The evolution of the japanese developmental state institutions locked in by ideas

2 Industrial associations Gyo-kai dantai 4III Alternative arguments 6 1 The collective action theory 6 2 A power-based rationalist theory 8 VII Existing literature on the Japanese econom

in the context of war continued to influence the direction of the Japaneseeconomy over the following decades.

Analysing the institutional origin and evolutionary path of developmentalstate system, The Evolution of the Japanese Developmental State extendsexisting scholarship on the institutions that were at the heart of the develop-mental state system by focusing on not just why they were important, but alsohow and why they were originally built Based on extensive archival research

in both Japan and the USA, including Japanese-language collections notwidely known in the West, it will be of great interest to students and scholars

of political science, economic history, economics and Asian studies

Hironori Sasada is Associate Professor at Ritsumeikan University in Kyoto,Japan

1 The Police in Occupation Japan

Control, corruption and

Tai Yong Tan and Gyanesh Kudaisya

4 The Australia-Japan Political

6 The Triads as Business

Yiu Kong Chu

7 Contemporary Taiwanese Cultural

Nationalism

A-chin Hsiau

8 Religion and Nationalism in India

The case of the Punjab

Hans J van de Ven

11 Hong Kong in Transition

One country, two systems

Edited by Robert Ash, Peter Ferdinand,

Brian Hook and Robin Porter

12 Japan’s Postwar Economic Recoveryand Anglo-Japanese Relations,1948–62

Noriko Yokoi

13 Japanese Army Stragglers andMemories of the War in Japan,

1950–75Beatrice Trefalt

14 Ending the Vietnam WarThe Vietnamese communists’perspective

Ang Cheng Guan

15 The Development of the JapaneseNursing Profession

Adopting and adapting Westerninfluences

Aya Takahashi

16 Women’s Suffrage in AsiaGender nationalism and democracyLouise Edwards and Mina Roces

17 The Anglo-Japanese Alliance,

1902–22Phillips Payson O’Brien

18 The United States and Cambodia,

1870–1969From curiosity to confrontationKenton Clymer

19 Capitalist Restructuring and thePacific Rim

Ravi Arvind Palat

20 The United States and Cambodia,

1969–2000

A troubled relationshipKenton Clymer

21 British Business in Post-ColonialMalaysia, 1957–70

‘Neo-colonialism’ or

‘disengagement’?

Nicholas J White

Thai Absolutism

Kullada Kesboonchoo Mead

23 Russian Views of Japan, 1792–1913

An anthology of travel writing

David N Wells

24 The Internment of Western Civilians

under the Japanese, 1941–45

26 Nationalism in Southeast Asia

If the people are with us

28 A Colonial Economy in Crisis

Burma’s rice cultivators and the

world depression of the 1930s

33 Colonial Armies in Southeast Asia

Edited by Karl Hack and

Tobias Rettig

World MarketA.J.H Latham andHeita Kawakatsu

35 Japanese–German Relations,

1895–1945War, diplomacy and public opinionEdited by Christian W Spang andRolf-Harald Wippich

36 Britain’s Imperial Cornerstone

in ChinaThe Chinese maritime customsservice, 1854–1949

38 Japanese-American Civilian PrisonerExchanges and Detention Camps,1941–45

41 Christian Heretics in LateImperial China

Christian inculturation and statecontrol, 1720–1850

Edited by Rotem Kowner

44 Business–Government Relations inPrewar Japan

Peter von Staden

People, princes and colonialism

Edited by Waltraud Ernst and

47 The Quest for Gentility in China

Negotiations beyond gender and class

Edited by Daria Berg and Chloë Starr

48 Forgotten Captives in Japanese

Occupied Asia

Edited by Kevin Blackburn and

Karl Hack

49 Japanese Diplomacy in the 1950s

From isolation to integration

Edited by Iokibe Makoto, Caroline

Rose, Tomaru Junko and John Weste

50 The Limits of British Colonial

Control in South Asia

Spaces of disorder in the Indian

Ocean region

Edited by Ashwini Tambe and

Harald Fischer-Tiné

51 On The Borders of State Power

Frontiers in the greater Mekong

R.B Smith (Edited by Beryl Williams)

54 Port Cities in Asia and Europe

Edited by Arndt Graf and

Chua Beng Huat

55 Moscow and the Emergence of

Communist Power in China, 1925–30

The Nanchang Rising and the birth

of the Red Army

Bruce A Elleman

in Southeast AsiaThe Maria Hertogh controversy andits aftermath

Syed Muhd Khairudin Aljunied

57 Japanese and Hong KongFilm Industries

Understanding the origins of EastAsianfilm networks

Edited by Antony Best

63 Journalism and Politics in Indonesia

A critical biography of MochtarLubis (1922–2004) as editorand author

David T Hill

64 Atrocity and American MilitaryJustice in Southeast AsiaTrial by army

Louise Barnett

65 The Japanese Occupation of Borneo,

1941–1945Ooi Keat Gin

66 National Pasts in Europe andEast Asia

P.W Preston

A Journey to the West

Hsiao-ting Lin

68 New Perspectives on the History and

Historiography of Southeast Asia

Edited by Liping Bu, Darwin

Stapleton and Ka-che Yip

From enemies to alliesPeter Berton

73 Reforming Public Health inOccupied Japan, 1945–52Alien prescriptions?

Christopher Aldous andAkihito Suzuki

74 Trans-Colonial Modernities inSouth Asia

Edited by Michael S Dodson andBrian A Hatcher

75 The Evolution of the JapaneseDevelopmental State

Institutions locked in by ideasHironori Sasada

The Evolution of the Japanese Developmental State

Institutions locked in by ideas

Hironori Sasada

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

Simultaneously published in the USA and Canada

by Routledge

711 Third Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

© 2013 Hironori Sasada

The right of Hironori Sasada to be identi fied as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.

All rights reserved No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers.

Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identi fication and explanation without intent to infringe.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data

Sasada, Hironori.

The evolution of the Japanese developmental state : institutions locked in

by ideas / Hironori Sasada.

p cm – (Routledge studies in the modern history of Asia)

Includes bibliographical references and index.

1 Economic development –Japan–History–20th century 2.

Bureaucracy –Japan–History–20th century 3 Business and politics–Japan– History –20th century 4 Industrial policy–Japan–History–20th century 5 Japan –Economic policy–1945- 6 Japan–Politics and government–1945- I Title.

HC462.9.S2548 2012

338.952 –dc23

2011050174 ISBN: 978-0-415-50346-4 (hbk)

ISBN: 978-0-203-11437-7 (ebk)

Typeset in Times New Roman

by Taylor & Francis Books

2 Industrial associations (Gyo-kai dantai) 4

III Alternative arguments 6

1 The collective action theory 6

2 A power-based rationalist theory 8

VII Existing literature on the Japanese economy 18

VIII Outline of the book 20

I Industrialization in the Meiji period (1868–1912) 23

II Liberal market economy in the 1920s 25

III Zaibatsu dominance and party politicians 29

IV Conclusions 37

I Historical background 40

II Institutions of the Manchurian developmental state system 41

1 Pilot agency: the Planning Bureau 41

2 Special corporation (tokushu geisha) 45

III Explaining the establishment of the Manchurian developmentalstate system 47

IV Conclusions 75

4 The rise of a wartime economy in Japan (1937–45) 77

I Institutions of the Japanese wartime developmental state

system 78

1 Pilot agency: the Cabinet Planning Board 78

2 Control associations (to-sei kai) 83

II Explaining the rise of the wartime Japanese developmental

state system 85

III Conclusions 117

5 Institutional evolution in postwar Japan (1946–65) 120

I Core institutions of the postwar Japanese developmental

state system 121

1 Pilot agencies: the ESB and MITI 121

2 Industrial associations (gyo-kai dantai) 130

II Explaining the evolution of the postwar Japanese developmental

state system 134

III Conclusions 163

6 The evolution of the Japanese developmental state 165

I Evolution of developmental state systems in Manchuria, wartimeJapan, and postwar Japan 165

5.4 Positive feedback effect in the Japanese developmental

Tables

4.2 Production goals and actual results of important industrial

6.1 Pilot agencies of Manchuria, wartime Japan, and postwar Japan 166

I Introduction

In the latter half of the 20th century, most East Asian countries developedsuccessful economic systems that are, in many ways, different from those ofother industrial countries Many studies have analyzed the secrets behind the

“economic miracles” of those nations, and it is now widely believed that theirunique economic and political institutional settings, often referred to collec-tively as the “East Asian developmental state system”, were critical to theirrapid economic development.1 Such studies persuasively explain the causallink between economic growth and the economic system

However, their theoretical domain is limited because studies of politicaleconomy and economic development often overlook the issue of institutionalorigin and evolution, taking institutions as an exogenous factor and focusingprimarily on the causal link between institutional settings and economicperformance.2Waldner (1999: 5) explains the lack of attention paid to institu-tional origins in this type of study by saying, “Institutionalist scholars haveidentified neoliberal economists, who ignore the role of institutions, as theirprimary theoretical rivals; thus, their most pressing goal has been to demonstratethat institutions matter, a task that obviously precedes exploring the etiology

of institutional formation.” Just as neoliberal economists ignored the role ofinstitutions, studies into East Asian economies have neglected the issues ofinstitutional origin and evolution

An understanding of a country’s political economy and economic developmentcannot be complete if we cannot account for how its economic institutionswere built and later developed into their particular form Without suchknowledge, we do not know who built the institutions and for what purpose;what factors affected the institutions’ evolution; and whether the institution’scausal effects were intentional The Japanese economy went through a funda-mental transition from a liberal economy to a developmental state systemduring the wartime era, and the elements of the wartime economic systemremained in place as central features of the postwar Japanese economy,despite efforts by the American occupation forces to dismantle them afterWorld War II

In spite of the intense scholarly attention paid to the Japanese politicaleconomy, existing studies have not fully accounted for why the Japanesedevelopmental state system emerged during wartime and why it has persistedfor such a long period In this book, I attempt to unravel unsolved puzzlesrelating to the institutional origin and evolution of the Japanese develop-mental state system To accomplish this, I examine the process through whicheconomic institutions emerge and evolve from an historical institutionalistpoint of view,3 while focusing on the long-term impact of policy decisionsmade during a critical period of history on institutional evolution insubsequent periods.

The puzzles analyzed in this book are the following:first, what is the origin ofthe postwar Japanese developmental state system? Second, why did wartime andpostwar Japanese policymakers choose the developmental state system, asystem that combines characteristics of both planned and market-basedeconomies? Studies suggest that most of the core institutions governing thepostwar Japanese economy did not exist before the mid-1930s.4In the 1930sthe Japanese government adopted a new system of political economy that wassignificantly different from the economic systems of the previous periods (i.e theMeiji period and the 1920s)

In the postwar period, Japanese policymakers facing the urgent need for anew economic system had several options available These included Marxist,national-socialist, and liberal market-based economic systems Yet, the pol-icymakers repeatedly chose the developmental state system, rebuilding prewarinstitutions, despite strong pressure from American occupation forces toabolish them What explains the Japanese policymakers’ decision to introduce

a new system of political economy in the 1930s, and to rebuild that system inthe postwar period?

II Core institutions of the postwar Japanese economy5

In my analysis of the Japanese developmental state system, I focus on theevolution of two types of institutions, namely pilot agencies and industrialassociations These two institutions were of particular importance to theJapanese economy because they were critical components of the system ofstate control and played important roles in Japan’s rapid economic growthduring the postwar period Pilot agencies took charge of drawing blueprintsfor the nation’s economic growth and provided guidance to domestic industrybased on their plans (Johnson 1982, 1995; Anchordoguy 1988; Wade 1990;Gereffi and Wyman 1990; Woo 1991; Samuels 1994; Pekkanen 2003) Indus-trial associations assisted the pilot agencies in making and implementing theireconomic plans, and by promoting a cooperative relationship between thestate and industry (Samuels 1983; Lynn and McKeown 1988; Yonekura 1993;Tilton 1996; Schaede 2000) Thanks largely to these two institutions, Japan’sdevelopmental state system functioned effectively and achieved remarkablesuccess in the postwar period However, the creation and evolution of these

core institutions has not been widely studied Therefore, I will attempt tocontribute to the understanding of Japanese political economy by exploringthe institutional origin and evolution of these core institutions.6

1 Pilot agencies

The most distinguishing feature of the so-called East Asian developmentalstate system is the role of bureaucratic leadership in economic development.Because of this, the remarkable successes of industrial policies championed byvarious bureaucracies have been the focus of numerous scholars in thefield ofEast Asian political economy (Johnson 1982, 1995; Anchordoguy 1988; Okimoto1989; Wade 1990; Gereffi and Wyman 1990; Woo 1991; Samuels 1994; Evans1995; Waldner 1999; Pekkanen 2003) Even though the system generallyallows the nation’s market to perform its primary functions unhindered andguarantees private property rights, the bureaucracy does intervene in variouseconomic activities In other words, in a developmental state system, themarket system is fused with state control in a subtle balance This fusion ofmarket and state control differentiates the developmental state system fromboth Western free market and socialist economies in important ways

The postwar Japanese economy was based on the market system withgovernment guarantees of private property rights However, it differed from aliberal market-based economy in that state elites set economic goals designedspecifically to promote productivity and economic growth However, itremains distinct from a socialist economy because the government does notnationalize or manage private firms directly Instead, it compels them tooperate and compete within a framework of government regulations andindustrial policies Thus, the Japanese system combines features of both liberalmarket-based and socialist economies

In such systems, economic bureaucracies play a crucial role as “pilotagencies” that direct industrial development from strategic viewpoints Pilotagencies are distinguished from other bureaucratic bodies by their functions

in guiding industrialization and economic growth The defining functions ofpilot agencies include: composition of the long-term development plans for aspecific industry or the national economy as a whole; carefully calculatedallocation of scarce resources among industries; protection and nurturing ofinfant industries; manipulation of macroeconomic conditions in favor ofdomestic industries; and assistance in acquiring advanced technology fromabroad.7Many scholars acknowledge that such policies of economic bureaucratsmade the significant contributions to the economic success of East Asianeconomies, although they do not necessarily use the term “pilot agency”(Johnson 1982, 1995; Anchordoguy 1988; Okimoto 1989; Wade 1990; Gereffiand Wyman 1990; Woo 1991; Samuels 1994; Hatch and Yamamura 1996;Pekkanen 2003)

The Manchurian government under Japanese influence established such anagency, called the Planning Bureau, in 1932 The Planning Bureau drafted

“the Manchurian Industry Development Five-Year Plan,” and led the trialization of the newborn nation The wartime Japanese government institutedits own pilot agency, called the Planning Board, in 1937 In the postwarJapanese government, the Economic Stabilization Board (ESB) and the Ministry

indus-of International Trade and Industry (MITI) functioned as pilot agencies.8

This study focuses on the pilot agencies in Manchuria, wartime Japan, andpostwar Japan

Pilot agencies also emerged in other East Asian countries in the 1960s and1970s Some examples include South Korea’s Economic Planning Board andMinistry of Trade and Industry, and Taiwan’s Council on Economic Planningand Development and the Industrial Development Bureau of the Ministry ofEconomic Affairs Just like their Japanese counterparts, these Korean andTaiwanese pilot agencies took the initiative in guiding the industrialization ofthe newly developed countries (see Wade 1990; Gereffi and Wyman 1990;Woo 1991; Evans 1995 and Waldner 1999 for information on pilot agencies inSouth Korea and Taiwan)

2 Industrial associations (Gyo-kai dantai)

Industrial associations (a.k.a.“trade associations”) are cartel-like organizations

of privatefirms operating in the same industry In contemporary Japan, there

is one industrial association (gyo-kai dantai) in practically every industry, eventhough some industries are more tightly organized than others WartimeJapan and Manchuria had similar institutions, called control associations(to-sei kai), in a number of key industries, such as the steel, coal, automobile,railroad, and trade industries The obvious function of such associations is toserve as a political vehicle for memberfirms Much like pressure groups in theUnited States, they carry the voices of member firms to the policymakingprocess by lobbying the government or petitioning political parties

However, unlike their American counterparts, industrial associations inJapan have other functions Even though industrial associations are privateorganizations managed by their members, the government uses them as tools

of indirect control An industrial association coordinates interests amongmember firms and monitors their compliance with laws and governmentalguidance In some cases, when necessary, such associations even voluntarilyregulate their memberfirms In other words, they are partly an extension of agovernment agency within the private sector because they function as quasilaw-enforcement agencies These industrial associations allow the state tosuccessfully implement and enforce its industrial policies without creatingfriction with private actors

Industrial associations also support the government in making economicplans and industrial policies In order to draft effective economic plans andindustrial policies, the government needs to collect as much detailed andaccurate information regarding the industries as possible Industrial associa-tions assist government policymaking by providing information about their

member firms and by proposing policy recommendations In this sense,

“trade associations and cartels are critical intermediaries between the state(principally MITI) and the market (or individualfirms)” (Tilton 1996: 2)

To distinguish them clearly from American-style pressure groups, I define

an industrial association in a developmental state system as an organization

of firms in a given industry that first, lobbies the government for policiesreflecting the interests of member firms; second, helps the governmentenforce its administrative guidance by acting as a quasi-government agency;and third, assists the government in policymaking by providing valuableinformation regarding the industry In other words, while pressure groupsonly meet the first criterion (resulting in a one-way relationship with thegovernment), an industrial association meets all three criteria and is a vitalcomponent of a two-way relationship (Yonekura 1993)

The contribution of industrial associations to Japan’s rapid industrializationwas significant Lynn and McKeown (1988) argue that industrial associations(they use the term ‘trade associations’) made the Japanese economic systemmoreflexible in its ability to aggregate interests, and thus better able to work towardthe good of the society as a whole “The Japanese were able to subordinatemany other concerns effectively for rapid economic growth, particularly in the1950s and 1960s And the network of trade associations and peak associationsplayed a prominent role in policy making during this period” (Lynn andMcKeown 1988: 174)

Yonekura (1993) also claims that industrial associations “strengthenedthe effectiveness of government’s industrial policies by reducing the informa-tion asymmetry between the government and the industry” (Yonekura 1993:183) Furthermore, Samuels (1983) claims that the state’s control overindustry would have been impossible without the assistance of industrialassociations He argues,“While it is frequently acknowledged that committees

in the Diet [the Japanese legislature] and the ruling Liberal Democratic Party(LDP), having limited expertise and staff, are dependent upon the bureau-cracy for data and analysis, it is not as often recognized that the well-fabledJapanese bureaucracy is itself often dependent in the same way upon theindustrial associations and firms with which it works so closely” (Samuels1983: 499) In other words, while it is true that pilot agencies, such asMITI, played a critical role in leading Japan’s rapid economic growth, theycould not have performed their functions without the assistance of industrialassociations

Yet, despite the significance of industrial associations in Japan’s development,they have not been widely studied Lynn and McKeown (1988: xiii) contendthat“trade associations have not been the object of much scholarly research”

in the United States Also, “There is a relative abundance of governmentdocuments [in Japan] giving information on associations and their activities(most of which are not available in translation and have not yet been analyzed

by non-Japanese scholars) But even in Japan, there is a paucity of analyticwork on trade associations.” Tilton claims: “Debate whether the state or the

market guides Japan’s economy often overlooks the importance of tradeassociations” (Tilton 1996: 190; also see Schaede 2000: 2–3).

The origins of Japan’s industrial associations lie in the wartime period.9

In 1941 the wartime Japanese government established control associations(to-sei-kai) in six industries, including steel, coal, and automobiles, that wereconsidered strategically important In the following year, the governmentcreated control associations in 15 other industries From that point, thenumber of industries organized under control associations steadily increasedduring the war period The primary tasks of the control associations were toprovide data and information regarding the condition of their industries(production potential, raw material supply, etc.), allocate production quotasand raw materials among member firms based on government plans, and,when necessary, reorganize nonproductive firms The wartime pilot agenciesoutlined production quotas based on the information provided by the controlassociation and issued administrative guidance for the member firms tofollow

There is an unmistakable linkage between the wartime control associationand the postwar industrial associations (gyo-kai dantai) Even though theJapanese government was strongly pressured by the American occupationforces to dismantle such organizations, the government continued to allowindustries to reorganize themselves under similar groups While there weresome minor differences, the postwar industrial associations played a surprisinglysimilar role in assisting the government’s industrial policies

III Alternative arguments

How are such institutions built? What causes changes in institutional settings?What determines evolutionary path of institutions? Some scholarly groupstry to solve those puzzles by focusing on the interests of the primary actors,

an approach that is often referred to as applying rational choice theories.Those theories operate on the premise of individual rationality, whichassumes that actors are rational and behave in ways that maximize their uti-lity That is, when actors make decisions, they examine all the availablealternatives and choose those options that yield the most benefit Rational the-orists argue that institutions are built to advance the interests of the actors Inother words, an institution is built when it is more beneficial to build one than to

do without it Rationalists offer mainly two ways to address institutions Someposit that actors build institutions to achieve mutually beneficial outcomes,and others claim that powerful actors build institutions to maximize theirindividual gains

1 The collective action theory

The collective action theory posits that actors build institutions to achievemutually beneficial outcomes Many rationalists view politics in terms of

collective actions (Olson 1965; Bates 1981; Weingast 2002) Rationalists arguethat collective action is hard to organize due to problems such as the lack oftrust among actors, free riding, high transaction costs, and the absence of mon-itoring and enforcement mechanisms Since the individual utility maximizationbehaviors of actors make it hard for them to overcome the collective actionproblems, they tend to end up with suboptimal or inefficient outcomes.Rationalists view institutions as one of the solutions to collective action problemsbecause the rules and agreed procedures embedded in institutions decreaseuncertainties, provide enforcement mechanisms, reduce transaction costs,facilitate trade-offs, etc According to the rationalist view, this is the reasonwhy actors build institutions.

Weingast argues,“Appropriately configured institutions restructure incentives

so that individuals have an incentive to cooperate.” Therefore, he contends,

“the essence of institutions is to enforce mutually beneficial exchange andcooperation” (Weingast 2002: 670) Keohane (1984) argues that, in interna-tional politics, collective action problems often lead to mutually undesirableoutcomes (e.g market failure) States often fail to cooperate with each otherbecause of high transaction costs among them and due to the lack of inter-national laws Because of this, some states build institutions that facilitatecooperation by reducing transaction costs, promoting reciprocity in interac-tions among member states, and by establishing reliable monitoring andenforcement mechanisms.10

Rationalists apply this view of institutions to explain institutional continuity.They maintain that when all actors benefit from certain institutions that makecollective action possible, the institution becomes self-enforcing becausedefecting from the institutional arrangement could destroy the institution.Therefore, the institution develops a reproduction mechanism For example,Weingast analyzes democratic consolidation and explains why some demo-cratic institutions survive while others fail by focusing on the constitutionalarrangements that affect actor incentives He argues that the democracies withconstitutions that guarantee the fundamental rights of its citizens make it farless likely for citizens “to resort to extraconstitutional means to defendthemselves” (Weingast 2002: 682) Such a constitution becomes self-enforcing

by providing actors with incentives to abide by its provisions He claims:“Thepact becomes self-enforcing when all parties are better off under the pact andwhen all realize that unilateral defection from the pact implies that the otherswill also defect, destroying the pact” (ibid.: 682)

In short, institutions persist because the actors have incentives to abide bythe arrangement, as gains from maintaining the existing institutions aregreater than without it In the rationalist view, institutional change occurswhen it is no longer in the self-interest of actors to reproduce the existinginstitution In this view, actors design and build institutions as a solution to thecollective action problems because those institutions allow actors to cooperateand reach mutually beneficial outcomes Once actors create an institution in acertain form, the reproduction mechanisms discussed above continue to

encourage actors to maintain it in the same form Its reproduction cycle endswhen the institution ceases to serve the interest of the actors That, in turn,leads to institutional change because actors will seek to replace it with anew institution Based on the collective action theory, a hypothesis can beformulated as follows:

H1: Actors build institutions to overcome collective action problems andachieve mutual benefits The institutions continue to exist as long as theyperform such function and advance actor interests

2 A power-based rationalist theory

While some rational-choice theorists are concerned about collective-actionproblems, another group of rationalists focuses on the pursuit of poweramong actors when explaining the origin and evolution of institutions Forexample, Knight (1992, 1995) rejects the efficiency-based rationalist argumentand argues,“Institutions are not created to constrain groups or society in an

effort to avoid suboptimal outcomes, but, rather, are the by-product of stantive conflicts over the distributions inherent in social outcomes” (Knight1992: 40).11 From this viewpoint, it is considered likely that political leadersattempt to design institutions for the purpose of gaining strategic advantagesvis-à-vis other actors, who would resist such attempts The stronger the poli-tical leaders’ bargaining power in relation to other actors, the closer theinstitutional outcome gets to the political leaders’ intended design In thissense, institutional outcome reflects the asymmetries of the bargaining power

sub-of political actors.12

One advantage of this type of argument is that it explains inefficienciescaused by certain institutions that are extremely difficult for the efficiency-basedargument to explicate Knight claims that rather than focusing on collectivegoals, actors build institutions to advance their self-interests Thus, “Socialinstitutions may fail to achieve Pareto-optimal outcomes because it is not inthe interest of those who establish those institutions to do so” (Knight 1992:37–38).13

Ramseyer and Rosenbluth (1993, 1995) take a power-based rationalistapproach to explain institutional evolution in Japanese politics and argue thatpowerful actors manipulate the institutional framework to their privateadvantage They contend that political institutions in postwar Japan reflectedthe interests of LDP politicians, who had the power to manipulate the insti-tutional environment to their own advantage These institutions allowed LDPpoliticians to subordinate other social actors, such as bureaucrats andjudges.14Also, in their 1995 book, they echo their earlier argument and arguethat prewar Japanese bureaucrats were just agents of powerful actors(oligarchs in the Meiji period, party politicians in the Taisho- period, and themilitary during the wartime period), who had the power to manipulate theinstitutional environment to their advantage

Additionally, Silberman (1993) argues that political leaders in Japan’s Meijiperiod attempted to secure their power by transforming the government into

an autonomous bureaucratic state consisting of highly educated officials Inorder to become a bureaucrat, one had to pass an extremely competitivequalification exam, which was open to everyone, but which only a select fewhad the ability to pass This system provided political leaders with the legiti-macy to govern the nation while effectively denying other social groups themeans to challenge their power In short, these rationalist scholars argue thatpowerful political groups pursue the development of institutions for their ownbenefit, and that resulting institutions are used as instruments to keep otheractors under control and maintain the existing power structure A hypothesisdrawn from a power-based rationalist theory can be summarized as follows:

H2: Institutional arrangements reflect the power balance among actors.Powerful actors build institutions to secure or further increase theirpower The institutions continue to exist as long as the given powerstructure remains unchanged

in terms of critical juncture, path dependency, and ideas

1 Critical juncture

A critical juncture is defined here as an important moment in history marked

by fundamental transitions from the preexisting settings These can includeregime shift, revolution, democratization, and establishment of a new eco-nomic system Critical junctures often follow certain exogenous shocks thatinduce social transitions, such as war, economic crisis, and political turmoil

As such, exogenous shocks function as vital catalysts for institutional changebecause, if an existing institution fails to adapt sufficiently to the environmentthat follows a shock, its credibility will be significantly undermined That, inturn, motivates policymakers to build a new set of institutions

Various factors, however, will influence the shape of the new institutionalsetting Such factors include the domestic power balance, international politicaland economic conditions, intellectual discourse, and previous experiences ofthe policymakers Such a juncture can be deemed “critical” if “once a

particular option is selected it becomes progressively more difficult to return

to the initial point when multiple alternatives were still available” (Mahoney2000a: 513) Collier and Collier (1991) claim that all critical junctures sharethree basic components: first, antecedent conditions; second, cleavage; andthird, legacy “Antecedent conditions” refers to the conditions prior to thefundamental social transition These are disrupted by“cleavage” or crisis thattriggers change in the old system The “legacy” component corresponds tothe prolonged impacts of changes made during a critical juncture Scholarsbelieve that it is critical to study how policies made during a critical junctureestablish a trajectory for institutional evolution in subsequent periods.15

2 Path dependency

When analyzing institutional continuity, historical institutionalists exploremechanisms of institutional reproduction, which results in a condition oftenreferred to as “path dependency.” Mahoney (2000b: 507) defines it as “his-torical sequences in which contingent events set into motion institutionalpatterns or event chains that have deterministic properties.” Pierson (2004: 10)describes it as“the dynamics of self-reinforcing or positive feedback processes

in a political system.” Because of a self-reinforcing/reproduction mechanism,once a particular path becomes“locked in,” it becomes increasingly difficultover time to reverse it Scholars have pointed out several mechanisms of pathdependency that can contribute to the rigidity (or“stickiness”) of an institution.Studies of institutions explore various mechanisms of path dependency,such as increasing returns (North 1990; Arthur 1994; Thelen 1999) and sunkcosts (Keohane 1984; Collier and Collier 1991; Pierson 1994) For institutions,increasing returns refers to the condition in which the benefits of a giveninstitution increase as more actors take part in the institutional arrangement

As the institution generates more benefits, it self-reinforces its reproductionand makes the course of development harder to reverse Sunk costs, in terms

of institutions, refer to the unrecoverable time and effort invested in lishing the institutions themselves The larger the sunk costs of a given institution,the harder it becomes to replace because,“The depth and interrelatedness ofaccumulated investments may make the adoption of previously plausiblealternatives prohibitively costly” (Pierson 2004: 152)

estab-An increasing number of studies have analyzed various causal impacts ofpath dependency on institutional development (Thelen and Kume 1999;Mahoney 2002; Kato- 2003; Thelen 2004; Krauss and Pekkanen 2011) However,

as Hall points out, existing studies on political institutions have not fullyexplored the mechanisms of path dependency Hall claims that path dependencystudies contain an unresolved issue:“What features of the social, economic, orpolitical world give developments in the distant past such influence over sub-sequent actions?” He also contends, “We need better conceptualizations ofhow path dependence operates and more methodological sensitivity to thepossibility that it might be operating” (Hall 2009: 247) Responding to such a

demand in the field, I attempt to advance the discussion of positive feedback

effects in path dependency by focusing on ideational factors that can also erate positive feedback cycles and have significant impacts on the evolutionarypath of institutions

gen-3 Ideas in politics

Policymakers cannot always obtain sufficient information to make rationaldecisions because there is a high level of uncertainty in politics Since it is noteasy for policymakers to make rational decisions without intellectual assistance,they must rely on some kind of ideational guidance Such guidance plays acritical role when policymakers assess information before making decisions,16

because it helps them understand what current conditions could mean to thenation, and normally helps identify solutions to problems (for discussions onthe role of ideas, see Hall 1989, 1992; Weir 1989, 1992; North 1990; Sikkink

1991, 1993; Haas 1992; Goldstein and Keohane 1993; Goldstein 1993;Halpern 1993; Kier 1995; Keck and Sikkink 1998; McNamara 1998; Belandand Cox 2011).17

Ideational guidance helps actors make policies, but it also constrains theirdecision-making by limiting the availability of alternative options Onceactors adopt a given set of ideas as the foundation of their ideational gui-dance, they will make the types of policies that comply with that guidanceand will reject policies that contradict it As much as it helps actors makepolicies, ideational guidance constrains actors’ policy choices Therefore,ideational guidance can result in institutional continuity by locking actorpolicies into certain patterns (North 1990; Arthur 1994; Hall 1989; Weir1989; Goldstein 1993; Kier 1995; Pierson 2004; Beland and Cox 2011) Forexample, Weir (1989) argues that the British economic bureaucrats adoptedand kept their faith in Keynesian ideas in the 1930s, which continued to affectBritish economic policies through the 1970s Kier (1995) also demonstrateshow the military’s organizational culture continued to shape military doctrine

in Britain and France during the interwar period

This book attempts to extend this argument about ideas and institutionalcontinuity by presenting a mechanism in which ideational guidance cangenerate a positive feedback cycle, which then reinforces the lock-in effect

I argue that this positive feedback cycle occurs as follows: first, positivefeedback can occur when policymakers hand down ideational guidance to theirsuccessors The new generation of policymakers is likely to embrace theirpredecessors’ ideational guidance because they are products of the sameeducational system, professional training, and career experience as their pre-decessors Because policymakers continue to rely on ideational guidance oversuccessive generations, they continue to reproduce the existing institutionsthat were built based on the original guidance

Second, a positive feedback effect can also occur as the ideational guidance

of policymakers spreads into other social groups Some ideas may proliferate

into other social groups, as they interact with policymakers.18 As additionalsocial actors accept ideational guidance from policymakers, they begin tosupport existing institutions that were built based on that guidance Whenideational guidance spreads from policymakers to other social groups, thepolitical supports for existing institutions are reinforced This makes itincreasingly easier for policymakers to reproduce the existing institutions.Thus, ideational guidance generates a positive feedback effect and facilitatesthe reproduction of institutions for an extended period These reproductionmechanisms make the path of institutional evolution increasingly hard toreverse (seeFigure 1.1).

In the next section, I will present my viewpoints on institutional change andcontinuity in the Japanese economy, drawing on an historical institutionalistapproach

V The evolution of the Japanese developmental state system and the legacy of a wartime idea – “to-sei keizai ron”

Japan’s modernization started after the Meiji Restoration of 1868 In the Meijiperiod the government played a key role in industrialization by introducingmodern technologies from advanced Western nations, establishing state-ownedenterprises, and investing in industrial infrastructure However, in the early20th century the government’s roles gradually declined as private businessbecame the driving force behind the Japanese economy, causing it to evolveinto a liberal market-based economy by the 1920s Most of the institutionalfeatures of the developmental state system, such as pilot agencies and industrialassociations, did not exist prior to the mid-1930s.19 In fact, private firmsoperated with minimal government interference and the business leaders andparty politicians, who wanted to keep the market free, had more influenceover the policymaking process than the military officers and bureaucrats, whowanted tighter state control

However, three events in the late 1920s prompted a departure from thesystem of the 1920s First, the Great Depression severely damaged the JapaneseFigure 1.1 Positive feedback effect of ideas

economy, causing many people to begin thinking that the selfish profit-seekingactivities of privatefirms needed to be restrained Second, the success of theSoviet Union’s Five-Year Plan (1928–33) shocked Japanese policymakers andconvinced them that state economic planning was necessary Third, in the late1920s some Japanese policymakers started to believe that Japan was on aninevitable collision course with the Soviet Union or the United States, whichwould result in a total war with one of those nations These events stronglyencouraged Japanese policymakers to introduce a new economic system withenhanced state control.

A prototype of the developmental state system was built in Manchuria inthe early 1930s, where Japan had seized substantial control by 1932 Military

officers, who led Japan’s state-building efforts in Manchuria, used thatopportunity to establish an industrial base, which they believed would beindispensable for Japan’s future war against the United States When buildingManchuria’s new economic system, policymakers attempted to create asystem that combined the virtues of both liberal market and Soviet-stylecommand economies Evidence suggests that the Great Depression andthe success of the abovementioned Soviet Five-Year Plan had convincedJapanese policymakers of the need to put industry under state control However,they did not go as far as to abolish private ownership because they acknowledgedthat the vitality of private entrepreneurship was an important part of indus-trialization Therefore, the new system in Manchuria was virtually a fusion ofthe two systems

Manchuria was used as a breeding ground for the developmental statesystem because, at that time, such a policy experiment was not conceivable inJapan due to strong opposition from zaibatsu (corporate conglomerate) leaders–who were proponents of the liberal market economy Furthermore, sinceJapanese zaibatsu had already gained strong influence in Japan’s other colonies(Korea and Taiwan), the developmental state system could not emerge inthose places.20 Manchuria, on the other hand, did not have powerful privatecorporations to oppose tight state control, so military officers and bureaucratshad a free hand in designing its new economic system The policy entrepreneurs

in Manchuria included officers of the Kwantung Army – a division of theJapanese Imperial Army stationed in Manchuria21– and Japanese bureaucratsplaced in the Manchurian government The most influential figures were IshiwaraKanji and Miyazaki Masayoshi Ishiwara was a staff officer of the KwantungArmy in charge of constructing Japan’s puppet state in Manchuria Miyazaki,who assisted Ishiwara in his state-building efforts, was a Soviet economyspecialist employed in a Japanese-owned, semi-public railroad company inManchuria

My analysis of Manchurian state-building efforts indicates that the worldviewsand educational/career backgrounds of these two individuals had significantimpacts on their policies In the early stage of his military career, Ishiwarastudied the history of world warfare in Germany, where he became con-vinced that the world was approaching the age of total war Furthermore,

because he had come to believe that Japan would inevitably face total warwith the United States as early as the late 1920s, he recognized the urgentneed to build up Japan’s military and economic capabilities This, hebelieved, could be accomplished by combining the strengths of liberalmarket and planned economies in Manchuria, thus transforming the newcolony into an industrial base for Japan This idea, which was later called,

“to-sei keizai ron” (a theory of managed economy),22 became the ideationalfoundation for the developmental state systems in Manchuria and wartimeJapan

Ishiwara and Miyazaki created a bureaucratic agency called the PlanningBureau (Kikaku sho) as a pilot agency of the Manchurian government Itwas designed to function as the “economic general staff” (keizai sanbo-honbu), by formulating grand strategies for the Manchurian economy andguiding its industrialization, much like a general staff office does in the mili-tary This Manchurian agency became the prototype of the wartime andpostwar pilot agencies in Japan To control industry, the Manchurian pilotagency established the “special corporation system” (tokushu gaisha seido),

in which the government exercised tight control over the management ofeach firm

Around this time, military officers found a political ally in a governmentfaction consisting primarily of younger officials, called the “reformistbureaucrats” (kakushin kanryo-) The Kwantung Army summoned a largenumber of those reformist bureaucrats from Japan to manage the Manchuriandevelopmental state system Many of those Japanese military and bureau-cratic officials, who had gained policymaking and administrative experience

in Manchuria, would return to Japan during the war and build anotherdevelopmental state system there Ishiwara returned to the Imperial Army’sGeneral Staff Office in Japan in 1935 and quickly attained tremendous influ-ence over the Japanese political arena Ishiwara’s ideas of the inevitable totalwar and to-sei keizai ron spread quickly among military officers and youngbureaucrats, stimulating them to think seriously about the need to prepare theJapanese economy for the upcoming war

To-sei keizai ron competed with three other economic models (i.e Marxism,national socialism (kokka shakaishugi),23 and free market liberalism), inthe intellectual discourse of this period Each model had enthusiastic propo-nents Leftist activists, labor organizations, and Marxist scholars advocatedMarxism, while right-wing activists and one faction in the military supportednational socialism For their part, big business leaders and party politicianstried to defend the liberal market economy There was severe competitionamong these groups in determining the guiding principles of the Japaneseeconomy By the late 1930s, however, the proponents of to-sei keizai ron haddefeated their rivals, and had begun introducing an economic system based

on their ideas

Ishiwara and his followers established a pilot agency– the Planning Board(kikaku in) – in Japan in 1937 Their successful experience in Manchuria

encouraged them to replicate the Manchurian model in wartime Japan.However, the Manchurian model, composed of pilot agencies and the specialcorporation system, required some modifications for suitability in Japan.Unlike Manchuria, where industry was still at a primitive stage, Japaneseindustry was much more developed and consisted of numerous privatefirms.Furthermore, even though the political power of big business was declining,the introduction of the special corporation system seemed unattainablebecause it would require private firms to give up a significant degree of theirmanagerial freedom.

Although Ishiwara lost his political influence in Japan shortly after theestablishment of the Planning Board, his ideas and reform plans were carried

on by the reformist bureaucrats One of those reformist bureaucrats was KishiNobusuke, who had gainedfirst-hand experience in managing a developmentalstate system as the Secretary of Industry in the Manchurian government Kishireturned to Japan and became Vice-Minister of Commerce and Industry in

1939 The bureaucrats in pilot agencies established a type of industrial ciation, called the “control association” (to-sei kai) system in 1941 Thiscontrol association system utilized indirect control, which some of thosebureaucrats had studied in Germany Although it was also a system of statecontrol, this system was considered less threatening to the business because itallowed the members of industrial associations to control themselves onbehalf of the government The bureaucrats thought this method of indirectcontrol was a viable alternative to the special corporation system In sum-mary, it can be said that Japan’s wartime economic system evolved from theManchurian model with slight modifications necessary to suit the conditions ofJapanese industry at the time

asso-In the postwar period, the American occupation forces, which werestrongly influenced by liberal antitrust ideology, tried to “democratize” theJapanese economy by dismantling Japan’s wartime economic system As apart of the economic democratization policies, the Americans successfullydismantled the various zaibatsu and carried out a major agricultural land-reform program However, while the Americans achieved their economicdemocratization goals in those areas, they were unsuccessful in dismantlingmany institutions of the wartime developmental state system Initially, theSupreme Commander for the Allied Powers (SCAP) agreed to allow theJapanese government to rebuild institutions of indirect state control, includingpilot agencies (the Economic Stabilization Board) and industrial associations(gyo-kai dantai), for one year after the end of World War II, as a temporarymeasure to stabilize the war-torn economy However, when SCAP instructedthe government to dissolve those industrial associations after the one-yearmoratorium, Japanese policymakers, many of whom had served as economicbureaucrats during the war, passively sabotaged SCAP’s efforts by delayingcompliance until the end of the occupation As a result, the wartime institu-tions survived SCAP’s reform plans and continued to operate throughout thepostwar period

Why did the postwar Japanese policymakers choose to maintain thosewartime systems? This question is particularly intriguing considering that thewartime Japanese developmental state system did not expand the productioncapabilities of Japanese industry or ensure political advantages for bureau-crats and military officers The keys that explain the reemergence of thewartime control institutions are the personal and ideational linkages thatconnect the Manchurian government, the wartime regime, and the postwargovernment First, there was the important personal linkage connecting theofficials of the Manchurian, wartime Japan, and postwar Japanese governments.Those bureaucrats who built the developmental state systems in Manchuria and/

or wartime Japan continued to occupy important positions in the postwarJapanese government, even though many military officials, politicians, andbusiness executives were purged by the American occupation forces Further-more, a number of wartime bureaucrats became leaders of postwar politicalparties and business organizations This important continuity in the powerstructure of Japanese politics partly explains the reemergence of wartimeinstitutions in postwar Japan

Second, in addition to the continuity in the power structure throughoutthese periods, the policymakers’ continuing confidence in the validity ofwartime ideational guidance – to-sei keizai ron – was equally important inexplaining the institutional continuity This is noteworthy because just havingthe same group of individuals does not always result in institutional continuity.Policy shifts can occur even if there is no change in the power structure.The wartime institutions were reproduced and continued to function in thepostwar period because the Japanese policymakers kept the same ideationalguidance

The wartime Japanese policymakers established such institutions as pilotagencies and control associations to embody key to-sei keizai ron concepts.Evidence shows that the postwar Japanese policymakers maintained theirwartime ideational guidance and attitudes, such as distrust of a liberal marketeconomy and the commitment to state control, and that those ideas wereinherited by the postwar generation of bureaucrats This ideational guidancealso motivated them to reestablish those institutions in the postwar period.Pilot agencies were rebuilt to alleviate the shortcomings of a liberal marketeconomy by preparing long-term plans for the national economy Industrialassociations were tasked with assisting the enforcement of pilot agency plans,coordinating industrial interests, and preventing market competition fromveering out of the government control

Furthermore, bureaucratic ideational guidance began to spread amongother actors, such as party politicians and business leaders, many of whomhad objected to state control in the past Although not all business leaderscame to share such views, many postwar business leaders welcomed governmentintervention and assisted the government in its policy implementation pro-grams Members of conservative parties, such as the LDP, began to embracethe ideas of a developmental state system as well As ideational guidance

took root in the new generation of bureaucrats, and gained acceptance fromother actors such as party politicians and business leaders, a strong positivefeedback effect was generated With increasing numbers of actors advocatingthe idea, opposition to the developmental state system weakened, making

it easier for policymakers to maintain the institutions This positive feedbackeffect generated by ideational guidance further facilitated the reproduction

of the developmental state system Thus, personal and ideational linkagesamong the governments of Manchuria, wartime Japan, and postwarJapan explain the prolonged life of those institutions for an extendedperiod of time Based on this understanding, I formulated the followinghypothesis:

H3: Policymakers rely on their ideational guidance when building tions and institutional arrangements reflect their ideas Institutionscontinue to exist and institutional change becomes less likely as thoseideas are passed on to the next generation of policymakers and spreadamong other actors

institu-VI Points of contention

There are two fundamental differences between the rationalist arguments and

my argument First, while the former maintains that decisions by actors tomaintain an institution are driven by their perceptions of self-interest, thelatter posits that such decisions are constrained by their ideas For rational-ists, self-interest incentives among actors (the pursuit of a mutually beneficialoutcome) shape their decisions and institutions persist for extended periodsbecause the actors who built them have incentives to maintain them I arguethat the ideas held by actors constrain their decisions, so institutions persistbecause their ideas make it difficult for them to choose alternative options.24

This does not mean that actors have no self-interests at stake They do, andtheir ideas and interests may synchronize on some occasions, so their decisionsmay seem to serve their individual self-interests as well However, their ideasand interests may conflict with each other on other occasions This is thesecond difference between the rationalist arguments and my argument Somerationalists argue that the self-interest of actors will prevail under such con-ditions because they believe that ideas are only useful when they justify theirinterests.25Other rationalists claim that actor ideas are shaped by the institu-tions, which also reflect their self-interests.26 Therefore, rationalists do notbelieve that ideas can have a significant impact on institutions In contrast,

I argue that the decisions by actors regarding institutions will reflect theirideas and that when their ideas and self-interests conflict, they are more likely

to make decisions based on their ideas – even if it negatively affects theirindividual interests This is because under uncertain conditions, actors do notalways have perfect information available to assist them in making rationaldecisions, so they often must rely on what their ideas tell them to do They are

thus constrained by their ideas and may not be able to choose options thatmore readily serve their individual interests.

Although I test the validity of each argument in my case study of theJapanese developmental state system, I do not expect tofind a single argumentthat will explain all cases and invalidate the rest I suspect that all factorsinfluenced institutional evolution to one degree or another Thus, the expla-nation I am trying to present for the developmental path of institutions ismulti-causal, not mono-causal, and I expect each perspective to help explainthe developmental path of Japanese economic institutions to a certain degree.However, even though I present a multi-causal explanation, I will also discussthe weaknesses of each perspective in order to clarify the boundaries of itstheoretical domain A multi-causality explanation may not be as parsimo-nious as a mono-causal explanation, but I believe that it will allow me toavoid exaggerating the explanatory power of a single theory, and thus present

a more accurate and comprehensive account of institutional evolution

VII Existing literature on the Japanese economy

Several studies have analyzed the important links between the wartime economicsystem and the postwar Japanese economy (Johnson 1982; Okazaki andOkuno 1993; Yonekura 1993; Kobayashi et al 1995; Noguchi 1995; Gao1997) While I draw upon somefindings from these studies, I also attempt toadvance the study of the Japanese economy by presenting an analysis ofinstitutional change in the 1930s, and the process of institutional evolutionfrom Manchuria to wartime Japan and postwar Japan Although existingstudies present some evidence to support their claims of similarities inthe industrial policies of wartime and postwar Japan, they fail to present aconvincing explanation as to what caused those institutions to change, whysuch similarities exist, and why postwar Japanese policymakers reestablishedthose wartime economic institutions

For example, Johnson (1982), Yonekura (1993), and Kobayashi et al.(1995) all claim that one of the reasons postwar Japan reestablished the war-time institutions was the personal linkage between bureaucrats of the wartimeand postwar governments They argue that many of the wartime bureaucratswho administered the managed economy assumed important positions

in postwar Japan, and that this personal linkage facilitated the reproduction

of wartime economic institutions However, as my analysis will show in

Chapter 5, personal linkage cannot fully explain the reproduction process ofthe Japanese developmental state system

In addition, my study presents the significant roles played by private firms

in the policymaking and implementation processes, through their tion in industrial associations Johnson’s study focuses primarily on the role ofbureaucrats in controlling industry and leading economic development, andunderestimates the roles played by private firms (Samuels 1983; Schaede2000) This study aims at presenting an account of the development of the

participa-Japanese economy that captures the roles played by state and private-sectoractors, as well as the significant conflicts between them.

A study of Japanese industrial policies by Gao (1997) has an analyticalinterest similar to this study and addresses the significant impacts of eco-nomic ideology Gao argues that an economic ideology that is inspired by thegovernment’s industrial policies, which he calls “developmentalism,” exists inJapan However, a major difference exists between this book and Gao’s book.While Gao is interested primarily in a series of changes in Japan’s economicideas, I focus on the continuity in economic ideas from wartime Japan to the1960s That is not to say that economic ideas did not change at all duringthe period Gao rightly points out that Japan had to adapt to new marketenvironments after the end of World War II He contends that there werethree distinctive policy paradigms in Japan– the managed economy betweenthe 1930s and 1945, export promotion in the 1950s, and market liberalization

in the 1960s – and that such changes in industrial policies were induced bymodifications in ideas

Nonetheless, I believe such changes are of minor importance compared

to what did not change This is particularly true when looking at the coreinstitutions of the Japanese developmental state system because policies aremore susceptible to changes than the overall functions of institutions In otherwords, I am more interested in the continuity in the core institutions of theJapanese developmental state system than the minor changes in industrialpolicies My analysis of Japanese economic institutions shows that the func-tions of pilot agencies and industrial associations in Japan showed remarkablylittle change from wartime through the 1960s

Furthermore, although there were some changes in the industrial policies ofpilot agencies during the abovementioned period, there was also significantcontinuity in the industrial policies The postwar pilot agencies (i.e theESB and MITI) maintained exactly the same types of industrial policies(e.g economic plans and industry rationalization policies) as their counterparts

in Manchuria and wartime Japan However, Gao’s study underestimates thisimportant continuity in Japanese industrial policies and does not pay sufficientattention to intellectual discourse among state elites.27

Another significant contribution of this study is its analysis of the importantimpact of Japanese policymaker experiments in Manchuria, as well as onpolicymaking in wartime and postwar Japan Many scholars have studied thedevelopmental path of the Japanese economy, but very few have explored thelink between Manchuria and postwar Japan Johnson refers to the experi-ences of some Japanese bureaucrats in Manchuria very briefly (Johnson 1982:

Chapter 4) However, his analysis of those Manchurian experiences does not

go beyond a few short references.28Gao also makes brief references to Japanesepolicymakers’ experiences in Manchuria, but his account on Manchuria iseven more limited than Johnson’s He pithily introduces Ishiwara and Miyazaki,but does not look into any of their policy plans, writings, or other documents

to prove his points (Gao 1997: 104–5) Despite his acknowledgement of the

significance of the Manchurian case, Gao, like Johnson, does not analyze thecritical link between Manchuria and Japan.

In this book, I will present a more detailed examination of Manchurianand Japanese industrial policies I explore three types of industrial policies ofpilot agency in Manchuria, wartime Japan, and postwar Japan: economicplanning, industrial rationalization, and market stabilization To illustratethat there is indeed a critical link between them, I examine how these industrialpolicies evolved from Manchuria through postwar Japan I also examine theinstitution-building processes in wartime and postwar Japan and show howthose processes reflected the experiences gained by policymakers in Manchuria

In addition, this study looks into the context behind the modification of theManchurian model in wartime Japan These issues have not been examined inthe existing studies

In summary, even though several studies have explored some of the issuesanalyzed in this study, this study claims to make important contributions Myanalysis of the institutional change and evolution, along with a theoreticalframework of historical institutionalism, offers a new way to understand thedevelopmental path of the Japanese political economy in a long-term viewpoint.This study is perhaps thefirst attempt to apply these theories of historical insti-tutionalism to an analysis of the Japanese developmental state system It aims

at capturing the significant links between Manchuria, wartime Japan, andpostwar Japan; the interaction between state and private-sector actors;and the vital impact of ideational guidance on the continuity of the coreinstitutions of the Japanese developmental system

VIII Outline of the book

The rest of this book is structured as follows: Chapter 2 presents a briefcomparative analysis of Japan’s economic systems in the Meiji period, the1920s, and the post-1930s Through this analysis, I demonstrate that therewere fundamental differences among those systems and that the develop-mental state system did not exist before the 1930s This is a critical finding

of this study because it suggests that the origin of the developmental statesystem can be found in the 1930s and also allows me to draw a set ofempirical questions

Chapters 3,4, and5are the empirical chapters of this study, where I presentcase studies of Manchuria under Japanese control (1932–45), wartime Japan(1937–45), and postwar Japan (1946–65) In each of these chapters, I analyzethe roles of pilot agencies and industrial associations in promoting economicgrowth and the process by which the policymakers of those periods madedecisions regarding establishment of those institutions Particular attention isgiven to the personal and ideational linkages among the three cases by evaluat-ing how decisions made in previous case(s) affected subsequent institutionalevolution My analyses employ an analytical method known as “processtracing,”29 which investigates every step of the causal sequence between the

explanatory variable and the dependent variable, explicating how the formerbrought about changes in the values of the latter.

The concluding chapter compares and contrasts the industrial policies ofpilot agencies and the functions of industrial associations in the three cases toillustrate how these institutions evolved throughout the periods It then revisits

my arguments on the prolonged causal influence of the decisions made atcritical junctures and assesses the soundness of alternative explanations com-pared to my arguments Then, I discuss the implications I can draw from thisstudy concerning Japanese political economy and the study of institutions

2 The early stages of Japan ’s

industrialization

This chapter reviews Japan’s political economy prior to the 1930s while focusing

on the industrial policies of the Meiji period (1868–1912) and the conditions

of economy and politics of the 1920s The primary argument of this chapter isthat Japan’s political and economic structures prior to the 1930s were sig-

nificantly different from those after the 1930s Japan’s economic systems afterthe 1930s were quintessential examples of the developmental state system, which

is often associated with particular characteristics, such as powerful state elites,close state–business relations, and a number of unique business practices.First, meritocratic state elites (i.e economic bureaucrats) had strong influ-ence over the management of privatefirms and controlled the flow of capitaland other scarce resources in order to achieve rapid industrialization Second,the developmental state system entails close state–business relations through anetwork of retired bureaucrats who were reemployed in privatefirms, and throughindustrial associations that facilitated the cooperation of private firms withthe state Last, a developmental state system is marked by a number of uniquebusiness practices, such as bank-centeredfinancial systems, lifetime employmentsystems, seniority-based wage/promotion systems, and a cordial relationshipbetween labor and management (Johnson 1982, 1995; Wade 1990; Hatch andYamamura 1996)

However, Japan’s economic systems prior to the 1930s did not have thosecharacteristics The Meiji government’s approach to industrial developmentwas mainly through the establishment of state-owned enterprises and invest-ment in industrial infrastructure However, it is uncommon for economicbureaucrats in a developmental state system to establish state-owned cor-porations because they already have ways to control the management of pri-vatefirms and the flow of capital, and do not need to own businesses directly.Yet, the Meiji-period bureaucrats lacked such control over privatefirms, andthus, direct ownership was the only option available to them, even though itwas very costly for the government

Additionally, since they were recruited primarily through personal ties topowerful politicians rather than passing by competitive qualifying examina-tions, Meiji bureaucrats were not as competent as their wartime or postwarcounterparts, who dominated the policymaking process with their policy

expertise Furthermore, the 1920s economy was a liberal market-based economy,

in which government intervention was kept to a minimum Therefore, it wasmore similar to the economic systems in Great Britain and the United Statesthan a developmental system, and the power of economic bureaucrats was muchweaker before the 1930s than in later periods– particularly during the 1920swhen business leaders and politicians dominated the political scene

The nonexistence of the developmental state system before the 1930s iscritical to this study because it indicates that the developmental state systememerged in Japan during or after the 1930s In the following pages, I brieflyreview the economic systems of the Meiji period and the 1920s, focusing onthe power of the economic elite vis-à-vis other social groups, their industrialpolicies, and business practices among privatefirms From the comparison ofJapan’s political and economic structures in different time periods, I draw aset of empirical puzzles, which is explored in subsequent chapters

I Industrialization in the Meiji period (1868 –1912)

The Japanese government began its modern state-building process after theMeiji Restoration of 1868 In the early stage of industrialization, the governmentcarried out several policies to nurture industry by investing in state-ownedenterprises, building industrial infrastructure, and encouraging technologytransfers from advanced European countries by hiring foreign engineers.Although the Meiji government played an active role in facilitating indus-trialization, its industrial policies were fundamentally different from those ofgovernments after the 1930s

The primary industrial policy of the Meiji government was called“shokusanko-gyo-” (production promotion) policy, which primarily emphasized threeareas:first, development of industrial infrastructures; second, establishment ofstate-owned enterprises; and third, facilitating acquisition of foreign technology.The Meiji government first began its modernization project with industrialinfrastructure development, particularly in thefinancial, railroad, and maritimetransportation sectors The government considered development in these threesectors indispensable to modernization of the Japanese economy Throughthe 1882 establishment of the central bank, the Bank of Japan, the governmentunified the currency system, which was fragmented and disorganized duringthe Edo period (1603–1867) The government also approved establishment of

153 private banks, which created a network of modern financial institutionsthroughout the nation Then the government initiated a massive campaign todevelop a nationwide transportation network Not only did the governmentinvest heavily in railroads and the maritime transportation industry, it alsoencouraged private investment in these sectors by introducing a modern privatecorporate system According to Nakamura, the financial sector received78.1% of total private investment in 1884, while investments infinance, railroad,and maritime transportation sectors accounted for about 77.1% of privateinvestments in 1894 Thesefigures make a clear contrast with the low level of

private investment in the manufacturing sector (2% of private investment in

1884 and 11.4% in 1894) (Nakamura 1978: 74), which was the central focus

of industrial policies in later periods

Second, the Meiji government established a number of state-owned enterprises

in a wide range of industries including railroads, steel, mining, shipbuilding,postal service, and telecommunications Development in such industriesrequired a large amount of capital and was associated with high risks Privatecompanies were unwilling to take such risks and incapable of taking suchheavyfinancial burdens in the late 19th century Therefore, the government chose

to establish state-owned enterprises Examples of state-owned enterprises includeTomioka Silk Mill, Japan National Railway, shipyards in Nagasaki and Hyogo,Yahata Steel, Takashima Coal Mine, Miike Coal Mine, Kamaishi Iron OreMine, and many more The government invested public funds heavily in thesestate-owned enterprises, and the Ministry of Domestic Affairs (Naimu sho-)and Ministry of Industry (Ko-bu sho-) were directly involved in the manage-ment of those enterprises As a result, the corporate structure and productionsystems of state-owned enterprises became a model for private firms thatwould enter those industries later

Direct management by the government was the primary instrument of itsindustrial policy until the 1880s However, once a business was well estab-lished, the government transferred its ownership stake to private companies.Such state-owned enterprises were sold at relatively low prices, and the gov-ernment offered private companies easy payment plans with low interest rates.These attractive deals were given to “seisho-” (political merchants) a termgiven to merchants who had close ties with the government officials andpoliticians The most successful seisho- first harvested state-owned enterprisesthat had been nurtured by the government They later established corporateconglomerates called zaibatsu For instance, the government transferredMiike Coal Mine and Tomioka Silk Mill to the Mitsui zaibatsu, NagasakiShipyard and Takashima Coal Mine to the Mitsubishi zaibatsu, and Hyo-goShipyard to the Kawasaki zaibatsu Those enterprises grew into some ofJapan’s largest firms in the later periods and played critical roles in Japan’s earlyindustrialization This privatization process caused rent-seeking activities by bigbusiness and became a source of the crony relationships between business andpoliticians that are discussed in the next section

Third, the government facilitated acquisition of foreign technology andmachinery The Japanese government invited a number of engineers fromEurope and the United States to work for the government in the construction

of factories, railways, and telecommunications In 1874, as many as 213 foreignengineers were employed in state-owned enterprises (Nakamura 1978: 77).This contrasts strongly with the Edo government whose closed-door policyallowed no foreigners on Japanese soil except for some Chinese and Dutch.Additionally, the government imported foreign machines and technologies andthen leased them to private firms For example, the government importedenough spinning machines to run 10 factories It then used some of them in

two state-owned enterprises, but leased the rest to private textile firms tofacilitate development of the industry (Nakamura 1978).

There were significant differences between the Meiji government andthe governments of the wartime and postwar periods in their industrial poli-cies and their relations to industry First, the primary focuses of industrialpolicies were different The Meiji government’s main goal of industrial policieswas development of industrial infrastructure, while expansion of productioncapacity– particularly in manufacturing industry – was the primary objective

of industrial policies in the wartime and postwar periods, as will be discussed

in the following chapters Second, the level of state control over privatebusinesses was much lower in the Meiji period because the government hadlimited policy tools to control them The Meiji government had neither pilotagencies that had authority to plan and direct industrialization, nor industrialassociations that allowed the government to coerce industry to assist inthe enforcement of its policies Those institutions would not emerge until thewartime period, so the Meiji government was unable to influence the manage-ment of private firms directly Thus, when the government wanted to leaddevelopment of a certain industry, it needed to establish a state-owned enterprise.Yet, many state-owned enterprises were privatized by the end of the Meijiperiod because it was too expensive for the government to operate such busi-nesses by itself The post-Meiji governments rarely established state-ownedenterprises.1

II Liberal market economy in the 1920s

The Meiji government’s industrial policies focused on development of industrialinfrastructures, establishment of state-owned enterprises, and acquisition offoreign technologies and machineries However, private firms became thecentral players of the Japanese economy by the 1910s because the Meijigovernment had privatized many of its state-owned enterprises Thus, theJapanese economic structure developed into a liberal capitalist system inwhich the government stayed out of market activities; economic bureaucratsdid not play significant roles; and private business activities were largely freefrom government regulations It can be seen, therefore, that the Japaneseeconomic structure in the 1920s was fundamentally different from that ofthe postwar periods (Okazaki and Okuno 1993; Werner 2001; Miwa andRamseyer 2002; Hoshi and Kashyap 2001) Okazaki and Okuno contendthat,“The Japanese economy before the 1930s was basically an Anglo-Saxontype of classical market economy” (Okazaki and Okuno 1993: ii–iii)

First, labor-management conditions were significantly different management conditions in the postwar period were marked by peculiar laborpractices, such as lifetime employment, a seniority-based wage and promotionsystem, enterprise unions, egalitarian income distribution, and cordial labor-management relations Labor-management relations were quite hostile between

Labor-1919 and 1938 Labor unions organized workers in different companies and

were very active and hostile to companies, resulting in a number of labordisputes Statistical data collected by Nihon Keizai To-kei Kenkyu-jo (JapanEconomic Statistics Research Institute) show that there were as many as 2,388labor disputes in 1919– a sudden increase from 417 cases during the previousyear This number remained high until 1938, the year Japan shifted to awartime economy (see Figure 2.1) Additionally, scholars suggest that theorigin of the lifetime employment and seniority wage system can be found inthe government’s labor policies during the wartime period (Okazaki andOkuno 1993; Kobayashi et al 1995; Noguchi 1995).

Second, while firms in the post-1930s periods relied heavily on indirectfinancing through bank loans, most companies in the 1920s financed themselvesthrough the equity market,2and individual stockholders exerted strong influenceover corporate governance This is important because the postwar governmentused bank loan-based financing as a tool to control private firms By main-taining tight control over banks, the postwar government could indirectly

influence the management of private firms, which heavily relied on bank loans

to finance their business (Allen and Gale 2000; Hoshi and Kashyap 2001;Miwa and Ramseyer 2002; Hayes 2004) Hayes describes the heavy reliance

of postwar Japanesefirms on government by saying, “Japanese enterprises areprimarily financed with borrowed money, a practice that differs from thatfound in most other industrialized countries Whereas Western companieshave somewhere around 60% equity investments, leaving 40% borrowedmoney, the Japanese have only 20% equity and 80% debt” (Hayes 2004: 163).However, according to data from Okazaki and Okuno, on average between

1966 and 1970, firms raised 81.2% of their funds from bank loans and only6.8% from equities, while firms raised 119.0% of their funds from equitiesand–25.5%3from bank loans in the period between 1931 and 1935 (Okazakiand Okuno 1993: 10) These data suggest that stocks were the major source offinance for Japanese firms until 1935, and the heavy reliance on bank loansstarted in the later periods The primary reason behind the popularity ofFigure 2.1 Labor disputes, 1910–45

Source: Nihon Keizai To-kei Kenkyu-jo (1958: 296)

directfinancing through the equity market was its lax regulatory environment.Compared to the post-1930s period, it was much easier for firms – evensmaller enterprises – to list their stock on the market (Okazaki and Okuno1993; Allen and Gale 2000) Most regulations in the financial sector wereintroduced after the outbreak of war with China in 1937,4 and, until then,firms were able to raise funds in the equity market with minimal restrictions.Hoshi and Kashyap claim, “The late 1930s and early 1940s was the timewhen bank financing became the dominant funding source for most of theindustrial firms involved in the war effort” (Hoshi and Kashyap 2001: 4).Bank loan-based financing was not popular before the mid-1930s because ofthe nation’s low savings rate, the relatively free equity market, and the lack ofthe so-called “main-bank system” during the pre-1930s period (seeTable 2.1

and Figure 2.2) Therefore, unlike their post-1930s counterparts, firms wereforced to focus on short-term rather than long-term profits because of strongstockholder pressure and threats from take-over bids

Table 2.1 Corporatefinance sources (%)

(Okazaki and Okuno 1993: 10)

Figure 2.2 Household savings and consumption rates, 1930–64

Source: Bank of Japan (1966: 50–51)