: Colin, 1954).

  224 It should be noted that a firm has no incentive to treat ‘investment’ expenditures of the kind referred to above as anything other than costs of current operations unless it wants to adjust its earning figures to show a better profit position for the sake of attracting investors or enhancing its general reputation. We have argued that firms prefer to retain as much of their earnings as possible for reinvestment in the firm. To treat certain types of reinvestment expenditures as costs may permit a higher total retention of earnings (as is well known for depreciation) both because it may reduce the pressure to pay higher dividends and because it reduces taxable income of the firm where corporate income is taxed. So long as profit figures are ‘satisfactory’ there is, on the contrary, an incentive to increase expenditures which are in fact for the purpose of increasing profits in the future and to call these expenditures ‘current costs’.

  225 The following discussion is not relevant to the situation, originally analysed by E. H. Chamberlin and Joan Robinson, in which ‘excessive entry’ may reduce economic efficiency by raising the costs of existing producers. In their models, existing producers were forced to contract their output because of competition from newcomers, and no producer had any competitive advantage. On this subject, see the remarks of Fritz Machlup, op. cit., pp. 312 ff. In our discussion, however, we are concerned only with restrictions on the ability of firms to take advantage of opportunities for expansion which would otherwise be neglected, and we assume that the larger producers have a competitive advantage over smaller ones.

  226 ‘... in the modern industry shared by a few large firms, size and the rewards accruing to market power combine to insure that resources for research and technical development will be available. The power that enables the firm to have some influence on prices insures that the resulting gains will not be passed on to the public by imitators (who have stood none of the costs of development) before the outlay for development can be recouped. In this way market power protects the incentive to technical development’. J. K. Galbraith, American Capitalism: The Concept of Countervailing Power (Boston: Houghton Mifflin, 2nd ed., 1956), p. 88. For an attempt at a statistical investigation of this argument see Almarin Phillips, ‘Concentration, Scale and Technological Change in Selected Manufacturing Industries 1899-1939’, The Journal of Industrial Economics, Vol. IV, No. 3 (June 1956), pp. 179
-93. Using census data this study attempted to test the hypothesis that large firms are in a better position to innovate than are small firms. It was found that the ‘hypothesis that industries with large numbers of small firms tend to be technologically more progressive, while not disproved, received no support from the data’ (p. 193); in general the conclusion was drawn, though ‘with caution’, that ‘industries in which production is concentrated in a few firms and in which firms are relatively large’ do show more signs of technical change than industries in which these conditions do not exist’ (p. 192).

  227 This is clear from the most cursory glance at the literature, and is evident in the very phrase ‘New Competition’ which has been widely adopted to describe the relations between big business units. ‘Competition, always the mainspring of our economy and of the dynamics of American life, in mid-twentieth century America has been stimulated and quickened by Big Business. As a consequence competition has taken on a renewed vitality and diversity, a new dimension and a new content.’ David E. Lilienthal, Big Business: A New Era (New York: Harper, 1952), p. 47. See also Oswald Knauth, Business Practices, Trade Position, and Competition (New York: Columbia University Press, 1956); Peter Drucker, Concept of the Corporation (New York: John Day, 1946), pp. 122 ff., 247 ff., Herrymon Maurer, Great Enterprise: Growth and Behavior of the Big Corporation (New York: Macmillan, 1955), pp. 171 ff.; and A. D. H. Kaplan, Big Enterprise in a Competitive Society (Washington D.C.: Brookings Institution, 1954) Chs. VIII, IX, X.

  228 It may be that the seeds of the ‘new case’ are already being planted by those who stress the influence on the decisions of the large firms of a strong sense of responsibility towards the public, and sometimes go so far as to give the impression that the large firm operates primarily as a public service.

  229 A recent American study of barriers to entry found that economies of scale or ‘absolute cost advantages of established firms’ were less significant in most concentrated industries than ‘the advantage to established sellers accruing from buyer preferences for their products’, especially when the size of the United States market was considered. See Joe S. Bain, Barriers to New Competition (Cambridge: Harvard University Press, 1956), p. 216.

  230 It is under these circumstances that the strictures of J. M. Clark might have their greatest significance. He called knowledge the ‘ ... industrial instrument with unlimited capacity, which shows increasing economy with increasing utilization, and never reaches the stage of “diminishing return”. To be efficient, the enterprise of extending the frontiers of industrial and economic knowledge must be carried out on a large scale; yet the results must also be made available for the benefit of smaller scale producers, unless we are to submit to a ruinous waste of overhead in this respect. If industrial research becomes the sole perquisite of the concern which can afford an expensive laboratory, there is an end of economic freedom, and as a long-run result, perhaps, an end of economic efficiency, owing to bureaucratic stagnation.’ J. M. Clark, Studies in the Economics of Overhead Costs (Chicago: University of Chicago Press, 1923), p. 141. See also a discussion of an analogous problem from an international point of view in E. T. Penrose, The Economics of the International Patent System (Baltimore: The Johns Hopkins Press, 1951), Ch. VI.

  231 Another, rather different approach would be to break up the big firms into several smaller firms. This might have temporary disruptive effects but would have no long-run adverse effects on economic progress if economies of growth rather than of size account for their size. The strongest case to be made for the big firms rests on the costs of research, and in this respect the Chairman of the United States Steel Corporation has recently stated that the largest firms in the United States today are still too small to achieve the maximum rate of progress. But see the discussion of dominance later in this chapter.

  232 G. W. Nutter, I think, was quite correct in his observation that it is ‘ ... not unreasonable to view mergers as a force counteracting expansion of markets, and hence as leading to less change in the structure of dominance than is usually believed. ... The early merger movement is important in the history of industrial concentration because it made concentration, taken in a relevant sense, greater than it otherwise would have been, irrespective of whether it actually increased concentration or not’. G. Warren Nutter, ‘Growth by Merger’, Journal of the American Statistical Association, Vol. 49, No. 267 (Sept. 1954), p. 456.

  233 For example, Butters, Lintner, and Cary, speaking of merger activity between 1879 and 1903 state, ‘Even though many of these early mergers subsequently proved to be unsuccessful, the evidence is clear that this wave of merger activity in itself largely created the characteristic twentieth-century pattern of corporate concentration’. J. K. Butters, J. Lintner, and W. Cary, Effects of Taxation on Corporate Mergers (Harvard: 1951), p. 289. And Jesse Markham states, ‘The conversion of approximately 71 important oligopolistic or nearcompetitive industries into near monopolies by merger between 1890 and 1904 left an imprint on the structure of the American economy that fifty years have not yet erased’. ‘Survey of the Evidence and Findings on Mergers’, Business Concentration and Price Policy (Princeton: Princeton University Press for the National Bureau of Economic Research, 1955), p. 180.

  234 The ‘holding company’ device is very widely used in Australia to consolidate ownership interests, although some of the largest companies scarcely meet our definition of an industrial firm. See the paper by L. Goldberg and D. M. Hocking, ‘Holding Companies in Australia’ delivered before the 1949 meeting of the Australian and New Zealand Association for the Advancement of Science. Some of the companies originally only holding companies have subsequently converted themselves into operating firms with much closer administrative structure, for example, Felt and Textiles, Ltd., of Melbourne. (See the Annual Report of this firm for 1954).

  235 Gideon Rosenbluth, Concentration in Canadian Manufacturing Industries (Princeton: Princeton University Press for the National Bureau of Economic Research, 1957), p. 102.

  236 George Stigler has maintained that since the 1900s in the United States merger has been a significant force in a shift from monopoly to oligopoly in significant industries: ‘The change has been most striking in the industries which were merged for monopoly at the beginning of the century. The merger firm has declined continuously and substantially relative to the industry in almost every case. The dominant firm did not embark on a new program of merger to regain its monopolistic position, however; the new mergers were undertaken by firms of the second class. The industry was transformed from near-monopoly to oligopoly’. ‘Monopoly and Oligopoly by Merger’, American Economic Review, Vol. XL, No. 2 (May 1950), p. 31, J. F. Weston challenges Stigler’s conclusions on the ground that other forces, notably antitrust dissolutions and internal expansion of rivals, were more important in the emergence of oligopoly than was merger. J. Fred Weston, The Role of Mergers in the Growth of Large Firms (Berkeley: University of California Press, 1953), pp. 35–7. (See, however, G. Warren Nutter’s review of Weston’s argument with Stigler in the Journal of the American Statistical Association, loc. cit.). We are not here concerned with the motives for merger (whether ‘for’ monopoly or ‘for’ oligopoly) as were Stigler and Weston, but rather with its effects, and as to the latter, there seems little question that the later merger movements in the United States did increase oligopoly. Cf. Jesse Markham, loc. cit. p. 179.

  237 Whereas some of the earlier students of the subject held that there was a close correlation between mergers and the business cycle (for example, Willard Thorp, ‘The Persistence of the Merger Movement’, American Economic Review, Vol. XXI, No. 1, p. 85, more recent research has cast much doubt on the existence of any correlation. See Jesse Markham’s excellent summary of the research on this subject, loc. cit., pp. 146–54.

  238 ‘Whereas the high security prices of the 1920s, for instance, encouraged merger activity by making outside capital relatively cheap, the low security prices of recent years (in relation to asse
t values) have also made the merger route of expansion attractive to many companies by providing an attractive outlet for the accumulated financial resources of the acquiring companies’. J. K. Butters, J. Lintner, and W. Cary, op. cit., p. 312.

  239 I am indebted to Donald Whitehead for raising this point.

  240 The available information on the contribution of merger to the growth of small and large firms in the several merger ‘movements’ in the United States is well discussed by Butters, Lintner, and Cary, op. cit., Ch. X. No purpose would be served by summarizing it here.