As discussed in earlier chapters, the inhospitality tradition regarded nonstandard contracting practices as presumptively unlawful. Nonstandard internal forms of organization have also been regarded with deep suspicion. The same technological orientation to economic organization plainly informed both of those approaches. Unless a clear-cut technological justification for the con- tracting practices or organizational structure in question could be discerned, antitrust specialists were quick to ascribe antisocial purpose and effect. Trans- action cost economics regards nonstandard forms of market and internal orga- nization differently. For one thing, anticompetitive concerns ought to be reserved for the subset of conditions for which a condition of preexisting monopoly power exists. For another, the possibility that economies of transaction cost are realized ought to be admitted. Rather, therefore, than regard organizational innovations with suspicion and hostility, such innovations are assessed on the merits instead. Real economies of all kinds, transaction cost included, warrant respect.
1. The Conglomerate
The conglomerate form of organization has been subject to a variety of in- terpretations. Some of them are sketched here, after which a transaction cost interpretation is advanced. The matter of tradeoffs is then briefly addressed.
1.1. EARLIER INTERPRETATIONS
The antitrust enforcement agencies were among the first to venture an unfavorable assessment of the conglomerate. Thus the staff of the Federal Trade Commission held:
With the economic power which it secures through its operation in many diverse fields, the giant conglomerate may attain an almost impregnable position. Threatened with competition in any one of its various activities, it may well sell below cost in that field, offsetting its losses through profits made in its other lines—a practice which is frequently explained as one of meeting competition. The conglomerate corporation is thus in a position to strike out with great force against smaller business. [U.S. Federal Trade Commission, 1948, p. 59]
Robert Solo subsequently characterized the conglomerate corporation as a “truly dangerous phenomenon” and argued that it “will probably subvert management effectiveness and organizational rationale for generations” (1972 pp 47-48). Others advised that the large conglomerate was a hazard to competition “in every line of commerce in every section of the country” (Blake, 1973, p- 567). Bogeyman economics became fashionable. Procter & Gamble, for example, was repeatedly described as a “brooding omnipresence” in a court of law.103 Even those who regarded the conglomerate form more sympathetically referred to it as a puzzle (Posner, 1972, p. 204).
To be sure, Morris Adelman (1961) advanced a more favorable in- terpretation. He observed that the conglomerate form of organization had attractive portfolio diversification properties. But why should the conglomerate appear in the 1960s rather than much earlier? After all, holding companies, which long predated the conglomerate, can accomplish portfolio diversification. And individual stockholders, through mutual funds and otherwise, are able to diversify their own portfolios. At best the portfolio diversification thesis is a very incomplete explanation for the postwar wave of conglomerate mergers.
1.2. AN INTERNAL CAPITAL MARKET INTERPRETATION
As set out previously, Alfred P. Sloan, Jr., and his associates at General Motors were among the first to perceive the merits of the M-form structure. But while the divisionalization concept was well understood and carefully implemented within General Motors, those same executives were fixated on the notion that General Motors was an automobile company.
Thus Sloan remarked that “tetraethyl lead was clearly a misfit for GM. It was a chemical product, rather than a mechanical one. And it had to go to market as part of the gasoline and thus required a gasoline distribution system.”105 Accordingly, although GM retained an investment position, the Ethyl Corporation became a free-standing entity rather than an operating division (Sloan, 1964, p. 224): Similarly, although Durant had acquired Frigidaire and Frigidaire’s market share of refrigerators exceeded 50 percen the 1920s, the position was allowed to deteriorate as rivals developed market positions in other major appliances (radios, ranges, washers, etc.) while Frigidaire concentrated on refrigerators. The suggestion that GM get into air conditioners “did not register on US, and the proposal was not adopted” (Sloan, 1964, p. 361). As Richard Burton and Arthur Kuhn conclude, GM’s “deep and myopic involvement in the automobile sector of the economy [prevented) product diversification opportunities in other market areas—even in product lines where GM had already achieved substantial penetration—[from being] recognized” (1979, pp. 10-11).
The conglomerate form of organization, whereby the corporation con- sciously took on a diversified character and nurtured its various parts, evidently required a conceptual break in the mind-set of Sloan and other prewar business leaders. That occurred gradually, more by evolution than by grand design (Sobel, 1974, p. 377), and it involved a new group of organizational innovators—of which Royal Little was one (Sobel, 1974). The natural growth of conglomerates, which would occur as the techniques for managing diverse assets were refined, was accelerated as antitrust enforcement against horizontal and vertical mergers became progressively more severe. Conglomerate acquisitions—in terms of numbers, assets acquired, and as a proportion of total acquisitions—grew rapidly with the result that “pure” conglomerate mergers, which in the period 1948-53 constituted only 3 percent of the assets acquired by merger, had grown to 49 percent by 1973-77 (Scherer, 1980, p. 124).
As developed more fully elsewhere (Williamson, 1975, pp. 158-162), the conglomerate is best understood as a logical outgrowth of the M-form mode for organizing complex economic affairs. Thus once the merits of the M-form structure for managing separable, albeit related, lines of business (e.g. a series of automobile or a series of chemical divisions) were recognized and digested, its extension to manage less closely related activities was natural. That is not to say that the management of product variety is without problems of its own. But the basic M-form logic, whereby strategic and operating decisions are distinguished and responsibilities are separated, carried over. The conglomerates in which M- form principles of organization are respected are usefully thought of as internal capital markets whereby cash flows from diverse sources are concentrated and directed to high-yield uses.
The conglomerate is noteworthy, however, not merely because it permitted the M-form structure to take that diversification step. Equally interesting are the unanticipated systems consequences that developed as a byproduct. Thus once it was clear that the corporation could manage diverse assets in an effective way, the possibility of takeover by tender offer suggested itself. The issues here are developed in Chapter 12.
The term “M-form” is reserved for those divisionalized firms in which the general office is engaged in periodic auditing and decision review and is actively involved in the internal resource allocation process. Cash flows, therefore, are subject to an internal investment competition rather than automatically reinvested at their source. The affirmative assessment of the conglomerate as a miniature capital market presumes that the firm is operated in such a way. Not all conglomerates were. In particular, firms that in the 1960s were referred to as “go-go” conglomerates did not respect M-form principles. Their merits, if they had any, presumably resided elsewhere.
Inasmuch, however, as the organizational logic of the M-form structure runs very deep—serving, as it does, both to economize on bounded rationality (the information processing interpretation) and safeguard the internal resource allocation process against the hazards of opportunism (which is what the general office concept adds), the rationale for conglomerate structures in which M-form principles are violated is gravely suspect. Indeed, one would expect, and events have borne the expectation out, that the ‘‘go-go” conglomerates would become unglued when adversity set in—as it did in the late 1960s. Those firms found it necessary to reorganize along M-form lines, to simplify their product lines, or to do both.
Note in that connection that the M-form conglomerate engages in a depth- for-breadth tradeoff. As Alchian and Demsetz put it: “Efficient production into heterogeneous resources is not a result of having better resources but in knowing more accurately the relative productive performance of those resources” (1972, p.789; emphasis in original). Plainly, diversification can be .taken to excess. As the capacity to engage knowledgeably in internal resource allocation becomes strained, problems of misallocation and opportunism intrude. That conglomerate firms voluntarily engage in divestiture is presumably explained by that condition.
Lest I be misunderstood, I do not mean to sugges that opportunities to express managerial preferences in ways that conflict with the preferences of the stockholders have been extinguished as a result of the conglomerate form. The continuing tension between management and stockholder interests is reflected in numerous efforts that incumbent managements have taken to protect target firms against takeover (Cary, 1969; Williamson, 1979; Benston, 1980). Changes in internal organization have nevertheless relieved managerial discretion concerns. A study of the economic institutions of capitalism that makes no allowance for organization form changes and their capital market ramifications will naturally overlook the possibility that the corporate control dilemma posed by Berle and.’Means has since been alleviated more by internal than it has by regulatory or external organizational reforms.
3. Multinational Enterprise
The discussion of the multinational enterprise (MNE) that follows deals mainly with recent developments and, among them, emphasizes organizational aspects—particularly those associated with technology transfer in manufacturing industries. As Mira Wilkins has reported, direct foreign investment, expressed as a percentage of GNP, was in the range of 7 to 8 percent in 1914, 1929, and 1970 (Wilkins, 1974, p. 437). Both the character of this investment and, relatedly, the organization structure within which this investment took place were changing, however. It is not accidental that the term MNE was coined neither in 1914 nor in 1929 but is of much more recent origin.
Thus whereas the ratio of the book value of U.S. foreign investments in manufacturing as compared with all other (petroleum, trade, mining, public utilities) was 0.47 in 1950, that had increased to 0.71 in 1970 (Wilkins, 1974, p. 329). Also, “what impressed Europeans about American plants in Europe and the United States [in 1929] was mass production, standardization,.and scientific management; in the 1960s, Europeans were remarking that America’s superiority was based on technological and managerial advantage [and] that this expertise was being exported via direct investment” (Wilkins, 1974, p. 436).
The spread of the multinational corporation in the post-World War II period has given rise to considerable scrutiny, some puzzlement, and even some alarm (Tsurumi, 1977, p. 74). One of the reasons for this unsettled state of affairs is that transaction cost economizing and organization form issues have been relatively neglected in efforts to assess MNE activity.12
Organization form is relevant in two related respects. First is the matter of U.S.-based as against foreign-based investment rates. Yoshi Tsurumi reports in this connection that the rate of foreign direct investments by U.S. firms increased rapidly after 1953, peaked in the mid-1960s, and has leveled off and declined since (Tsurumi, 1977, p. 97). The pattern of foreign direct investments by foreign firms, by contrast, has lagged that of the United States by about a decade (pp. 91-92).
Recall that the conglomerate uses the M-form structure to extend asset management from specialized to diversified lines of commerce. The MNE counterpart is the use of the M-form structure to extend asset management from a domestic base to include foreign operations. Thus the domestic M- form strategy for decomposing complex business structures into semi- autonomous operating units was subsequently applied to the management of foreign subsidiaries. The transformation of the corporation along M-form lines came earlier in the United States than in Europe and elsewhere. U.S. corporations were for that reason better qualified to engage in foreign direct investments at an earlier date than were foreign-based firms. Only as the latter took on the M-form structure did that multinational management capability appear. The pattern of foreign direct investments recorded by Tsurumi and reported above is consistent with the temporal differences of U.S. and foreign firms in adopting the M-form structure.
That U.S. corporations possessed an M-form capability earlier than their foreign counterparts does not, however, establish that they used it to organize foreign investment. John Stopford and Louis Wells have studied that issue. They report that while initial foreign investments were usually organized as autonomous subsidiaries, divisional status within an M-form structure invariably appeared as the size and complexity of foreign operations increased (Stopford and Wells, 1972, p. 21). The transformation usually followed the oiganization of domestic operations along M-form lines (p. 24). The adoption of a “global” strategy or “worldwide perspective”—whereby “strategic planning and major policy decisions” are made in the central office of the enterprise (p. 25)—could be accomplished only within a multidivisional framework.
Even more interesting than those organization form issues is the fact that foreign direct investments by U.S. firms have been roncentrated in a few industries. Manufacturing industries that have made substantial foreign direct investments include chemicals, drugs, automobiles, food processing, electronics, electrical and nonelectrical machinery, nonferrous metals, and rubber. Tobacco, textiles and apparel, furniture, printing, glass, steel, and aircraft have, by comparison, done little foreign direct investment (Tsurumi, 1977, p. 87).
Stephen Hymer’s “dual” explanation for -the multinational enterprise is of interest in this connection. Thus Hymer observes that direct foreign investment “allows business firms to transfer capital, technology, and organizational skill from one country to another. It is also an instrument for restraining competition between firms of different nations” (Hymer, 1970, p. 443).
Hymer is surely correct that the MNE can service both of those purposes, and examples of both kinds can doubtless be found. It is nevertheless useful to ask whether the overall character of MNE investment, in terms of its distribution among industries, is more consistent with the efficiency purposes to which Hymer refers (transfer of capital, technology, and organizational skill) or with the oligopolistic restraint hypothesis. Adopting a transaction cost orientation discloses that the observed pattern of investment is more consistent with the efficiency part of Hymer’s dual explanation.
For one thing, oligopolistic purposes can presumably be realized by portfolio investment coupled with a limited degree of management involvement to segregate markets. Put differently, direct foreign investment and the organization of foreign subsidiaries within an M-form structure are not needed to effect competitive restraints. Furthermore, if competitive restraints were mainly responsible for those investments, then presumably all concentrated industries— which would include tobacco, glass, and steel—rather than those associated with rapid technical progress, would be active in MNE creation. Finally, although many of the leading U.S. firms that engaged in foreign direct investment enjoyed “market power,” that was by no means true for all.
By contrast, the pattern of foreign direct investments reported by Tsurumi appears to be consistent with a transaction cost economizing interpretation. Raymond Vernon’s 1971 study of the Fortune 500 corporations disclosed that 187 of them had a substantial multinational presence. R&D expenditures as a percentage of sales were higher among those 187 than among the remaining firms in the Fortune 500 group. Furthermore, according to Vernon, firms that went multinational tended to be technological innovators at the time of making their initial foreign direct investments.
That raises the question of the attributes of firms and markets for accom- plishing technology transfer. The difficulties with transferring technology across a market interface are of three kinds: recognition, disclosure, and team organization (Arrow, 1962; Williamson, 1975, pp. 31-33, 203-7; Teece, T977).13 Of those three, recognition is probably the least severe. To be sure, foreign firms may sometimes fail to perceive the opportunities to apply technological developments originated elsewhere. But enterprising domestic firms that have made the advance can be expected to identify at least some of the potential applications abroad.
Suppose, therefore, that recognition problems áre set aside and consider disclosure. Attempts to transfer technology by contract can break down because of the “paradox of information.” A very severe information asymmetry problem exists, on which account the less informed party (in this instance the buyer) must be wary of opportunistic representations by the seller.14 Although sometimes the asymmetry can be overcome by sufficient ex ante disclosure (and veracity checks thereon), that may shift rather than solve the difficulty. The fundamental paradox of information is that “its value for the purchaser is –mot known.until he has the information, but then he has in effect acquired it ..without cost” (Arrow, 1971,,p. 152).
Suppose, arguendo, that buyers concede value and are prepared to pay for information in the seller’s possession. The incentive to trade is then clear, and for some items this will suffice. The formula for a chemical compound or the blueprints for a special device may be all that is needed to effect the transfer. Frquently, however, and probably often, new knowledge is diffusely distributed and is poorly defined (Nelson, 1981). Where the requisite information is distributed among a number of individuals all of whom understand their speciality in only a tacit, intuitive way15 a simple contract to transfer the technology cannot be devised.
Transfer need not cease, however, because simple contracts are not feasible. If the benefits of technology transfer are sufficiently great, exchange may be accomplished either by devising a complex trade or through direct foreign investment. Which will be employed depends on the circumstances. If only a one-time (or very occasional) transfer of techology is contemplated, direct foreign investment is a somewhat extreme response.16 The complex contractual alternative is to negotiate a tie-in sale whereby the technology and associated knowhow are transferred as a package. Since the knowhow is concentrated in the human assets who are already familiar with the technology, this entails the creation of a “consulting team” by the seller to accompany the physical technology transfer—the object being to overcome startup difficulties and to familiarize the employees of the foreign firm, through teaching and demonstration, with the idiosyncrasies of the operation.
Inasmuch as many of the contingencies that arise in the execution of such contracts will be unforseen, and as it will be too costly to work out appropriate ex ante responses for others, such consulting contracts are subject to consider- able strain. Where a succession of transfers is contemplated, which is to say when the frequency shifts from occasional to recurring, complex contracting is apt to give way to direct foreign investment. A more harmonious and efficient exchange relation—better disclosure, easier reconciliation of differences, more complete crosscultural adaptation, more effective team organization and reconfiguration—predictably results from the substitution of an internal governance relation for bilateral trading under those recurrent trading circumstances where assets, of which complex technology transfer is an ex- ample, have a highly, specific character.
The upshot is that while puzzlement with and concerns over MNEs will surely continue, a transaction cost interpretation of the phenomenon sheds insight on the following conspicuous features of multinational investment: (1) the reported concentration of foreign direct investment in manufacturing in- dustries where technology transfer is of special importance; (2) the organization of those investments within M-form structures; and (3) the differential timing of foreign direct investment between U.S. and foreign manufacturing enterprises (which difference also has organization form origins).
Source: Williamson Oliver E. (1998), The Economic Institutions of Capitalism, Free Press; Illustrated edition.