MITI and the Japanese miracle

by Chalmers Johnson

  ninka

  ) the cartel. The government could also change the terms or nullify the agreement. And it could force nonparticipants in the cartel agreement to abide by its terms if they did not do so voluntarily. As a result of an MCI-sponsored amendment to the law (introduced in the Diet during September 1932 and passed during 1933), the ministry obtained the powers to approve investments that would expand facilities by cartel members and to approve members' decisions to curtail production. Needless to say, all members of an industry were required to submit frequent reports on their investment plans and activities to the government. It is in this law that we find the origins of the government's licensing and approval authority and of the practice of "administrative guidance," which together became the heart of postwar industrial policy. To assuage skeptics in the Diet, the law had a five-year limitation written into it; on August 15, 1936, it was extended for another five years but it was superseded by the National General Mobilization Law before the second time period expired.

  As a result of the law cartels were organized in some 26 designated "important industries," including silk thread, rayon, paper, cement, wheat flour, iron and steel, and coal. In industries suffering from excess capacity, such as cotton spinning, shipbuilding, and electrical machinery, it helped curtail competition and restore profitability. But Yoshino believes that the law did not actually work as he had intended it to. This was partly because only a month after it came into effect, the army seized all of Manchuria and the entire economy shifted to a war-preparatory footing. Even without this unforeseen development, however, the cartels seemed to work less in favor of "order" than in favor of strengthening and expanding the scope of zaibatsu operations (the

  keiretsuka

  of the economy, as the Japanese put it, or the structuring of the economy into conglomerates). In three industries, for examplepetroleum, newsprint, and cementMCI cartels promoted the interests of gasoline refiners over those of taxi companies, of the paper trusts over the newspapers, and of the cement industry over the lumber industry. All of the favored industries were areas of expanding zaibatsu strength.

  48

  Page 111

  The zaibatsu had made it clear in the TIRB's Control Committee that they were less interested in cartels on the model of the industrial unions for medium and small enterprises than they were in easing competition through mergers and in the reduction of the number of competitors. Many mergers followed in the wake of the law: in May 1933 three companies merged to form Oji * Paper, in December the Sanwa Bank came into being as a result of a three-way merger, in January 1934 Yawata and five private companies united to become Japan Steel, in June Mitsubishi Heavy Industries appeared, and the following year Sumitomo Metals was created. Each of these cases fostered concentrations of economic power that came close to being monopolies rather than cartels. The zaibatsu believed that cartels did not allow for change and would prove unworkable over time. They would have preferred an individual industrial development law for each industry (for example, the Petroleum Industry Law of 1934 and the Automobile Manufacturing Law of 1936), that would also protect them from international competitors. As it turned out, this kind of law became more important than the original control law as the thirties progressed, but for reasons the zaibatsu did not foresee; the militarists wanted them, and even though they did not like working with the zaibatsu, they had no other choice.

  At the time the control law was deliberated and enacted, Yoshino clearly stood for self-imposed control in industry and for government cooperation with private control agreements. However, the year 1931 ended up giving self-control a bad name, particularly insofar as it involved the zaibatsu, and this led to powerful demands for its opposite, namely, state control (

  kokka

  tosei

  *). The reason was the "dollar-buying" scandal. On September 21, 1931, England announced that it was leaving the gold standard because of the depression. This should have signaled the failure of Inoue's policy; Japan could not maintain the gold standard if its main trading competitor did not also abide by it. However, the Japanese gold embargo was not reimposed until three months later, and in the interim the zaibatsu banks engaged in an orgy of dollar buying, using yen they knew were soon to be devalued. Mitsui alone is said to have made $50 million in international currency transactions. The speculation ended in December with the return to power of the Seiyukai* (the last political party regime until after 1945) and Takahashi's reversal of Inoue's policy.

  The effect on Japanese public opinion of the zaibatsu's dollar buying during the height of the depression was one of outrage. Many groups concluded that the zaibatsu were so irredeemably avaricious they were quite prepared to debase their own country's currency in the

  Page 112

  name of profits. Right-wingers assassinated Dan Takuma, the chief executive officer of Mitsui, and he was succeeded by the Harvard-educated Ikeda Seihin, who became MCI minister in 1938. After taking over at Mitsui, Ikeda carried out a public

  tenko

  * (conversion to patriotism) on behalf of his company. One important reason Ikeda took this action was that the military and some members of the economic bureaucracy were beginning to think in terms not of working with the zaibatsu in cartels but of dominating themperhaps even nationalizing themthrough the exercise of state power.

  On December 21, 1931, as part of the Seiyukai's* housecleaning after taking over from the Minseito*, Yoshino Shinji was named vice-minister. He held the post until October 7, 1936. Although the TIRB had been associated with the Minseito party, the Seiyukai* retained it because of the positive appeal of its ideology and because Yoshino, who was thought to have Seiyukai sympathies even though he was publicly neutral, was now in charge of the ministry.

  Within trade and industry circles the years 1931 to 1936 came to be known as the era of the Yoshino-Kishi line. This meant government promotion of heavy and chemical industrialization and a stress on industrial rationalization as the main objective of MCI policy. During his vice-ministership, Yoshino first promoted Kishi to the post of chief of the Industrial Policy Section (January 1932, the month after Yoshino took office), then chief of the Documents Section (December 1933), and finally chief of the Industrial Affairs Bureau (April 1935). Kishi was clearly on the "elite course," and he was expected to advance to the vice-ministership shortly after his mentor gave it up. Kishi did eventually become vice-minister, but a three-year Manchurian interlude intervened first. For although there was an orientation within the ministry that could be called a Yoshino-Kishi line, there were also differences between the two men. If Yoshino will always be identified with industrial "self-control," Kishi will always be identified with ''state control" of industry.

  The first phase of modern Japanese industrial policy seems remote from the postwar economic miracle, but it is, in fact, directly relevant to it for several reasons. During the 1920's Japan faced economic problems comparable in kind and in severity to those of the early 1950's: the need to restore competitive ability in international trade, the need to reorganize industry in order to achieve economies of scale and to take advantage of new technological developments, and the need to increase the productivity of the labor force. During the period from

  Page 113

  the creation of MCI to the passage of the Important Industries Control Law in 1931, the Japanese invented and experimented with the first of their three characteristic approaches to industrial policy, approaches that have remained in their repertoire to the present day. The first approach was the attempt to replace competition with self-control of an industry by the enterprises already established in it. The institutional form of this approach, state-licensed cartels, remains big business's preferred form of industrial policy down to the present day. Its major weakness, the tendency of cartelization to lead to zaibatsu domination and monopoly, was already fully visible by 1931; and this weakness in turn elicited demands for the opposite of self-control, namely, state control, that dominated the rest of the 1
930's.

  Another theme of this early period was the search for criteria of managerial and enterprise performance other than short-term profitability. Via the industrial rationalization movement, which began in the late 1920's but reached its full flowering only during the 1950's, the Japanese began self-consciously to think about how to build into enterprises and whole industries incentives to promote labor peace, job security, capital formation, increased productivity, and the development of new products. Although the earliest efforts at rationalization were largely frustrated by zaibatsu power and interests, a concern with rationalizationthe attempt to gain a competitive edge through superior organization, labor peace, and cost cuttingis the most consistent and continuous feature of Japanese industrial policy throughout the Showa * era. The greatest achievement of the early days of MCI was to begin seriously to forge a government-business relationship that was oriented to cooperation and development and that took the position of the whole Japanese economy vis-à-vis competitive foreign economies as its primary frame of reference.

  The ideas and institutional innovations of this early period are not merely some "heritage" that had to be transmitted from one generation to the next. The generation that was to lead industrial policy during the 1950's and 1960's was already on the scene during the late 1920's and early 1930's. One of the most startling facts about the history of Japanese industrial policy is that the managers of the postwar economic "miracle" were the same people who inaugurated industrial policy in the late 1920's and administered it during the 1930's and 1940's. Unlike other nations defeated in World War II or torn by revolution in the wake of World War II, Japan did not experience a radical discontinuity in its civilian bureaucratic and economic elites. Men such as Yoshino Shinji, Kishi Nobusuke, Shiina Etsusaburo*, Uemura

  Page 114

  Kogoro *, and Inayama Yoshihiro were active in the formulation and execution of industrial policy before, during, and after the war. Equally important, all of MITI's vice-ministers during the 1950's entered the bureaucracy between 1929 and 1934. Thus, in studying the early origins of industrial policy, we are also studying the formative years of the officials who applied it with such seemingly miraculous effect during the 1950's. Not surprisingly, the institutions and policies first discussed in the Temporary Industrial Rationality Bureau bear more than a passing resemblance to the institutions and policies of the later period of high-speed growth.

  This theme of historical continuity also draws attention to the fact that industrial policy is rooted in Japanese political rationality and conscious institutional innovation, and not primarily or exclusively in Japanese culture, vestiges of feudalism, insularity, frugality, the primacy of the social group over the individual, or any other special characteristic of Japanese society.

  Economic crisis gave birth to industrial policy. The long recession following World War I, capped by the panic of 1927, led to the creation of MCI and to the first attempts at industrial policy, just as the need for economic recovery from World War II, capped by the deflation panic of 1949, led to the creation of MITI and to the renewal of industrial policy. All of the political and bureaucratic problems of the developmental stateincluding conflict between the bureaucracy and the central political authorities, and conflict among elements of the bureaucracy itselfappeared in this early period, just as they would reappear during the 1960's and 1970's. That the Japanese solved (or suppressed) these problems more effectively during the postwar period than during the 1930's is greater testimony to their ability to profit from experience than to any fundamental change in the situation they faced.

  During the late 1920's the Japanese began to build new forms of state intervention in the economy, forms that differ in critical ways from those of either the command economy or the regulatory state. These initial efforts were soon overwhelmed by recurring crisesand contained unforeseen consequences that dismayed their inventors. As a result, the leaders of industrial policy were led to attempt a different approach, direct state control of the economy, that carried them to disaster. The bitterness of the era of the Yoshino-Kishi line was more than enough to warn those who managed both the state and private enterprise after the war that catastrophes could occur if they did not transcend both self-control and state control in favor of gen-

  Page 115

  uine public-private cooperation. Nonetheless, it should not be thought that these painful early experiences were wholly negative. The prototype of industrial policy did not fly well, and the improved model crashed, but the suitably modified production version of the 1950's amazed the world by its performance. From this perspective the early years of industrial policy were a period of indispensable gestation in the evolution and perfection of a genuine Japanese institutional invention, the industrial policy of the developmental state.

  Page 116

  Four

  Economic General Staff

  During the period that Yoshino and his colleagues were discovering rationalization and inching their way toward industrial policy, a different group of Japanese officials was taking up parallel questions. They were military officers and civilian bureaucrats working in military or cabinet-level agencies. Their concerns centered on Japan's preparedness for war, particularly since Japan had not experienced the total mobilization during 191418 that had so deeply influenced the general staffs of the Europeans. They were also concerned about the severe economic constraints that had appeared during Japan's last major war (with Russia during 19045), the economic mobilization and growing industrial strength of the country's potential enemies (particularly Soviet Russia), and the emergence of problems in their planning for national security as it concerned primary resources (particularly petroleum, but including other materials necessary for modern armaments).

  These men were coming to believe that Japan needed an industrial policy in order to ensure its military survival, not just to overcome the depression. At an absolute minimum they wanted an "economic general staff" (

  keizai

  sanbo

  *

  honbu

  ) that would provide guidance for the economy from the point of view of Japan's military needs and industrial and resource deficiencies. During the 1930's this stream of inspiration for industrial policy flowed into and merged with the civilian MCI stream, and both were transformed in the process.

  The Japanese military's first thoughts about mobilizing the whole private economy for war came during World War I. On April 17, 1918, based on its understanding of Germany's mobilization effort and on the actions of the United States after entering the war, the govern-

  Page 117

  ment of Prime Minister (General) Terauchi enacted the Munitions Industries Mobilization Law (Gunju Kogyo * Doin* Ho*). It was Japan's first basic law relating to industrial control during wartime. It defined military supplies broadly and authorized the government after a declaration of war to supervise, use, or expropriate the industries producing them. Most of its provisions were never enforced during World War I, but it was still on the books in 1937, and it was implemented during the early stages of the "China Incident" (the Diet asked for a distinction between a "state of war" and a "state of incident") before the National General Mobilization Law of 1938 replaced it.

  1

  The 1918 law was virtually an afterthought of Japan's participation in World War I. But in order to prepare for the possible need to implement the law, the government on May 31, 1918, set up a Munitions Bureau (Gunju Kyoku) as a semidetached unit of the cabinet to prepare economic mobilization plans and to gather statistics on munitions industries. Its first chief, Hara Shoichiro* of the navy, worked hard at these tasks, but he found it almost impossible to get cooperation from the established ministries. On May 15, 1920, the government sought to lower the visibility of the bureau by merging it with the cabinet's Statistical Bureau to create a new agency called the Census Board (Kokusei-in). This idea did not work any better than the first onethe military officers and statisticians squabbled over turfand on N
ovember 30, 1922, with the military somewhat in disgrace because of the Siberian expedition and with the government trying to cut costs, the Census Board was abolished. The government transferred all the mobilization plans and accumulated statistics to MAC and from it to MCI, where they greatly enhanced the resources of the Secretariat's Statistical Section. Murase Naokai, the vice-minister of MCI from 1936 to 1939 and a very important figure in our later discussions in this chapter, was working in the cabinet at the time the Census Board was abolished. He says that he recognized that these mobilization materials would be useful to his ministry in administering industrial policy, and he implies that he had a hand in having them transferred there.

  2

  During the mid-1920'sthe period of "Taisho* democracy"the military was forced to drop its efforts to plan for economic mobilization, but by 1927 interest had revived. Many military officers had had a chance to study and absorb the lessons of World War I, and they were concerned about the growing economic might of Russia after the consolidation of the Bolshevik revolution. During 1927 General (then Major) Ishiwara Kanji, the chief economic architect of Manchukuo, wrote, "If national mobilization was taken to mean that the Japa-