MITI and the Japanese miracle

by Chalmers Johnson

  The chief mechanisms of the cooperative relationship are selective access to governmental or government-guaranteed financing, targeted tax breaks, government-supervised investment coordination in order to keep all participants profitable, the equitable allocation by the state of burdens during times of adversity (something the private cartel finds it very hard to do), governmental assistance in the commercialization and sale of products, and governmental assistance when an industry as a whole begins to decline.

  This form of the government-business relationship is not peculiarly or uniquely Japanese; the Japanese have merely worked harder at perfecting it and have employed it in more sectors than other capitalist nations. The so-called military-industrial complex in the United States, to the extent that it identifies an economic relationship and is not merely a political epithet, refers to the same thing. If one were to extend the kinds of relationships that exist between the U.S. Department of Defense and such corporations as Boeing, Lockheed, North American Rockwell, and General Dynamics to other sectors of industry, and if one were also to give the government the power to choose the strategic sectors and to decide when they were to be phased out, then one would have a close American approximation of the postwar Japanese system. The relationship between government and business in the American national defense industriesincluding the unusual management and ownership arrangements for the nuclear weapons

  Page 312

  laboratories and the existence of such official agencies as the former Atomic Energy Commission and the National Aeronautics and Space Administrationis thought by Americans to be exceptional, whereas it was the norm for Japan's leading industrial sectors during high-speed growth. It is also perhaps significant that aviation, space vehicles, and atomic energy are all sectors in which the United States is preeminent, just as Japan is preeminent in steel production, ship-building, consumer electronics, rail transportation, synthetic fibers, watches, and cameras.

  As noted earlier, the cooperative government-business relationship in the capitalist developmental state is very difficult to achieve and maintain. Even with such deeply entrenched social supports for cooperation as a shared outlook among government and industrial leaders because of common education (for instance, at Todai * Law) and an extensive cross-penetration of elites because of early retirement from government service and reemployment in big business, the Japanese have difficulty in keeping public-private cooperation on the tracks. Industry is quite willing to receive governmental assistance, but it does not like government orders (as the steel and automobile industries illustrate). Government is often frustrated by the excessive competition and preemptive investment of industries it is trying to foster (as the petrochemical and textile industries illustrate). Nonetheless, the Japanese have worked hard to create cooperative relationships and have developed numerous unusual institutions through which to pursue them. These include the official "deliberation councils" such as the Commerce and Industry Deliberation Council of 1927, the Cabinet Advisers Council of 1943, the Industrial Rationalization Council of 1949, and the Industrial Structure Council of 1964; MITI's vertical bureaus and the corresponding officially sanctioned trade associations for each industry; the temporary exchange of officials between the state and private enterprise (for example, the posting of young MITI officers to Keidanren headquarters); the formal "discussion groups" implemented in the wake of the failure of the Special Measures Law; and the practice of administrative guidance, in which government officials and representatives of banking and industry can coordinate their activities unconstrained by law and lawyers.

  In addition, the Japanese have fostered social supports for cooperation. We have already mentioned two of themthe essentially bureaucratic education of both public and private managers and the extensive "old boy" networks. It should not be thought that these are the only social supports or that they are not duplicable in other societies. They would be very hard to duplicate in other societies, since

  Page 313

  they rest on long-entrenched practices, but they are not pure cultural givens. As this book has sought to show, there was more consensus and cooperation in Japan during the 1950's than during the 1930's, which suggests that the reasons for this difference are to be found in changed historical circumstances and political consciousness and not in something as relatively unchanging as cultural mores. Some other social supports for government-business cooperation include the virtual impotence of corporate stockholders because of the industrial financing system; a work force fragmented among labor aristocrats enjoying semilifetime employment, temporaries, small-scale subcontractors, and enterprise unions; a system of collecting private savings through the postal system, concentrating it in government accounts, and investing it in accordance with a separate, bureaucratically controlled budget (FILP); some 115 government corporations covering such high-risk areas as petroleum exploration, atomic power development, the phasing out of the mining industry, and computer software development (these corporations are the successors to the national policy companies of the 1930's, the eidan of the wartime era, and the kodan * of the occupation); and a distribution system that serves not only to retail goods but also to keep the unemployed, the elderly, and the infirm working, thereby weakening demands for a welfare state in Japan.

  In Japan, as compared with the United States, one of the most powerful social supports for private managers' cooperation with the government is that Japanese managers enjoy freedom from being judged exclusively in terms of short-term financial performance. Just as the essential spirit of Japanese industrial policy after the late 1920's lay in the search for ways to replace competition with cooperation without a drastic loss in efficiency, the industrial rationalization campaigns sought criteria of good management other than short-term profitability. These included the maintenance of full employment, increased productivity, expansion of market share, cost reduction, and the management of long-term innovation.

  Morita Akio, chairman of Sony Corporation, believes that the emphasis on profitability has been a major cause of American industrial decline. He asserts, "The annual bonus some American executives receive depends on annual profit, and the executive who knows his firm's production facilities should be modernized is not likely to make a decision to invest in new equipment if his own income and managerial ability are judged based only on annual profit."

  1

  Morita believes that the incentive structure of postwar Japanese business has been geared to developmental goals, whereas the incentive structure

  Page 314

  of American business is geared to individual performance as revealed by quick profits. The result is not simply a lack of long-term planning in the United States but also exorbitant executive salaries, private corporate aircraft, palatial homes, and other major discrepancies between the rewards of labor and management. In postwar Japan the living standards of top executives and ordinary factory workers have differed only slightly (Morita observes that the American president of Sony's U.S. subsidiary makes more in corporate salary than Morita himself receives from Sony). On the other hand, it might be noted that managers in Japan have access to corporate entertainment funds of a size unparalleled in any other economy. The National Tax Agency calculated that during 1979 corporate social expenses amounted to some ¥2.9 trillion, or $13.8 billion, which meant that corporate executives were spending $38 million per day on drink, meals, golf fees, and gifts for their colleagues and customers.

  2

  The point is that Japan's more flexible means of evaluating managers contributes to smoother labor-management relations than in some other countries and also avoids disincentives to cooperate with other enterprises and with the government. These Japanese practices came into being as a result of postwar conditions. According to Morita, "There is nothing in Japanese history to suggest that smooth labor-management relations came naturally"; prewar Japanese capitalism was "stark in its exploitation of labor." The postwar leveling of all Japanese incomes because of inflation a
nd national adversity made possible the relative equality of rewards that existed during high-speed growth, as well as the emphasis on measures other than profitability for managerial performance. These social conditions are of considerable advantage to Japan in competing with countries such as the United States, but obviously they would be very difficult to transplant: although the salaries of American managers might be reduced, the institution of some other measure of performance than short-term profitability would require a revolution in the American system of allocating savings to industry through stock markets.

  The priorities and social supports for cooperation among the Japanese might not be replicable in other societies, but it is easy to imagine that they might be matchedthat is, a different society might be able to manipulate its own social arrangements in ways comparable to those of postwar Japan in order to give top priority to economic development and to provide incentives for public-private cooperation. If this were the case, then such a society would need an abstract model of the Japanese high-growth system to use as a guide for its own concrete application. Specialists on modern Japan will differ as to the pre-

  Page 315

  cise elements and the weight to be attached to each element in such a model, but the following, based on the history of MITI, is my own estimation of the essential features of the Japanese developmental state. For purposes of this discussion, I stipulate that Japan's particular history would not have to be reexperienced, and that the social inputs of popular mobilization and the incentives to cooperate already exist in the society trying to emulate Japan (assumptions that are not necessarily realistic, as this study has sought to demonstrate).

  The first element of the model is the existence of a small, inexpensive, but elite bureaucracy staffed by the best managerial talent available in the system. The quality of this bureaucracy should be measured not so much by the salaries it can command as by its excellence as demonstrated academically and competitively, preferably in the best schools of public policy and management. Part of the bureaucracy should be recruited from among engineers and technicians because of the nature of the tasks it is to perform, but the majority should be generalists in the formulation and implementation of public policy. They should be educated in law and economics, but it would be preferable if they were not professional lawyers or economists, since as a general rule professionals make poor organization men. The term that best describes what we are looking for here is not professionals, civil servants, or experts, but managers. They should be rotated frequently throughout the economic service and retire early, no later than age 55.

  The duties of this bureaucracy would be, first, to identify and choose the industries to be developed (industrial structure policy); second, to identify and choose the best means of rapidly developing the chosen industries (industrial rationalization policy); and third, to supervise competition in the designated strategic sectors in order to guarantee their economic health and effectiveness. These duties would be performed using market-conforming methods of state intervention (see below).

  The second element of the model is a political system in which the bureaucracy is given sufficient scope to take initiative and operate effectively. This means, concretely, that the legislative and judicial branches of government must be restricted to "safety valve" functions. These two branches of government must stand ready to intervene in the work of the bureaucracy and to restrain it when it has gone too far (which it undoubtedly will do on various occasions), but their more important overall function is to fend off the numerous interest groups in the society, which if catered to would distort the priorities of the developmental state. In the case of interests that cannot

  Page 316

  be ignored, deflected, or satisfied in symbolic waysor upon which the perpetuation of the political system dependsthe political leaders must compel the bureaucracy to serve and manipulate them.

  A non-Japanese example of the kind of relationship we are looking for would be something like the American legislative branch's relationship to the wartime Manhattan Project or to the postwar nuclear submarine development program. The political system of the developmental state covertly separates reigning and ruling: the politicians reign and the bureaucrats rule. But it must be understood that the bureaucrats cannot rule effectively if the reigning politicians fail to perform their positive tasks, above all, to create space for bureaucratic initiative unconstrained by political power.

  There are several consequences of this type of political system. One is that groups without access to the system will on occasion take to the streets to call attention to their disaffection (this occurred in Japan in 1960 in the antisecurity treaty riots, in the student revolts of the late 1960's, in the demonstrations against the new Tokyo airport and the government's nuclear ship project, and in the campaign against industrial pollution). These demonstrations may arise out of important interests that cannot be indefinitely ignored by the state, or they may simply reflect demands for political participation. Whatever the case, when they occur the political leaders are called upon to exercise "safety valve" functions, forcing the bureaucracy to alter priorities just enough to calm the protesters but taking most of the "heat" of the demonstrations themselves. Clever politicians will anticipate eruptions of this sort (Sato * Eisaku's strategy for the renewal of the security treaty in 1970). As long as the developmental projects are succeeding and their benefits are being equitably distributed, the political leaders should be able to deal with these problems symptomatically. Projects to call attention to the development effort and to instill pride in its successes may also be recommended (the Japanese Olympics of 1964, EXPO 70).

  A major political difference between the capitalist developmental state and the communist dictatorship of development is that the capitalist state simply ignores the nonstrategic sectors of the society, whereas the communist state attempts directly and forcibly to demobilize them. The first is preferable because it avoids the unintended consequences of the presence of large numbers of police and the full apparatus of repression, which is not only wasteful of resources but is also incompatible with effective international commerce. This is certainly one lesson the Japanese learned from the 1940's.

  The Japanese political system should also be distinguished from the

  Page 317

  bureaucratic-authoritarian regimes of Argentina, Brazil, Chile, and Uruguay. In these states the ruling elites seek to promote industrialization by excluding from power the previously mobilized economic groups and by developing collaborative relationships with multinational corporations. They do this through a technocratic political arrangement that relies heavily on coercion to enforce the rules of the game.

  3

  Japan differs in that it is a democracy in which the politicians are chosen by the votes of the majority; its stability has rested on the ability of the ruling political party to forge a coalition of voters committed to economic growth and effective management. During high-speed growth this coalition reflected the widespread recognition of Japan's situational imperatives; during the early 1970's the coalition began to weaken markedly, but it appears to have been reinvigorated by the energy crisis and the acceptance of the need for a new shift of industrial structure. Until very recent times the Japanese also have not been hospitable to collaboration with foreign capital. The Korean developmental state, by contrast, seems to share some of the bureaucratic-authoritarian characteristics and should to that extent be distinguished from the postwar Japanese case.

  Postwar Japan also differs from the market-socialist states (chiefly Yugoslavia and Hungary), where various experiments have been undertaken to try to synthesize market economics with political control. The contradictory tensions inherent in these systems are more like those of Japan during the period of attempted state control (the late 1930's and 1940's) than they are like Japan after it had regained its independence and launched high-speed growth based on public-private cooperation.

  4

  In addition to occ
asional protest demonstrations, a probable further consequence of the capitalist developmental system will be the periodic occurrence of "corruption" scandals. These arise because of the separation between reigning and ruling and because of the opportunities this condition gives some insiders to exploit the development programs. As long as these scandals occur primarily among politicians and not among bureaucrats, and as long as the development effort is proceeding to the benefit of the society as a whole, these scandals will be tolerated as unfortunate but not too serious imperfections in the overall system. However, if they occur among the bureaucracy, they signal the need for quick surgery and reconstitution of the system.

  The third element of the model is the perfection of market-conforming methods of state intervention in the economy. In implementing its industrial policy, the state must take care to preserve competi-

  Page 318

  tion to as high a degree as is compatible with its priorities. This is necessary to avoid the deadening hand of state control and the inevitable inefficiency, loss of incentives, corruption, and bureaucratism that it generates. It is probable that the market-conforming methods that actually work cannot be discovered a priori but will have to emerge from conflict between the managers of the state and the managers of the privately owned strategic industries. A cooperative relationship between the state and private enterprise is not a natural one: the state inevitably will go too far, and private enterprise inevitably will resent state interference in its decisions. When either the state or private enterprise becomes clearly dominant over the other, as occurred in Japan during the late 1940's (state dominance) and during the early 1970's (private-enterprise dominance), development will falter. One clear lesson from the Japanese case is that the state needs the market and private enterprise needs the state; once both sides recognized this, cooperation was possible and high-speed growth occurred.