II. FROM COMBINATION TO RATIONALIZATION

Plant’s conviction, outlined by Coase, that “governments often served special interests, promoted monopoly rather than competition” was probably an echo of an established tradition in British economic and political thought that dates from the end of the eighteenth century, and that delayed the adoption of legislation in defense of competition until well after WWI (Swann et al. Reference Swann, O’Brien, Maunder and Howe1974). Works by Henry W. Macrosty (Reference Macrosty1907) and Hermann Levy (Reference Levy1911) had shown that industrial combination as a way of evading competition was a widespread practice in British industry. Since 1890, Marshall had argued that combination is not necessarily the opposite of competition, and one of his pupils, David Hutchison Macgregor, developed this view in his first monograph.^{Footnote 2} Since 1893, however, Marshall had been aware that combination was frequently pursued for the sake of saving the least efficient firms, and in Industry and Trade he made it clear that this tendency had been evident in British industry before the war.^{Footnote 3}

The war strengthened this tendency. The reports of the Royal Commission appointed to investigate “Trusts,” and of the standing committees appointed on the basis of the Profiteering Acts of 1919 and 1920, demonstrated that competitive conditions had become more the exception than the rule. However, the general philosophy that informed the work of the commissioners was not that market power was to be eliminated; it was enough to ensure that it did not lead to ‘abnormal profits.’^{Footnote 4} Connected with this view was the idea that British firms were, on the whole, too small, that this was the main reason for their lack of competitiveness, and that combinations and amalgamations could therefore be a solution. During the post-1920 crisis, and especially after the return to gold in 1925, when the difficulties of British exports became even more acute, these views generated a vast movement of opinion in favor of ‘rationalization.’

The term “rationalization” was generally employed to refer to a reduction of the productive capacity of a whole industry by means of the elimination of the less efficient firms and the concentration of production in larger and more efficient ones. In order to obtain this result, strong support was given to horizontal and vertical integration of industries, and schemes of combination and amalgamation were adopted. The rationale for this choice was found in the economies of scale that larger units could obtain in production as well as in marketing.

This meaning of the term, however, was just one of a much wider range that Robert A. Brady (Reference Brady1932) cataloged in his analysis of the concept of rationalization. It was associated with the idea that it was possible to reduce average costs per unit of output by means of larger firms without necessarily incurring the problem of monopoly power—an idea that enjoyed wide circulation in the British press at that time. Henry Clay (Reference Clay1929, p. 171) wrote that rationalization “implies industrial combination with the object of securing not monopoly prices, but certain productive economies.” Writing about the situation in Lancashire in 1926, John Maynard Keynes sympathized with “what the Germans are calling ‘rationalisation’, that is, the concentration of demand on the most efficient plants, which are worked at full stretch and the rest closed down” (1971–89, XIX, p. 579; hereinafter CWK); “a ‘rationalising’ process designed to cut down overhead costs by the amalgamation, grouping or elimination of mills” (CWK, XIX, p. 584). Keynes, and Clay as well, were of the opinion that this would not happen spontaneously. More generally, this was the Liberal Party view on the subject as put forth in the Yellow Book:

Behind these proposals and analyses lay the ever increasing conviction that the post-war crisis was not a passing phase, and that it reflected an irreversible change in British overseas trade, which entailed a reduction of the extent of the market for the staples industries like cotton, coal, iron and steel, and shipbuilding. The Balfour Committee had expressed this view in its 1925 Report (Balfour Committee, vol. I), and the same point was made by Keynes in 1926 (CWK, XIX, p. 579).

This idea of rationalization as inevitable, since it was connected with irreversible processes, was not undisputed. Even among supporters of rationalization there were economists who thought that the crisis might be temporary—Macgregor (Reference Macgregor1927) expressed this view, for example (see also Macgregor et al. Reference Macgregor and Ryan1930). On the other hand, there was a substantial consensus that whatever else was the case, free competition was not the solution. This consensus was in large part the legacy of the pre-war combination movement and of the war economy. Several publications, many of which employed the reports of the Commission on Trusts and Profiteering as their empirical basis, recorded, and on the whole celebrated, the ever increasing scope of combination,^{Footnote 5} and Walter Meakin (Reference Meakin1928) went even further, claiming that rationalization was nothing less than The New Industrial Revolution. In this context and especially for those who, like Plant, were more loyal to the idea that markets are the most efficient allocators of resources, a certain “fixation” with the problem of monopoly was not out of place.^{Footnote 6} But, at least in the case of Britain in the interwar period, this did not result, as Coase would suggest, in a lack of attention to the study of the organization of industry.^{Footnote 7} Quite the contrary; the unconditioned emphasis that rationalizers were placing on the absolute need for larger firms stimulated a reaction, on the part of Plant and a group of Cambridge industrial economists, in which the evils of monopoly were momentarily left aside and in which the main focus was put, in a very Marshallian fashion, on the managerial costs of organizations.

III. ARNOLD PLANT AND THE CAMBRIDGE ECONOMISTS ON RATIONALIZATION

In a nation increasingly skeptical as to the possibility of its industries returning to a competitive status, some economists remained loyal to a conception of society as based on the automatic functioning of the market economy. Against Clay (Reference Clay1929), Edwin Cannan (Reference Cannan1930) argued that precisely because the crisis of British overseas trade was irreversible, any attempt to save the old staples industries was a waste of time and resources, and that the obstinacy of the ‘rationalizers’ was delaying the spontaneous development of new industries in new areas.^{Footnote 8} For Plant, a former student of Cannan's, despite all the efforts that had been made to keep rationalization and the search for monopoly power as separate as possible, rationalization was no more than another step in “the transition from the conception of intervention to prevent monopoly to that of intervention to create monopoly” (Plant Reference Plant1932a, p. 58).

In his inaugural lecture as holder of the newly created Chair of Commerce at the LSE, Plant argued against any attempt to overrule the spontaneous adaptation of industry to the new conditions. He took issue with “the half-conscious attempt to displace the controlling power which ultimately determines the nature and volume of all production, that is, the demand of ultimate consumers, and replace it by some other ill-defined criterion,” and he added that “[s]ome of the manifestations of planning and of rationalization are of this nature” (1932a, p. 46). Plant quoted the passage from Arthur Salter (Reference Salter1921) that Coase (Reference Coase1937, p. 587) would make so famous, in which it is stated that “[t]he normal economic system works itself. For its current operation it is under no central control” (Plant Reference Plant1932a, p. 51). And he lamented the impact of the war on the economic opinion of the country: “we have carried over from war a desire to plan and to rationalize, not in the sense of perfecting the responsiveness of our economic system to ubiquitous change, but with the effect of frustrating the spontaneous experimentation which perfect responsiveness implies” (Plant Reference Plant1932a, p. 57). In this perspective, the absolute priority was to preserve the market economy from the hypertrophic combinations of the period, rather than to explain how firms could exist in the ideal world depicted by Salter, as Coase (Reference Coase1937) would subsequently do.

Compared to the radically market-oriented views of Cannan and Plant, the position of the Cambridge economists was more complex. From a strictly theoretical point of view, their Marshallian upbringing gave them no reason to ask how firms could exist in a market economy and not the other way around.^{Footnote 9} Their main preoccupation was rather with the fact that the crisis of the textile industry constituted a particularly harsh blow to Marshall’s vision, in which Lancashire was identified as providing a vivid representation of ‘automatic’ cooperation within a localized industry. Now, outside Cambridge, Marshall’s external economies were sometimes employed as an explanation for the protracted delay in responding to the new conditions.^{Footnote 10} Moreover, together with coal mining, the cotton-spinning industry had become one of the main fields of rationalization. Keynes’s opinion, as expressed in a series of articles published between 1926 and 1929 in The Nation and Athenaeum and written in support of the Cotton Yarn Association (a cartel) and the Lancashire Cotton Corporation (the amalgamation eventually created by the Bank of England), has been summarized as a “microeconomic case against laissez-faire” (Marchionatti Reference Marchionatti1995) and presented as an example of an “anti-Marshallian” approach to industrial policy (Belussi and Caldari Reference Belussi, Caldari, Raffaelli, Nishizawa and Cook2011, pp. 154–155).

Cambridge views on rationalization were, however, not generally based on the idea that bigger is necessarily better. The point of view expressed by the Cambridge economists was rather that, in the specific circumstances of the ongoing crisis, larger units would certainly reduce costs in Lancashire and several other producing areas. However, their main preoccupation was not so much to support the claim for bigger units as to explain what was not working in the spontaneous mechanism that, according to Marshall’s time period analysis, should have regulated the transition from short- to long-period adaptations.

A readaptation of Marshall’s time-period analysis that could explain this unpredicted outcome was proposed by Richard Kahn in his fellowship dissertation in 1929.^{Footnote 11} Kahn’s argument was that, assuming conditions of imperfect competition, the individual expectations of non-coordinated entrepreneurs hampered firms in their attempt to expand their size in order to exploit scale economies. For Kahn, the obstacle in the way of the growth of firms’ average size was a strategic situation that could today be assimilated to the prisoner’s dilemma. While each firm should reduce prices in order to attract new customers and then expand output, this cannot be done without risk unless the reaction of other firms can be predicted. Especially when the number of firms within the same industry is very high, as in the case of the Lancashire cotton industry during the 1920s, the most rational decision was, according to Kahn, to keep each firm's output fixed at a constant level. A similar argument was also proposed by Shove (Reference Shove1930), who spoke of unexploited economies “of rationalization” and “concentration,” and then by Robinson in The Structure of Competitive Industry (Robinson [1931] Reference Robinson1935). Like Shove, Robinson abandoned Marshall’s life-cycle hypothesis and proposed a new solution to Marshall’s reconciliation (or Antoine-Augustin Cournot’s) problem of explaining the coexistence of unexploited economies and competitive conditions. Likewise in Macgregor (Reference Macgregor1906) and also Robertson (Reference Robertson1928), Robinson’s definition of competition was well within the Marshallian ground and therefore intrinsically imperfect, as it assumed the existence of a special but contestable market for each firm. And in the same fashion as Shove (Reference Shove1930), the theory proposed by Robinson is couched in terms of limited exploitation of internal and concentration economies, with market imperfection creating a cost in transferring customers from one firm to another (Robinson Reference Robinson1935, pp. 120–122). Along with the “five forces” (technical, managerial, and financial forces; the influence of marketing; risk and fluctuations) that determine a firm’s efficiency, this cost creates a costly and discontinuous pattern along which firms cannot always grow to the “optimum size.”^{Footnote 12}

These works of Kahn, Shove, and Robinson shifted attention from rationalization as a panacea to a readaptation of Marshall’s time-period analysis. Nevertheless, in Robinson’s book, and in other works more focused on industrial organization, such as Lavington (Reference Lavington1927) and Robertson ([1923] Reference Robertson1928), the Cambridge industrialists can be seen being at pains to keep alive the Marshallian idea that there is no one-way trend in industrial history, and that the ‘bigger is better’ banner is no more than an anti-scientific and ideologically driven misconception. Thus, even though Plant and the Marshallians were in opposite camps on industrial policy, with regard to the analysis of industrial organization they shared a common view that occasioned a small but nonetheless visible reaction to the idea that efficiency is always positively related to the size of business.

All these economists agreed that the problem was much more complicated than was generally assumed. Even more importantly, while they did not entertain a common opinion on rationalization, they based their arguments on a common analytical approach, for they shared a basic theory of organization and its costs. This theory, in which the costs of organization contribute to determining the size of firms, had an immediate antecedent in Marshall’s works, specifically in Industry and Trade, and roots in Adam Smith’s account of the division of labor. But this shared theory was more than mere repetition of old ideas because the increasing size of British firms was posing a new problem: provided that the increasing costs of organization counteracted scale economies on the side of production, thus putting a limit on the efficient size of business, it was far from clear how far this upper limit could be raised by the new methods of management. With this general issue in view, the Marshallians at Cambridge were working on a better analysis of coordination costs. Plant developed a similar approach, but, at least on one occasion, he went even further by introducing the idea that larger business dimensions could sometimes be the outcome of ill-conceived market regulations that increased the cost of using the market.

IV. THE THEORY OF THE BOUNDARIES OF THE FIRM IN ENGLAND BEFORE COASE (Reference Coase1937)

In relation to general opinion, the Economic Journal and Economica, Cambridge and the LSE, were the main centers of a more analytical, less ideological approach to the practical issues connected with rationalization, as well as to the theoretical problem of determining the size and boundaries of firms. For instance, “Problems of Rationalisation” (Macgregor et al. Reference Macgregor and Ryan1930) is the report of discussions held at the LSE, on the occasion of the Annual Meeting of the Royal Economic Society, with Arthur Cecil Pigou in the chair and Macgregor, John Ryan (the managing director of the Lancashire Cotton Corporation), Keynes, and Philip Sargant Florence among the speakers. Plant sometimes took advantage of his book reviews, both in Economica and the Economic Journal, to adopt a more detached and reasoned approach to proposals connected with rationalization (e.g., Plant Reference Plant1933, Reference Plant1934, Reference Plant1935). And it was in Economica that Lavington, certainly one of the most loyal Marshallians among the Cambridge men, published “Technical Influences on Vertical Integration. Based on Dr. Marshall’s Industry and Trade,” which is the natural starting point for the present reconstruction of the various theories of the size of firms before 1937.

In order to accord with the general idea that it was both possible and desirable to increase firms’ average dimensions, and also that it was possible to thereby gain the advantage of efficiency without incurring any cost in terms of monopoly power, Lavington started by “setting aside the desire of monopoly” (Lavington Reference Lavington1927, p. 27). On this premise, Lavington’s main contentions were that there was no “general ‘tendency to vertical integration,’” that “the representative economies of modern industrial methods work towards the dissociation of processes,” and that “the presence of such vertical integration as now exists is to be explained by the peculiar technique of one or two conspicuous industries and by the presence in other industries of certain technical conditions which partially neutralise the effect on the business unit of the main forces working for the vertical dissociation of processes” (1927, p. 27).

Lavington’s theory was an application of the principle of specialization. It consisted of the idea that costs of coordination grow with the number of processes performed by a firm and that management is more efficient when a larger volume of output is produced through a narrower range of processes. Accordingly, Lavington (Reference Lavington1927, pp. 29–30) distinguished between vertical and lateral integration—namely, the association of different processes at different stages or at the same stage of production, and horizontal combination, which is the association of like processes at the same stage—and postulated that business organization would spontaneously tend towards both vertical and lateral “dis-integration” and “horizontal combination.”

The assumption of this theory is that organization as a factor of production is the product of human faculties and that these faculties are inherently limited. Accordingly, the main economies of organization come from “concentrating human faculties on a narrower range of tasks,” and are associated (following Marshall Reference Marshall1919) with the application of “powerful appliances of production” to “uniform continuous process[es]” (Lavington Reference Lavington1927, p. 28).

Lateral and vertical specialization are limited only by the “effective range of marketing” (Lavington Reference Lavington1927, p. 29) and by a restricted group of “technical influences.” First, there may be technical relations between successive processes, which make it necessary for these stages to be carried out in close physical relation with one another, the iron and steel industries being the typical example. Second, a low degree of standardization of the products hampers the introduction of uniform continuous processes and the use of mechanization. Third (and again connected with a low degree of standardization), the adaptation of one stage of production to the scale of the next and the preceding ones in the chain sometimes makes it necessary for these stages of production to be integrated vertically within the same business unit.

Standardization is a key element in the theories of business size before Coase. For Lavington, as well as for Marshall, standardization permitted the specialization of firms on a narrower range of processes. For Lavington, standardization as opposed to individuality of product permitted the production of larger volumes of output through a narrower range of processes, thus reducing the average managerial costs per unit of product. In addition to this, for Marshall, standardization could also open up new avenues for very small firms, which could reduce their task to assembling standardized components in the manufacture of new products.^{Footnote 13} By replicating the analysis of the American case put forward in Industry and Trade, in which large and homogeneous markets permit a high level of product standardization and the adoption of large-scale production, Robertson provided an explanation of the limits to the growth of firms in different industries that is couched in terms of the degree of standardization, which itself is said to be “conditional on the growth of communications and the widening of markets” (1928, p. 20). The main advantage of large-scale production is the reduction of costs that can be obtained only by a “large firm”—one that can introduce “more highly specialized machinery, and keep it occupied more continuously than a small” industry (Robertson Reference Robertson1928, p. 21). For Robertson, however, the “economies of large-scale government rather than of large-scale technique ... dictate the size of the modern business unit” (1928, p. 25). This latter statement by Robertson introduced a further and more controversial element: the question of innovation in management and its impact on coordination costs. In itself, Robertson’s statement was potentially innovative because it indicated at least the possibility that larger managerial structures could be more efficient than privately owned firms. In true Marshallian fashion, however, Robertson was not saying that there was an absolute advantage on the side of big business. Rather, he explained, “the battle between the large firm and the small is not one which is ever fought to a definite finish,” and furthermore, the advantages that could be reaped from large-scale technique and management varied from one industry to another.

A similar view was adopted in his more refined work by Robinson ([1931] Reference Robinson1935). A very critical transition in the passage from one local optimum to a better one in Robinson’s theory is the “departure from individual management and a local market, in the direction of organised and coordinated management, and a national market” ([1931] 1935, p. 122). Like Robertson, Robinson was a rather conservative Marshallian on this point, and though he admitted that management had been the major field of innovation in British industry over the last decade, he nevertheless remained loyal to a faith in the “flexibility” and “energy” of small businesses, and of the primacy of individuals over organization. Hence, while he saw no upper limit to the growth of technical efficiency, Robinson affirmed that the managerial optimum sets “not only a lower but also an upper limit to the scale of operation” (Robinson [1931] Reference Robinson1935, p. 53).

This latter position of Robinson collided with that developed in Sargant Florence’s The Logic of Industrial Organization (1933). As in Lavington’s scheme, Florence argued that maximum technical efficiency is reached when the highest output is placed under a single managerial unit; but Florence’s managerial unit was an “organization,” not an individual. Moreover, he placed no upper limit upon the optimum managerial unit: “In the assumption ... of the economic advantage in specialization of men and equipment, and of long-run conditions when factors of production can be adjusted and reorganized, there is in my view no limit to the increase in the physical return obtainable by larger-scale operation” (Florence Reference Florence1933, p. 24).

This disagreement between Robinson and Florence, who had graduated from Cambridge in 1914 and was now professor of commerce at Birmingham, resulted in a quarrel in the Economic Journal (see Robinson [Reference Robinson1934] and Florence [1934]). Though rather sterile, this argument was nonetheless revealing of the questions that bigger businesses, with larger managerial structures, were posing to economists.^{Footnote 14} The division of labor was now working heavily in management as well, and this was creating an increasing demand for business education at a high level. But how far business administration could go in reducing managerial costs was far from clear, and Plant, like Lavington and Robinson, remained loyal to the idea that “it is one thing to increase the speed and range of contacts within a firm, and quite another to evoke indefinitely a greater and greater volume of response from the (still only human) organism which endeavours to direct the business from the centre” (Plant Reference Plant and Plant1937, p. 15).

The intrinsic limits of human faculties in dealing with complex processes and structures were frequently taken as the starting point for any theory of business administration, with the result that the latest of such theories sometimes looked like mere repetition of Lavington (Reference Lavington1927). We will soon see that this is true of Plant (Reference Plant and Plant1937); indeed, the correspondence with Ronald Fowler reported in Coase (Reference Coase1988b, pp. 9–11) shows that even Coase in early 1932 had for some while been adopting the same approach.

On March 24, 1932, Coase wrote to his friend that he was considering the problem of the “bringing together under one control of different functions.” Just as with Lavington’s distinction between lateral and horizontal integration, “[w]hat is important is that different functions are in fact brought together under one control, what stage they are in being of little account.” And, probably stimulated by this analogy, Fowler in his reply wrote that “Lavington, I believe it was, wrote an article denying any general tendency towards vertical integration (I mean apart from attempts to obtain monopoly)”; but, apparently, Coase was not acquainted with this article. At this stage, Coase was not “so sure how far all economies are as a rule due to greater specialization.... I started with this attitude but found it was not very helpful since it is not clear really whether modern business is more or less specialized than it used to be.” Fowler’s comment on this point insisted on the arguments put forward by Lavington, though without mentioning them explicitly, and this time Coase responded “with enthusiasm”: “I think your remark about decreasing returns to managerial ability arising not because the firm is more specialized but because it is less hits the nail on the head. There may be technical advantages in increasing complexity but it is decreasing returns to managerial ability which seems to set the limit.” But, in the end, Coase abandoned this theory because, as he wrote (in 1988b, p. 11), it “was going nowhere.” He now inaugurated his own approach.

Albeit implicitly, in “Centralize or Decentralize,” Plant (Reference Plant and Plant1937)^{Footnote 15} adopted Lavington’s approach, combining it with an analysis of ‘make or buy’ decisions^{Footnote 16} that is not very far from the idea that these decisions depend on coordination costs within the firm as well as the cost of using the market. Plant’s theory is divided into two parts. The first reproduces (without quoting, maybe because the paper was written for a non-academic audience) Lavington’s (Reference Lavington1927) argument that the spontaneous trend of any industry will be a general movement towards specialization rather than integration. In the second part, processes of centralization are considered for those gains these may procure “which do not depend upon the acquisition of power to influence prices” (Plant Reference Plant and Plant1937, p. 19). Both parts of Plant’s paper are framed in terms of make or buy decisions. But while in the first part (as in Lavington Reference Lavington1927) the focus is exclusively on the variations of managerial costs, the second part moves from the observation that the cost of using the market can also vary according to different circumstances, thus reproducing a scheme that is closer to the one adopted by Coase (Reference Coase1937).

The general principle expressed in the first part is that the opportunities for decentralization will be least “where the success of the firm depends mainly upon the personal qualities of the central authority” (Plant Reference Plant and Plant1937, p. 9). The number of individual decisions that have to be taken by the central authority can be reduced only to the extent to which decisions can be standardized in the form of “standing orders”:

This is a replica of Lavington’s argument, in which the relative degree of homogeneity of demand determines the feasible “standardization of output” and, therefore, the volume of output that can be produced at the maximum coordination cost, which corresponds to “the working day of the head of a business” (Plant Reference Plant and Plant1937, p. 12). As usual, therefore, the fundamental trade-off is that between production (technical) costs and coordination costs: “As business grows ... the time approaches when the capacity of the co-operating specialists in the firm to serve its clients will be greater if the attempt at maintaining central ‘co-ordination’ of all of their activities is abandoned, and parts of the firm are disintegrated into more easily manageable units” (Plant Reference Plant and Plant1937, p. 15).

The first part of Plant’s argument was, to a large extent, expressed in terms of make or buy decisions, taken as the consequence of the growing coordination costs of an expanding business. Thus, for instance, “[i]t is not accidental that motor-car manufacturers in this country, whose business rapidly expands, generally tend at the same time to ‘buy in’ more and more parts, instead of continuing to make them for themselves” (Plant Reference Plant and Plant1937, p. 17). Another example, which Plant considered in more detail, is the trend of development of department stores in retail distribution (1937, pp. 15–18). Plant noticed that the degree of central control in department stores tends to be highest in the case of standardized merchandise, while store managers are allowed some degree of discretion when merchandise has to be adapted to a local demand. In the latter case, the contracts of the local stores with central management come very close to a “virtual leasing of the department from the firm.” “In such cases it is a very short step to the complete ‘disintegration’ of the risk by the actual leasing of departments as independent businesses” (1937, p. 16). In this case, make or buy decisions are considered within the Lavingtonian scheme, but in the second part of the paper, Plant presented an analysis that is closer to that of Coase (Reference Coase1937).

Taking “the second standpoint,” which was the one connected with “the advantages of centralization,” Plant presented a scheme of analysis for the integration of processes in which the notion of a cost in using the market was at least implicit:

The remaining part of Plant’s paper consists of exemplifications of the same general principle that can be found in Coase (Reference Coase1937): “Specialist accountant service is usually easily acquired in the open market from professional firms; but here again, there may be an economy to the larger firm or amalgamation which can offer full-time specialized employment to a professional accountant” (Plant Reference Plant and Plant1937, p. 20); “the market in human effort is not yet so well organized that the qualities of workers offering themselves for employment can be accurately and expeditiously tested except by specialists” (p. 22); “No doubt if there were an open market, the delivery function would frequently be better disintegrated to specialist haulage contractors, but the Rail and Road Traffic Act of 1933 has put an end to the open market in road goods transport services; and large firms which centralize their delivery service secure an economy of pooling which their smaller competitors cannot share” (p. 26).

The latter example was connected with Plant’s previous work on transport legislation. As Coase explained, “Plant was critical of the rationalization schemes that were at that time advocated for a number of industries in Britain, and he was particularly hostile to proposals for the coordination of the various means of transport, a subject on which he presented a paper to the Institute of Transport in 1931” (1988b, p. 7). The paper mentioned by Coase is probably related to Plant (Reference Plant1932b).^{Footnote 17} As Coase observed, at this stage “Plant argued that competition would provide all the coordination needed. Yet we had in economics a factor of production, management, whose function was to coordinate” (Coase Reference Coase1988c, p. 7). What has been shown is that Plant himself, towards the end of 1936 or the beginning of 1937, was at least beginning to reason in terms of make or buy decisions, and that he was also beginning to comprehend these decisions in terms of a balance of the costs of two ways of coordination, the price mechanism within the market and management within the firm; but it still remains true that Coase’s path-breaking article had no immediate significant impact on the literature. This might have been due to the fact that it was a good idea at the wrong time; or, maybe better, a good idea on the wrong side of the controversy on rationalization.

V. A GOOD IDEA AT THE WRONG TIME?

The concept of a cost in using the market, along with the idea that business size depends on the balance between this cost and coordination costs, is implicit but nonetheless rather clear in Plant’s argumentation. Moreover, there was no apparent obstacle in the way of direct transmission of ideas from Coase to Plant, because Coase was a former pupil of Plant's and a member of the same commerce studies group at the LSE.^{Footnote 18} But this notwithstanding, and even though the last step that was still to be taken in order to arrive at Coase’s theory was by no means large, only Coase himself actually took this final step, and this poses the question of why the reception of “The Nature of the Firm” did not take shape until very much later. One possible answer to this question, which can be proposed at the end of the present study, is that for more than one reason, Coase’s article appeared in a context that was not favorable to its immediate reception.

This is not a complete novelty in the literature. Klaes (Reference Klaes2000) has already proposed the idea that “The Nature of the Firm” aroused little interest until the term “transaction cost,” which was not introduced by Coase in 1937, took shape in other contexts, and more specifically as an economic explanation in terms of costs that could replace the metaphor of ‘friction’ that was so often employed in the economic literature, and especially in the literature on money and the money markets. In this perspective, Coase (Reference Coase1937) had no immediate impact because his proposal contained the answer to a question that had not been posed yet, while it acquired its outstanding prominence only as a result of later contributions and discussions, including the one generated by Coase (Reference Coase1960) himself. Another element, which is not altogether new, is related to the debate on socialist planning. Within the contemporary literature on these topics, Coase (Reference Coase1937) was at a disadvantage for at least two reasons. The first, which emerges rather clearly in Campbell and Klaes (Reference Campbell and Klaes2005) and Bylund (Reference Bylund2014), is that Coase was proposing a view that reconciled the price mechanism with planning, while all the other participants in the discussion were neatly divided in a polarized debate between socialist planners and free market ideologists. If, as Bylund contends, Coase’s view was inspired by the idea of resource heterogeneity, which is at the roots of the Austrian theory of capital, the same view was not Hayekian in deriving from this intuition the other idea that, precisely for this reason, the market is not always better than planning, as Bylund also observes in his paper. Furthermore, if “The Nature of the Firm” gave planning its due place within the economy, it nonetheless left to competition the task of determining “the volume of planning (and therefore of market allocation),” as emphasized in Campbell and Klaes (2005, p. 270). This was an idea that must have been fairly unpalatable for the socialists. What Coase actually had to offer was neither a new theory of socialism nor a new analytical support for the specific Hayekian free market standpoint. Admittedly, Coase’s new theory was couched in the terms of the planning vs. market debate—indeed, it was presented as an explanation for the existence of planning within a market economy—but it still remained a theory of the firm.

It is perhaps at this point that the present research can provide a new element in the picture that is progressively emerging in the literature. As Coase (Reference Coase1937) presented a new theory of the firm, his approach could certainly have been awarded greater attention in the debate on rationalization. As shown in this paper, it was within the debate on rationalization that a theory of the firm (and not just a mere description) had been taking shape during the 1920s and the early 1930s. This might have placed Coase in the position to enter the group of economists who were developing this theory, but there is no evidence, at least in his later recollections of the past, that Coase ever took part in this discussion. It is revealing, in this respect, that in 1932 Coase was only indirectly familiar with the argument put forth by Lavington (Reference Lavington1927), finding it in a letter from a friend (Fowler), and that although as early as 1932 he developed his own approach as an alternative to it, he did not mention Lavington’s contribution. Very plausibly, this was a further side effect of the discussion on socialism vs. competition. In the late 1930s, the latter topic had replaced rationalization as the central issue of debate, being obviously of much more interest for a young socialist who was pursuing his very personal route towards a more liberal political stance. Had Coase (Reference Coase1937) appeared some ten years earlier, a fair reception of it would certainly have been easier. But while this remains only a hypothesis, there was also a conceptual distance between Coase and the other theorists of the firm, because they were concerned with different problems. While Coase was trying to lead a rather abstract theory of prices into contact with reality, Plant, Lavington, and Robinson were involved in a debate on industrial policy. And while Coase ignored this specific topic, setting out to deal with (what was in his view) the more important problem of explaining why firms existed in a market economy, at least one of the others, Plant, lamented that a market economy as Coase interpreted it in his article of 1937—that is, an economic system completely based on the price mechanism—simply could not exist in the Britain of his time. In the light of the existing historical and institutional context, Plant and his Cambridge colleagues were at pains to show that ‘bigger’ is not necessarily ‘more efficient’ and, it should be borne in mind, with this specific aim in view the idea that using the market had a cost was scarcely of any help to them. The cost of using the organization, rather than the cost of using the market, is the point upon which Plant, Lavington, and Robinson were forced to insist. In the abstract, the concepts employed in the two theories were complementary rather than rival, and it is therefore unfortunate that the Marshallian approach, with its emphasis on organization as a factor of production, and the Coasean approach, with its emphasis on the costs of market transactions, parted company instead of joining their explanatory powers. Plant (Reference Plant and Plant1937) is a case in point. With its replica of Lavington (Reference Lavington1927) in its first part and its quasi-Coasean approach in its second, this work came very close to combining the two theories. But, unfortunately, this catholic approach was not developed further, and, given the surrounding circumstances, this was a predictable outcome. While it would be silly to pretend to have advanced ‘the’ explanation of the missed reception of “The Nature of the Firm,” it seems reasonable to affirm that this element, among the others, should be taken into account.