Economicdevelopmentandinternationaltrade1990pdfa [611128]

ECONOMIC DEVELOPMENT
AND INTERNATIONAL TRADE
THE JAPANESE MODEL
BY IPPEI YAMAZAWA
TRANSLATED AND REVISED BY IPPEI YAMAZAWA

EAST-WEST CENTER
RESOURCE SYSTEMS INSTITUTE
HONOLULU, HAWAII

THE EAST-WEST CENTER is a public, nonprofit educational institution
established in Hawaii in 1960 by the United States Congress with a mandate
"to promote better relations and understanding among the nations of Asia,
the Pacific, and the United States through cooperative study, training, and
research.'*
Some 2,000 research fellows, graduate students and professionals in busi­
ness and government each year work with the Center's international staff on
major Asia-Pacific issues relating to population, economic and trade poli­
cies, resources and development, the environment, culture and communica­
tion, and international relations. Since 1960, more than 25,000 men and
women from the region have participated in the Center's cooperative programs.
Principal funding for the Center comes from the United States Congress.
Support also comes from more than 20 Asian and Pacific governments, as
well as private agencies and corporations. The Center has an international
board of governors.

Economic Development
and International Trade

ECONOMIC DEVELOPMENT
AND INTERNATIONAL TRADE
THE JAPANESE MODEL
BY IPPEI YAMAZAWA
TRANSLATED AND REVISED BY IPPEI YAMAZAWA
EAST-WEST CENTER
RESOURCE SYSTEMS INSTITUTE
HONOLULU, HAWAII

Copyright © 1990 by the Resource Systems Institute, East-West Center.
All rights reserved. No portion of this book may be reproduced, by any process or
technique, without the express written consent of the publisher.
First published in the Japanese language in 1984 as Ninon no Keizai Hatten to Kokusai
BungyO by Toyo Keizai Inc., Tokyo, Japan.
English-language edition first published in 1990.
Resource Systems Institute, East-West Center, 1777 East-West Road, Honolulu, Hawaii
96848.
Printed in the United States of America.
97 96 95 94 93 6 5 4 3 2
Library of Congress Cataloging-in-Publication Data
Yamazawa, Ippei, 1937-
(Nihon no keizai hatten to kokusai bungyo. English]
Economic development and international trade : the Japanese model
by Ippei Yamazawa ; translated and revised by Ippei Yamazawa.
p. cm.
Translation of: Nihon no keizai hatten to kokusai bungyo.
ISBN 0-86638-121-X
1. Japan—Economic conditions—1945- 2. Japan—Economic
policy—1945- 3. Japan—Commerce. I. Title.
HC462.9.Y33513 1990
330.952'04-dc20 90-8394
C1P

Contents
Tables and Figures vii
Preface xi
Acknowledgments . xv
Part I
Japan's Economic Development and International Trade
1 Economic Development and Trade Structure: The Long
View 3
2 Interaction Between Trade and Development: A Japanese
Model 27
Part II
Industry Growth and International Trade
3 Raw-Silk Exports and Japanese Economic Development 51
4 Catching-Up Product Cycle Development in the Textile
Industry 69
5 CPC Development of the Iron and Steel Industry 87
6 Trading Companies and the Expansion of Foreign Trade 109
Part III
The Role of Government in Industrial Growth
7 Japan's Industrialization and Protection Policy Before
the Second World War 141
8 Trade and Industrial Policies During the High-Growth
Period 165

v/ Contents
9 Trade Conflicts and Structural Adjustment 189
Part IV
Japan's Development Experience and Contemporary
Developing Countries
10 Is the Japanese Model Applicable? 231
Notes 247
Bibliography 257
Index 265

Tables and Figures
Tables
1.1 Development of Japan's Foreign Trade 4
1.2 International Comparison of Trade Expansion: Average
Annual Growth Rates 5
1.3 Commodity Composition of Japan's Exports 8
1.4 Commodity Composition of Japan's Imports 9
1.5 Composition of Industrial Output 14
1.6 Composition of Industrial Goods Imports 15
1.7 Composition of Industrial Goods Exports 16
2.1 Estimation of Equations for Export, Import, and Related
Variables 40
2.2 Japan's Balance of Payments 46
3.1 Direction of Japan's Raw-Silk Exports 54
3.2 Sources of U.S. Silk Imports 55
3.3 Japan's Raw-Silk Export Growth 58
3.4 Constant Market Share Analysis of Japan's Silk Exports
to the United States 59
3.5 Estimates of U.S. Import and Japanese Export Functions
of Raw Silk 61
3.6 Foreign Exchange Earned Through Silk Exports 65
3.7 Production Effects of Silk Exports 66

viii Tables and Figures
4.1 The Textile Industry's Share of Japanese Manufacturing
Activity 70
4.2 Changes in the Export Market Structure of Japanese
Cotton Fabrics 76
4.3 Direct Foreign Investment by the Japanese Textile Industry 85
5.1 Average Annual Growth Rates of Steel Products 90
5.2 Government Policies Protecting the Iron and Steel Industry 91
5.3 Model of CPC Development 95
5.4 CPC Development of Steel Products: 1900-14, 1921-36 99
5.5 CPC Development of Steel Products: 1952-68 104
6.1 Shares in Total Exports, Direct Export Ratios, and Rates
of Increase in Exports by Major Commodity Groups 113
6.2 Shares in Total Imports, Direct Import Ratios, and Rates
of Increase in Imports by Major Commodity Groups 114
6.3 Regression of Direct Export Ratio and Export Expansion 121
6.4 Regression of Direct Import Ratio and Import Expansion 122
6.5 Direct Trade Ratios and Rates of Increase of Selected
Commodities 124
6.6 Features of Mitsui Branches by Region 127
6.7 Export Structure of Japanese GTCs: 1960 and 1970 130
6.8 Overseas Branches of Japanese GTCs: 1961 and 1970 134
7.1 Rice and Sugar Imports of Japan Proper: 1898-1933 145
7.2 Simple Average Tariffs on Individual Commodity Groups
by Industry 152
7.3 Simple Average Tariffs on Individual Commodity Groups
by Economic Use 153
7.4 DFI in Japanese Manufacturing Industries: 1899-1940 162
7.5 Decreased Import Dependence on Heavy Electrical
Machinery 164
8.1 Japan's Participation in Worldwide Economic Integration
after the Second World War 168
8.2 World Industrial Production and Trade in Industrial
Products: Market Economy Countries 170

Tables and Figures ix
8.3 Phased Implementation of Trade Liberalization 174
8.4 Tariff Structure at the Start of Trade Liberalization: April
1962 176
8.5 Export-Promotion Policy 181
8.6 Preferential Export Financing: 1955-72 183
8.7 Reduction of Tax Revenues under Export-Promotion
Measures 185
9.1 Growth and Trade Performance Variation During Global
Adjustment 193
9.2 Pattern of Japan's Economic Growth: 1974-87 198
9.3 Changes in the Industrial Structure 203
9.4 Changes in the Export and Import Structure 205
9.5 Reduction of Iron and Steel Production Capacity 206
9.6 Globalization of Japanese Firm Activities 212
9.7 Changes in the Employment Structure 224
10.1 Eligibility Requirements and Preferential Treatment for
Korean GTCs 241
10.2 Comparison of Korean and Japanese GTCs 243
Figures
1.1 Compositional Changes in Imports of Industrial Goods by
End Use 10
1.2 Import Substitution and Export Expansion of Two Major
Industries 19
1.3 Industrial Output, Exports, Imports, and Raw Material
Imports 22
1.4 Long-Term Swings in the Balance of Payments 23
2.1 Schematic Diagram of the Catching-Up Product Cycle
Development of an Industry 29
2.2 Terms of Trade, and Relative Export and Import Prices:
1885-1980 38
2.3 Long Swings in Manufacturing Output: Growth and
Structural Change 42

x Tables and Figures
2.4 Schematic Diagram of Alternative Correspondence
among Production (S), Trade (X and M), and the
Trade Balance (X/M) 44
3.1 Mechanism of Silk Export Growth 63
4.1 Catching-Up Product Cycle Development of Cotton Yarn
in Japan 72
4.2 Catching-Up Product Cycle Development of the Cotton
Textile Industry in Japan 73
4.3 Diversification in Textile Industry Development 80
5.1 CPC Development of Steel Products 89
5.2 Dependence on Pig Iron and Steel Product Imports 93
5.3 CPC Development of Major Industries 106
6.1 Japan's Direct Export and Import Ratios: 1874-1960 111
7.1 Average Tariffs of Japan: 1868-1980 143
7.2 Average Tariffs on Commodity Groups Classified by
Economic Use 155
8.1 Changes in the Frequency of Tariffs on Mining and
Manufacturing Products: Comparison of Japan, the EC,
and the United States 178
9.1 Fluctuations in the Global Economy 191
10.1 Catching-Up Industrialization of East and Southeast Asian
Countries 235
10.2 Alternative Mechanisms Underlying CPC Development 237

Preface
International trade has played a key role in Japan's economic development.
Through it Japan has obtained energy and resources unavailable domestically;
trade has also introduced new products and technology and has contributed
to the development of modern industrial production through import substi­
tution at home and export expansion abroad. As a late-starting industrial
country, Japan's industries have typically followed a catching-up product cy­
cle (CPC) process of development.
The catching-up product cycle development process is characterized by im­
port substitution followed by export expansion. Import substitution versus
export expansion has become the major choice in development strategy con­
fronting contemporary developing economies. Economists in the developed
countries, as well as those at multilateral institutions such as the World Bank,
suggest that many developing countries have overemphasized import substi­
tution, which thus hampers export expansion after import substitution is
achieved; many of these economists recommend that these countries shift their
policies toward export expansion through deregulation and promotion of ex­
ports. The CPC development process differs from this orthodox view by stress­
ing the continuity of import substitution while pursuing export expansion.
The term "Japanese model" used in this book describes not so much a
development experience unique to Japan but a rational pattern of industrial
development and trade in a late-starting industrial country under certain
resource and market constraints. This pattern is relevant for contemporary
newly industrializing economies (NIEs), and near-NIEs, in spite of the greater
technological gap and more severe international market situation confront­
ing them.
What role has the Japanese government played in Japan's successful im­
port substitution and export expansion? Since its first attempts at develop­
ment Japan has had a strong centralized government staffed by well-trained

xii Preface
bureaucrats. Some Foreign observers have attributed Japan's success to the
foresight and leadership of the government, as in the oft-cited notion of
"Japan Incorporated." However, many Japanese economists believe this view
overemphasizes the government's role and are inclined to give more credit
to private entrepreneurship. Nevertheless, there has long been closer cooper­
ation between government and private business in Japan than in the United
States and some European countries, and this does seem to have contributed
to Japanese industrial development. This book attempts to present a balanced
assessment of the government's role in Japan's economic development.
The Japanese economy managed to overcome the two oil shocks of the
1970s but it has accumulated huge trade surpluses and faces severe trade con­
flicts with its major trading partners, especially the United States. Japan's
main policy efforts have recently been geared toward restructuring its econ­
omy and industry so that further growth is compatible with that of its trad­
ing partners. Japanese economists are helping to shed light on the changes
by explaining these restructuring efforts and how they are being implemented.
They can be properly understood only with a precise knowledge of the
Japanese policy environment.
This book is organized into four parts. Part I gives a summary of Japan's
development process over the long term and analyzes the interaction between
production and trade structures with ample consideration given to statistical
evidence. Part II analyzes the strategic role of four industries at different stages
of Japan's economic development—silk exports during the initial stage; tex­
tile and steel industries during CPC development, which is typical of such
labor- and capital-intensive industries; and the trading companies in the trade
expansion period. Part III focuses on trade and industrial policies adopted
by the Japanese government during the pre-Second World War period, the
high-growth period of the 1950s and 1960s, and the post-oil shock period.
Part IV discusses the relevance of the Japanese development experience to
contemporary NIEs and near-NIEs and presents two case studies of the
catching-up product cycle and general trading companies in representative
developing countries.
This book is based on its Japanese edition (Nihon no Keizai Hatten to
Kokusai Bungyd), which was published in 1984 and won the Japan Economic
JournaPs Best Economics Book of the Year Award. However, Chapters 2,
9, and 10 have been revised significantly in order to include the recent changes
in Japan's role in the international environment in the 1980s. For other chap­
ters, beyond mere translation into English, explanations and English-language
references for English-language readers have been added. Readers with
knowledge of the Japanese language should refer to the 1984 edition for
Japanese sources that have been omitted from this edition. The interaction
between industrial growth and international trade in Japan has occasionally
been treated in the English-language economics literature (for example, Lock-

Preface xiii
wood, Allen, Akamatsu, Kojima, Shinohara, Tatemoto-Baba, and Ohkawa
and Rosovsky). This author's debt to these studies is great. To these has been
added this author's research, which is based on the recent Long-term Eco­
nomic Statistics of Japan series. This book attempts to present a complete
picture of the development mechanism and government policies in a long-
run historical perspective—this author's version of the Japanese model.
The concept of the catching-up product cycle provides the main analytical
framework for Japanese economic development into which such institutional
devices as industrial policy and trading companies are incorporated. Although
based on Vernon's version of the product cycle, CPC was first proposed by
Akamatsu in the early 1940s and is known to Japanese economists as the
"flying wild geese pattern" of industrial development. The concept was based
on Akamatsu's statistical study of the interrelated development of Japanese
industries and trade. The concept was developed further by Kojima and Shino­
hara. Since introduced by Ohkita at the 4th Pacific Economic Cooperation
Conference (PECC) in Seoul in May 1985, the theory has been attracting the
attention of economists in the Asia-Pacific region as the theory that best ex­
plains industrial transmission among the Asian economies. The phrase "fly­
ing wild geese" has become a popular way to describe economic cooperation
in the Pacific. This book endeavors to present to English-language readers
this Japanese model, with all its implications for the economic development
of late-starting countries.
Long-term Japanese economic statistics are used to the fullest extent to
describe the process of economic development; these are taken from the
coauthored work, Trade and Balance of Payments (Yamazawa and Yamamoto,
Volume 14 of the Long-term Economic Statistics of Japan series) and from
other publications in the series, which are available in English translation and
may be referred to by interested readers.
August 1990

Acknowledgments
I would like to express my sincere thanks to my long-time friend, Dr. Seiji
Naya, Director of the East-West Center's Resource Systems Institute, who en­
couraged me to translate my 1984 edition into English, offered a month's stay
at the Resource Systems Institute to work on the translation, provided me
with editorial assistance, and waited patiently for the completion of my
manuscript. A former graduate student at Hitotsubashi University, Dr. Gwend-
lyn Tecson, helped with my translation. I appreciate greatly the Resource Sys­
tems Institute's editor, Mr. David Puhlick, for his painstaking efforts to make
my English readable and his patience. I also appreciate the help of Dr. Pearl
Imada and Ms. Janis Togashi of the Development Policy Program staff,
Resource Systems Institute, for their assistance in helping to clarify economic
concepts for English-language readers and providing comments that enhanced
the overall presentation. I would also like to thank Ms. Ann Takayesu for
her editorial support in preparing the manuscript for production. The Graphics
and Production Services of the East-West Center designed and typeset this
edition.
The publication of the English-language edition would not have been pos­
sible without the generous permission of Toyo Keizai, Inc., the publisher of
the 1984 Japanese version. I would also like to acknowledge the other sources
and publishers upon which parts of this edition were based. The materials
in this edition, and the publications in which they appeared, are as follows:
Figure 10.2 appeared, in part, as Figure 16.1 in Chapter 16, by I. Yamazawa
and S. Tambunlertchai, Japan and the Developing Economies: A Compara­
tive Analysis, K. Ohkawa and G. Ranis, eds. (Basil Blackwell, Inc., Oxford,
England, 1985). Tables 6.7, 6.8, 10.1, and 10.2 appeared as Tables 18.1, 18.2,
18.3, and 18.5, respectively, in Chapter 18, by I. Yamazawa and H. Kohama,
of the same title. Sections of an earlier version of Chapter 5 of this edition
appeared as "Industry Growth and Foreign Trade: A Study of Japan's Steel

xvi Acknowledgments
Industry?' by I. Yamazawa (Hitotsubashi Journal of Economics, Vol. 12, No.
2, Feb. 1972, Hitotsubashi University, Tokyo), and an earlier, shorter version
of Chapter 7 appeared as "Industrial Growth and Trade Policy in Pre-War
Japan" by I. Yamazawa (The Developing Economies, Vol. 13, No. 1, 1975,
Institute of Developing Economies, Tokyo).
The Japan Foundation provided funding for this edition's publication.
Kunitachi, Tokyo
August 1990 Ippei Yamazawa

Part I
Japan's Economic Development
and International Trade

One
Economic Development and Trade
Structure: The Long View
At the core of Japanese economic development was Japan's rapid industri­
alization and great capacity for change. Japan was late in developing, yet
Japan's domestic production and trade structures underwent a remarkable
change from the period of the introduction of modern industries from the
United States and Europe to the period of export-oriented growth. For a coun­
try lacking in natural resources—except labor—the strategy of importing raw
materials and exporting finished products was most appropriate. Beginning
in the Meiji period (1868-1912), the interaction between industrialization and
changes in trade structure led to over a century of Japanese economic de­
velopment, and although severe balance-of-payments constraints character­
ized Japan's initial period of growth, these gradually eased.
This chapter overviews Japan's development over a century, focusing on
the interaction between industrialization and changes in trade structure.' This
interaction, which is seen through Japan's high capability of transformation,
characterizes Japan's economic development. Chapter 2 presents the interac­
tion as a "Japanese model," which provides the framework for the presenta­
tion of industry case studies in Chapters 3 through 6. The role of government
policy in the process of this successful interaction during the pre-Second World
War period, the high-growth period after the Second World War, and the peri­
od after the two oil shocks is analyzed. Towards the end of this book it is
pointed out, however, that Japan has outgrown this traditional pattern of
catching-up industrialization and now searches for a new pattern of economic
growth that corresponds to its position as a new industrial leader. Chapter
10 discusses the relevance of the Japanese development experience to con­
temporary developing countries.

4 Economic Development and International TYade
ECONOMIC DEVELOPMENT AND FOREIGN TRADE
Commodity trade grew steadily during the process of Japan's economic
development. Columns 1 and 2 in Table 1.1 show changes in total commod­
ity exports and imports. To facilitate observation of such shifts, the pre-Second
World War indices have been expressed in terms of overlapping decade aver­
ages; the post-Second World War indices have been expressed in terms of
Table 1.1
Development of Japan's Foreign Trade
Period Exports
0) Imports
(2) World
Trade
(3) Export/GNP
Ratio (<7o)
(4) Import/GNP
Ratio (%)
(5)
(1882-91 = 100)
1874-83 52.7 64.1 — — —
1877-86 66.2 69.1 — — —
1882-91 100.0 100.0 100.0 — —
1887-96 143.1, 164.3 116.0 3.2 6.0
1892-1901 202.0 275.0 130.6 3.9 8.8
1897-1906 295.7 400.5 159.8 5.2 11.6
1902-11 417.9 487.6 185.7 6.7 12.8
1907-16 643.7 592.8 192:8 9.2 13.9
1912-21 921.2 749.2 192.6 10.6 14.2
1917-26 1,049.1 1,067.9 223.2 10.1 16.9
1922-31 1,297.0 1,375.9 265.2 11.4 19.8
1927-36 2,027.5 1,558.3 251.9 14.9 18.8
1930-39 2,505.1 1,729.0 241.5 16.1 18.2
(1951-55 = 100)
1951-55 100 100 100 3.4 4.9
1956-60 284 229 157 6.0 6.8
1961-65 612 485 226 7.7 8.7
1966-70 1,328 900 360 10.0 9.6
1971-75 2,171 1,138 550 11.0 8.2
1976-80 3,242 1,543 729 13.0 8.7
Notes: The indices of Japan's exports (1), imports (2), and world trade (3) are calculated from
constant price series; the export (4) and import (5) ratios are calculated from constant price
series of exports, imports, and GNP.
Sources: Figures for exports, imports, and world trade are from Yamazawa and Yamamoto
(1979a); for GNP, from Ohkawa, Takamatsu, and Yamamoto (1974).

Economic Development and Trade Structure 5
consecutive five-year averages. Both are expressed in constant prices, using
as a reference the periods 1882-91 for the prewar series and 1951-55 for the
postwar series.
During the 65 years spanning the periods 1874-83 to 1930-39, exports grew
at an annual average rate of 6.8 percent compared to a growth rate of 5.8
percent for imports. Between the periods 1951-55 and 1966-70, exports ex­
panded at an average annual rate of 17.2 percent and imports at 14.8 percent.
These rates were about double the 8.5 percent growth rate of world trade during
the same period. Table 1.2 compares the growth performance of Japan's ex­
ports in the principal export markets of Western Europe, the United States,
and China during the pre-First World War and interwar years (columns 4
to 5). In all periods or markets, Japan's export and import growth rates were
relatively high. Although world trade practically came to a halt or declined
during the interwar years, Japan's export growth continued.
Table 1.1 shows the seesaw-like movements in growth of both exports and
imports. In the periods 1874-83 to 1882-91, exports were leading; in the peri-
Table 1.2
International Comparison of Trade Expansion:
Average Annual Growth Rates (°7o)
Period Number
of
years Exports
by
Japan
(1) Imports
by
Japan
(2) World
Trade
(3) Imports
by
Western
Europe
(4) Imports
by
the U.S.
(5) Imports
by
China
(6)
1874-83 to
1930-39 57 6.77 5.78
1882-91 to
1930-39 49 1.80 1.70 2.59 1.37
1882-91 to
1902-11 20 7.15 7.92 3.09 2.94 3.00 3.04
1922-31 to
1930-39 10 6.58 2.28 -0.94 0.03 -0.72 -3.68
1951-55 to
1966-70 15 17.2 14.8 8.5
1966-70 to
1976-80 10 8.9 3.0 7.1 — — —
Notes: The annual compound rates are calculated by dividing the last period values by those
of ihe first period. The periods 1907-16 to 1917-26 are omilted to avoid the influence of
the First World War (see Table 1.1).
Sources: Table 1.1, (his volume, and Yamazawa and Yamamoto (1979a).

6 Economic Development and International Trade
ods 1887-96 to 1902-11 imports were leading, then exports in the periods
1902-11 to 1912-21, imports in the periods 1912-21 to 1927-36, and exports
in the periods 1927-36 to 1930-39. During the postwar years, however, the
growth rate of exports was consistently higher.
Table 1.1 also shows the movements in commodity exports and imports rela­
tive to GNP. At first the import ratio exceeded the export ratio, but as both
ratios increased during the entire prewar period, exports overtook imports. 2
Reflecting the movements observed in the export and import growth trends,
the import ratio showed a greater increase than did the export ratio during
the periods 1882-91 to 1887-96 but was overtaken by the export ratio in the
periods 1902-11 to 1912-21. Import ratios again increased by greater amounts
during the periods 1912-21 to 1922-31, and export ratios, in the periods
1922-31 to 1930-39. After the Second World War both the export and im­
port ratios were at about the 4-5 percent level, but in the second half of the
1960s the export ratio exceeded the import ratio for the first time.
The rise in the export/GNP and import/GNP ratios followed a general trend
observed in the process of modern economic growth in Western European
countries. According to Kuznets (1967:18-26), Britain's foreign trade ratio
(exports-plus-imports/GNP ratio or X + M/GNP) was 14 percent in the
nineteenth century and rose to 43.5 percent in the period before the First World
War (1909-13). Similarly, France's ratio was 10 percent in the 1820s and rose
to 35.2 percent in the period 1908-10, and Germany's grew from 13 percent
in 1840 to 38.3 percent in the period 1910-13. 1 For Japan, the foreign trade
ratios were 9.2 percent in 1890 and 23.1 percent in the period before the First
World War (1907-16). The ratios rose to 34.3 percent in the period 1930-39.
The initial and final levels of the Japanese ratios were thus similar to those
of the other developed countries. Although the start of modern Japanese eco­
nomic growth came 50 to 100 years after that of other developed countries,
trade growth after the First World War continued, until in the 1930s it was
comparable to the level of trade growth of the developed countries.
Japan, like most developing countries, had a higher GNP growth rate than
the developed countries, but what was unusual was that it had an even higher
growth rate in trade, particularly exports. Export growth was sustained even
in the 1920s and 1930s when other countries' trade either declined or came
to a halt. What role did such rapid trade expansion, especially in exports,
play in Japan's economic development? What are the implications for the
role of trade in the development process as seen in the fluctuation of exports
and imports vis-a-vis GNP? Since these questions cannot be answered mere­
ly through observation of general trends, an analysis must begin by examin­
ing the interaction between changes in the export and import structures and
in domestic production.

Economic Development and Trade Structure 7
CHANGES IN THE STRUCTURE OF TRADE
Commodity Composition of Exports
The rapid growth in commodity export and import values was accompa­
nied by significant changes in commodity composition. Table 1.3 shows the
dramatic changes in Japan's commodity exports that took place over a pe­
riod of approximately one hundred years. All primary goods and specific
primary goods such as raw silk and copper ingots are included in the data.
Note that both of these goods are simply processed from indigenous raw
materials and are similar to the processed primary goods exports of contem­
porary developing countries; they are different from processed goods such
as cotton textiles and pig iron, which depend on imported raw materials. At
the same time, processed exports of indigenous materials are not limited to
these two types of goods. Export goods such as processed food, porcelain
pottery, and wooden and straw products also depend on indigenous raw
materials but are included in the manufactured goods category. The com­
parative advantage of these goods in export markets is at the processing stage,
since they do not depend as much on the decisive factor of raw material en­
dowments as do raw silk and copper ingots. Although any classification will
inevitably be somewhat arbitrary, the special treatment accorded to raw silk
and copper ingots is due to their importance as export products.
Aside from raw silk and copper ingots, other principal primary export com­
modities were tea, marine products, and coal for ships. The share of primary
goods in total commodity exports for the period 1874-83 was more than 80
percent, but, except for raw silk, primary goods rapidly became unimportant
as export commodities. Among the primary goods, raw silk led export growth
and made up one-third of total export value from 1874 to 1896. It was not
until the late 1920s that the share of primary goods fell below one-fourth
of total exports and became significantly less important than light industry
exports such as textiles (particularly cotton fabrics), processed food, and mis­
cellaneous manufactured goods (see Table 1.3). Exports of processed food
and other manufactured goods grew rapidly, and both were important ex­
port commodities until the Second World War, a period of more than 40 years.
The main exports in the heavy industrial category of chemicals, metals, and
machinery were primarily such goods as matches and camphor. Exports of
more sophisticated heavy industrial goods did not begin until the First World
War, after which they grew to replace light industrial goods as principal ex­
ports only in the second half of the 1950s. By the latter half of the 1970s,
the share of heavy industrial goods in total exports was 87 percent, a rather
high percentage, of which two-thirds was represented by machinery exports. 4

8 Economic Development and International Dade
Commodity Composition of Imports
Table 1.4 shows the change in the structure of imports. The most notice­
able change in structure is the decline in the share of industrial goods from
about 90 percent in the period 1874-86 to 50 percent during the First World
War. Another significant change is the rise in the share of imports of primary
products from 60 percent in the pre-Second World War period to 70 percent
Table 1.3
Commodity Composition of Japan's Exports (°/o in market prices)
Primary Goods 3 Industrial Goods b
Periods All
primary
goods
(1) Raw
silk
(2) Copper
ingots
(3) All
industrial
goods
(4) Textiles
(5) Chemicals,
metals,
and
machinery
(6)
1874-83 82.4 37.7 2.2 17.6 4.4 5.9
1877-86 79.4 36.8 3.1 20.6 6.1 6.7
1882-91 74.9 36.8 5.1 25.1 8.8 7.2
1887-96 65.5 34.1 5.1 34.5 14.8 8.3
1892-1901 55.1 29.3 4.8 44.9 23.3 8.2
1897-1906 47.7 26.2 4.9 52.3 27.4 9.0
1902-11 45.2 26.2 4.9 54.8 27.7 12.6
1907-16 41.8 24.6 4.9 58.2 28.9 12.5
1912-21 34.2 22.6 2.6 65.8 33.8 16.7
1917-26 36.5 28.4 0.8 63.5 35.2 14.3
1922-31 38.5 31.7 0 61.5 34.1 12.8
1927-36 27.2 20.5 0 52.8 36.3 19.7
1930-39 19.9 13.1 0 80.1 35.0 26.5
1951-55 4.7 — — 95.3 39.5 39.9
1956-60 4.5 — — 95.5 32.0 45.1
1961-65 3.5 — — 96.5 21.3 58.6
1966-70 1.7 — — 98.3 13.7 71.2
1971-75 1.4 — — 98.5 7.4 82.9
1976-80 0.9 — — 99.1 4.7 87.1
Notes: Column 4 includes processed foodstuffs, miscellaneous products, ceramics, and wood­
en products as well as textiles and chemicals, metals, and machinery.
a. Raw silk and copper ingots are the principal primary goods exports.
b. Textiles as well as chemicals, metals, and machinery are the principal industrial goods exports.
Source: Yamazawa and Yamamoto (1979a: table 1), but adjusted for raw silk. Copper ingots
reclassified as a primary good for the years before the Second World War.

Table 1.4
Commodity Composition of Japan's Imports (Vo)
Industrial goods Primary goods
Chemicals, Un­Raw materials
Total Industrial metals, Total processed Total Textile Metal
(2+3) goods machinery (5+6) food (7-9) materials ores. Fuels
Periods (1) (2) (3) (4) (5) (6) (7) (8) (9)
1874-83 91.2 69.9 21.3 8.8 0.7 8.1 0.7 — 5.0
1877-86 89.7 68.4 21.3 10.3 0.8 9.5 1.6 — 6.1
1882-91 81.3 54.7 26.6 18.7 5.0 13.7 5.8 • — 6.4
1887-96 71.8 42.8 29.0 28.2 7.1 21.1 14.8 — 4.9
1892-1901 63.6 31.0 32.6 36.5 9.9 26.6 20.8 0.1 4.4
1897-1906 56.9 24.1 32.8 43.1 13.8 29.3 22.9 0.1 4.6
1902-11 54.8 20.5 32.3 46.2 12.5 32.7 25.9 0.2 3.9
1907-16 50.0 15.6 34.4 50.0 10.3 39.7 32.6 0.7 2.7
1912-21 47.4 12.1 35.3 52.6 12.5 40.1 32.4 1.0 2.2
1917-26 45.7 14.9 30.8 54.3 16.1 38.2 29.5 0.8 2.9
1922-31 43.4 17.1 26.3 56.6 18.8 37.8 27.1 0.8 4.3
1927-36 39.0 13.8 25.2 61.0 19.0 42.0 28.6 1.4 5.9
1930-39 42.0 12.3 29.7 58.0 17.5 40.5 25.1 2.6 7.4
1951-55 14.4 4.0 10.4 85.6 25.0 60.6 27.6 6:8 11.0
1956-60 23.3 3.6 19.7 76.7 13.2 63.5 19.3 13.8 15.7
1961-65 27.7 5.5 22.2 72.3 13.5 58.8 12.6 13.0 18.3
1966-70 30.3 7.0 23.3 69.7 12.8 56.9 6.9 13.6 20.4
1971-75 27.2 9.5 17.7 72.8 14.7 58.1 3.8 9.2 33.9
1976-80 25.2 8.9 16.3 75.5 14.3 61.2 2.3 6.4 44.2
Notes: Percentage composition calculated from overlapping decade averages at current prices. (1) + (4) = 100; (2) + (3) = (1); (5) + (6) = (4);
but (7) + (8) + (9) < (6), the difference consisting of other agricultural materials, forest products, and other mineral materials.
Source: Yamazawa and Yamamoto (1979a: table 2).

10 Economic Development and International Trade
in the postwar era. Thus, over the past century, Japan has undergone a com­
plete transition from an industrial goods-importing country to a primary
product-importing country. What is important, however, is the change in the
composition of this industrial goods and primary products trade. Light in­
dustrial goods represented 70 percent of total imports in the period 1874-83
but rapidly declined thereafter to about 10 percent in the interwar period.
In contrast, heavy industrial goods imports rose in importance through the
First World War, reaching a maximum share of 35 percent of total imports
and falling to about a 30 percent share by the start of the Second World War
period.
Figure 1.1 shows the change in composition of industrial goods imports
classified according to end use. Beginning with the period 1874-83 consumer
goods imports represented close to 50 percent of total imports but rapidly
declined to about 10 percent by the interwar period. This declining share was
initially replaced by investment goods (producer durables and construction
materials), which leveled off at 20-25 percent, and later by intermediate goods
imports.
Figure 1.1
Compositional Changes in Imports of Industrial Goods by End Use
1877-86 1887-96 1897-1906 1907-16 1917-26 1927-36
80

% 40
20 \- Consumer
goods
Producer durables and
construction materials
I i l i
1874-83 1882-91 1892-1901 1902-11 1912-21 1922-31 1930-39
Periods
Note: Composition was calculated from the overlapping decade average at current prices.
Source: Yamazawa and Yamamoto (1979a: table 6).

Economic Development and Trade Structure 11
Although consumption goods do not correspond exactly to the light in­
dustrial goods category, import substitution in the manufacturing sector
brought about growth in essential investment goods imports (mostly heavy
industrial goods). The leveling off of the shares of heavy industrial and in­
vestment goods imports reflects the import substitution that was simultane­
ously under way in these industrial sectors.
The growth in primary goods imports was mainly in raw materials. Due
to the expansion in textile production in the pre-Second World War period,
imports of textile raw materials made up three-quarters of total raw material
imports, or one-third to one-fourth of total imports. With the development
of iron and steel and other heavy industrial goods production, however, the
shares of iron ore, coal, and other mineral product imports began to grow
in the 1930s and became the principal raw material imports from the second
half of the 1950s. The rapid growth of petroleum imports during the post-
Second World War period is particularly evident. In the 1970s, the price of
crude oil skyrocketed, and the share of petroleum imports rose to 44 percent
of total fuel imports. Because of the increase in total fuel imports, the share
of industrial goods declined to about 25 percent of total imports. The other
import category that rose in importance during the post-Second World War
period was forestry and other agricultural raw material imports (Table 1.1),
which together accounted for 13 to 14 percent of total imports, thus demon­
strating the trend toward diversification in raw material imports.
The share of basic food imports rose from less than 1 percent in the period
1874-83 to 16-19 percent during the interwar period. Even here, the change
in composition of food imports over these periods of time should be noted.
The agricultural food imports of the Meiji period fluctuated from year to
year. During years of bad harvests and wars (that is, 1879-80, 1890, 1897-98,
1903-13), food imports grew to several times the amounts of the preceding
and following years. As a result, the share of food imports during normal
years fell far below those shown in Table 1.4. At one point Japan became
dependent on Taiwan and Korea for imports of agricultural food products.
But in 1912, with the establishment of tariffs on rice and wheat, imports
declined. After the First World War, however, agricultural food imports gradu­
ally stabilized and represented 16-19 percent of total imports during normal
crop years. From 1955 through 1980 the share of food imports in total im­
ports also increased, from about 13 to just above 14 percent, but for the most
part this slightly increasing trend reflected import dependence on animal feeds
and cereals other than rice.
Geographical Composition
In addition to changes in the composition of trade, there was also a shift
in trading partners. Trade began with Western Europe, but East Asia and North

12 Economic Development and International TYade
America gradually became the principal trading regions. At the same time,
trading regions expanded, first to Southeast and South Asia and the Middle
East, and then, after the 1920s, to Oceania, Central and South America, and
Africa. Even during the post-Second World War years, the trend toward trade
diversification continued, beginning with the decline in importance of East
Asia, and, in turn, of Southeast and South Asia and the Middle East; however,
the high import shares for North America continued.
Behind these changes in the geographical origins and destinations of ex­
ports and imports, a shift in commodity composition is discernable. The first
industrial goods imports came from Europe, and as a result of an increase
in the volume of raw material imports, diversification of raw material im­
ports took place. Food imports were originally shipped from East Asia in
the early 1900s, especially from Japan's colonies. Raw silk was exported to
Europe and North America, and other textile products found export markets
in Korea, Taiwan, and China, moving gradually to Southeast Asia and to far-
flung markets such as South Asia, the Middle East, Africa, Central and South
America, and Oceania. Heavy industrial goods, such as iron, steel, and am­
monium sulfate, were first exported in the pre-Second World War years to
the East Asian colonies and finally to every region of the world during the
post-Second World War period. 5
Colonies such as Taiwan and Korea are usually assumed to have been sources
of food and raw material supplies as well as export markets for Japan during
the pre-Second World War period, but such a view is not necessarily accurate.
At first glance, this assumption fits the British colonial trade pattern and may
be considered natural for Japan, which is also a country lacking in raw materi­
als and space. However, although Taiwan and Korea were the major supply
sources of food imports, they were not Japan's primary sources of raw material
supplies (only 7 to 9% of which was made up of mineral products). Moreover,
although these two countries were important export markets for Japan, ab­
sorbing about 40 percent of its heavy industrial goods exports, the export/
production ratios of chemicals, metals, and machinery were only about 11
to 12 percent—hardly enough to represent the core of pre-Second World War
Japan's export-led industrialization. 6
CHANGES IN THE STRUCTURE
OF INDUSTRIAL PRODUCTION AND TRADE
The shift in the long-term trend of industrial production is demonstrated
in Shinohara's overlapping decade average series of industrial production
values at constant prices (Shinohara 1972: chap. 2). From the 1870s to the
1930s, the share of the modern sectors—such as textiles, chemicals, ceram­
ics, nonferrous metals, iron and steel, machinery, and printing—grew and the

Economic Development and Trade Structure 13
share of the traditional sectors—processed food and wood and miscellane­
ous products—declined. However, domestic prices in the modern sectors
noticeably declined relative to those in the traditional sectors. The data for
the relevant years show only small changes in shares, because changing rela­
tive prices tended to offset part of the changes in real shares. In the relation­
ship between the expansion of production shares of the modern industrial
structure and the decline in relative prices, Shinohara saw proof of Japan's
success in achieving comparative advantage in these industries and in import
substitution as well as in the export expansion of its products.
Originally termed the "flying wild geese pattern" of industrial develop­
ment by Akamatsu (1943) and later renamed the "catching-up product cy­
cle" (CPC) by Kojima (1973), the processes of import substitution and
subsequent export expansion by the modern industrial sector foreshadow simi­
lar developments in major modern industries. These processes provide the
basic analytical framework for the following chapters, which begin with an
examination of the correspondence of structural changes between industrial
output, exports, and imports.
Compositional Changes in
Industrial Output, Imports, and Exports
Tables 1.5-1.7 show the trend in composition of industrial output, imports,
and exports. For the pre-Second World War period, the overlapping decade
average series have been calculated from Shinohara's analysis of the same
13 periods (1874-83 to 1930-39) and the five-year average series for the post-
Second World War years (1951-55 to 1976-80). The computed distribution
is based on 1934-36 prewar and 1965 postwar price series, because changes
in composition based on current year price data might be distorted by changes
in relative prices.
As shown in Table 1.5, changes are evident in industrial output in terms
of the declining shares of processed food and wood products throughout the
entire series, and the declining share of miscellaneous products up to the Sec­
ond World War. These products were replaced by textiles whose share increased
up to 1939. In addition, during the periods 1874-83 through 1907-16 growth
begins in the heavy industrial sector in such products as metals and machinery
(and later chemicals), and a basic trend emerges in the period 1922-31. Ex­
cept for the period 1951-55, there is a firming of the trend toward share ex­
pansion by the metals, machinery, and chemicals sectors during the postwar
period. Among the heavy industrial goods, the expansion of machinery is
the most remarkable. While growth problems were encountered in chemicals
and metal products during the second half of the 1970s, the assembly indus­
tries of the machinery group continued to grow.

Table 1.5
Composition of Industrial Output (Vo)
Period Processed
food Textiles Wood
products Chemicals Ceramics Metals Machinery Miscellaneous
products
1874-83 58.5 10.8 6.3 11.0 1.9 1.4 1.3 8.8
1877-86 56.6 12.5 5.9 11.0 1.5 1.6 1.5 9.3
1882-91 53.4 17.9 4.7 9.5 1.3 1.8 1.4 9.8
1887-96 48.6 24.6 3.7 8.7 1.5 1.7 1.5 9.7
1892-1901 47.3 26.5 3.7 8.6 1.5 1.4 2.2 8.7
1897-1906 45.9 25.1 4.1 9.2 1.6 1.7 3.7 8.7
1902-11 42.1 25.2 3.7 9.7 2.0 2.6 5.6 9.1
1907-16 37.2 27.3 3.1 9.6 2.2 4.3 7.7 8.7
1912-21 32.0 27.6 2.6 9.2 2.1 6.8 12.5 7.7
1917-26 30.9 28.3 2.4 9.2 2.4 8.0 12.2 6.9
1922-31 28.2 29.8 2.6 10.7 2.7 9.0 10.5 6.6
1927-36 21.0 30.4 2.8 13.2 2.6 11.4 12.6 6.2
1930-39 17.3 28.1 2.8 15.2 2.5 13.0 15.5 5.7
1951-55 23.2 14.0 14.5 10.0 4.6 17.6 12.2 3.9
1956-60 18.0 12.3 9.0 13.5 4.2 17.4 19.7 5.9
1961-65 13.6 9.2 5.7 15.9 3.8 17.6 26.4 7.8
1966-70 10.9 7.4 3.9 15.7 3.4 17.9 32.9 7.9
1971-75 8.8 5.9 2.7 15.9 3.0 19.6 36.5 7.6
1976-80 8.1 4.9 2.1 15.4 2.8 18.5 40.2 7.4
Note: Computed from constant price series (base years: 1934-36 (prewar); 1965 (postwar)).
Source: Yamazawa and Yamamoto (1979a).

Table 1.6
Composition of Industrial Goods Imports (%)
Period Processed
food Textiles Wood
products Chemicals Ceramics Metals Machinery Miscellaneous
products
1874-83 6.1 65.7 0.1 5.4 1.4 6.0 4.9 10.4
1877-86 7.2 63.0 0.1 5.2 1.4 7.4 7.5 8.2
1882-91 7.8 53.2 0.1 6.0 1.8 9.8 12.9 8.4
1887-96 7.9 44.4 0.0 7.9 1.7 11.9 18.1 8.1
1892-1901 10.2 29.2 0.0 11.9 1.6 13.3 28.1 5.7
1897-1906 10.7 20.2 0.1 15.8 1.9 14.7 32.7 3.9
1902-11 9.6 15.8 0.1 20.3 2.3 17.0 31.5 3.4
1907-16 10.0 11.0 0.1 23.6 2.0 20.2 30.1 3.0
1912-21 11.8 6.8 0.4 25.7 0.9 25.1 27.4 1.9
1917-26 13.0 9.3 2.3 24.4 0.8 23.8 24.8 1.6
1922-31 15.4 10.1 3.6 24.4 1.3 21.3 22.0 1.9
1927-36 18.8 7.5 2.8 27.1 1.5 24.7 15.7 1.9
1930-39 17.4 5.7 1.3 25.1 1.2 34.2 13.5 1.6
1951-55 12.3 5.6 6.0 17.6 1.5 8.9 45.4 2.7
1956-60 6.5 2.4 3.2 21.3 1.6 20.6 41.7 2.7
1961-65 9.7 1.9 3.4 20.4 1.9 16.7 42.5 3.5
1966-70 7.5 3.8 3.3 19.7 3.2 19.8 36.5 6.2
1971-75 4.9 8.3 4.8 18.1 5.8 .12.0 36.9 9.2
1976-80 8.6 8.3 5.5 20.0 3.6 9.3 36.2 9.7
Note: Computed from constant price series (base years: 1934-36 (prewar); 1965 (postwar)).
Source: Yamazawa and Yamamoto (1979a).

Table 1.7
Composition of Industrial Goods Exports (<Vb)
Period Processed
food Textiles Wood
products Chemicals Ceramics Metals Machinery Miscellaneous
products
1874-83 5.9 59.8. 0.3 22.0 1.4 7.2 0.0 3.4
1877-86 5.3 56.0 0.3 22.8 1.6 9.5 0.0 4.5
1882-91 4.0 56.4 0.3 20.3 2.0 11.9 0.0 5.1
1887-96 3.2 59.2 0.5 17.1 2.0 10.9 0.2 6.9
1892-1901 3.6 61.3 1.2 15.3 1.9 7.9 0.3 8.5
1897-1906 6.2 59.2 2.3 13.6 2.3 6.4 0.8 9.2
1902-11 7.2 57.4 3.5 12.3 2.4 6.3 1.6 9.3
1907-16 6.8 56.2 3.0 11.9 2.3 6.4 2.8 10.6
1912-21 7.1 55.1 2.3 11.3 2.4 6.6 5.3 9.9
1917-26 7.1 60.1 1.8 9.9 2.6 5.7 4.5 8.3
1922-31 7.4 62.9 1;3 9.6 3.0 5.2 3.4 7.2
1927-36 7.1 57.4 1.2 10.2 3.3 7.2 6.3 7.3
1930-39 7.1 52.1 1.7 10.8 3.4 7.7 9.5 7.7
1951-55 6.0 41.4 2.0 4.9 5.7 20.6 13.2 6.2
1956-60 6.2 36.1 2.2 6.5 4.5 12.7 23.4 8.4
1961-65 3.6 21.8 1.5 10.1 3.8 18.2 32.6 8.4
1966-70 2.4 14.1 0.8 12.7 2.5 18.1 42.6 6.8
1971-75 1.2 8.6 0.5 13.0 1.4 17.9 53.2 4.2
1976-80 0.7 5.3 0.2 9.8 1.2 14.4 64.9 3.3
Note: Computed from constant price series (base years: 1934-36 (prewar); 1965 (postwar)).
Source: Yamazawa and Yamamoto (1979a).

Economic Development and Trade Structure 17
The change in import structure is even more striking (see Table 1.6). The
66 percent share of textile imports in the period 1874-83 declined drastically
to 11 percent by the period 1907-16 and to 2 percent by the period 1961-65.
Although its share was smaller, miscellaneous products followed a similar trend
up until the Second World War. On the other hand, the shares of three
sectors—chemicals, metals, and machinery—grew rapidly, cumulatively reach­
ing 70 percent in the period 1907-16 and leveling off thereafter until the post-
First World War years. Processed food, wood products, and ceramics showed
a slow but steadily rising trend.
The change in the export structure before the First World War is not as
remarkable as the changes that occurred in the import and output structures
(see Table 1.7). Up to the period 1922-31, the structure of industrial goods
exports changed relatively little. Textiles maintained its share between 50 and
60 percent; the share of chemicals showed a steady decline. After surging in
the periods 1877-86, 1882-91, and 1887-96, the share of metals dropped dra­
matically in the period 1892-1901 and did not regain its 1874-83 share until
the period 1927-36. The shares of wood products, processed food, and mis­
cellaneous products gradually rose. However, a significant change took place
in textiles—the share of raw silk fell during the periods 1874-83 to 1930-39
from 38.3 percent to 12.1 percent of total industrial exports (raw silk is in­
cluded in industrial exports in Table 1.7), but that of other textile products
(mainly cotton fabrics) rose from 21.5 percent to 40.0 percent.
Beginning in the period 1922-31 and continuing through the post-Second
World War years, the combined share of chemicals, metals, and machinery
exports rose and eventually replaced that of textiles in the period 1956-60
(see Table 1.7). This change occurred as the share of textiles declined from
63 percent (1922-31) to 52 percent (1930-39) to 14 percent (1966-70), while
the share of chemicals, metals, and machinery rose from 18 percent to 28 per­
cent to 73 percent in the same periods. The decline in the share of textiles
from the periods 1922-31 to 1930-39 was primarily due to the decline in the
share of raw silk (the remaining share of textile products plunged during the
post-Second World War years). Likewise, miscellaneous products, processed
food, and wood products held a 20 percent share in the periods 1902-11 to
1912-21; this was halved to 10 percent by the period 1966-70.
It is inaccurate to assert that import substitution and export expansion
proceeded in all modern industrial sectors before the Second World War by
merely comparing the situations in the periods 1874-83 and 1930-39. Whereas
import substitution was completed for textile products by the period 1907-16,
for chemicals, metals, and machinery import substitution occurred after the
period 1912-21. Although chemicals, metals, and machinery maintained a
combined share of 70 percent of total imports in the postwar period, it should
be noted that the share of industrial goods sharply declined as a percentage

18 Economic Development and International Trade
of total imports. As Table 1.6 shows, after the period 1956-60 the shares of
other industrial goods imports such as textiles, processed foods, and miscel­
laneous products began to rise.
Beginning in the period 1882-91, the share of other textile products (ex­
cept for raw silk) grew, rising to 40 percent in the period 1930-39. Again,
this share was displaced in the postwar period by those of chemicals, metals,
and machinery combined, whose shares began to grow after the period
1922-31. Within this group, the growth in machinery exports, beginning in
the 1970s, was indeed remarkable. Thus, although occurring at different times,
expansion in the shares of light industrial goods (particularly textile prod­
ucts—except raw silk) and heavy industrial goods (such as chemicals, metals,
and machinery) constituted the two largest shifts in structure, each following
the sequence of imports, production, and exports. Kojima referred to this
as "import-led structural change" (1958: chap. 3).
Interaction between Production and Trade
The change from import substitution to export-led growth should be con­
sidered not in terms of composition but, rather, in terms of the relationship
between exports and imports, domestic production, and domestic demand.
Figure 1.2 shows the change in import dependence—that is, the ratio of im­
ports to domestic demand—and the change in the export/production ratio
of textile products (except raw silk) as well as chemicals, metals (except cop­
per ingots), and machinery. As is shown in Chapter 2 (see Figure 2.1 (B)),
the decline in import dependence signals the advent of import substitution,
and the rise in the export/production ratio indicates development of the ex­
port stage. When the export/production ratio exceeds the import/demand ra­
tio, the sector moves to the position of being a net exporter; this is the
demarcation point between the import substitution and export stages (see
Figure 2.1). Using this standard, the two graphs in Figure 1.2 show that im­
port substitution in textile products progressed during the periods 1874-83
to 1892-1901, and the export stage was reached by the period 1892-1901. In
the other industries—chemicals, metals, and machinery—the import/demand
ratio continued to increase, reached a plateau during the periods 1892-1901
to 1902-11, and rapidly declined thereafter. It was not until the post-Second
World War years that the export/production ratio was greater than the im­
port/demand ratio in these industries. Thus, it appears that although import
substitution in heavy industrial goods grew from 1900 to the Second World
War, the export stage emerged only after the Second World War.
However, the export/production ratio of machinery alone has tended to
outstrip those of chemicals and metals in the 1970s. In the periods 1971-75
and 1976-80, the export/production ratio for machinery alone was 19.0 and

Economic Development and Trade Structure 19
24.6, respectively, whereas for chemicals and metals as a group the ex­
port/production ratio was 9.6 and 8.0 in the same periods (Yamazawa and
Yamamoto 1979a). This reflects the shift in Japan's comparative advantage
after the first oil shock in 1973 from the basic materials-producing sectors
to the processing-assembly sectors. Moreover, even within the machinery group
the emphasis shifted away from the highly labor-intensive industries such as
shipbuilding and radio assembly to the technology-intensive industries such
as production of automobiles, video cassette recorders (VCRs), and indus­
trial machinery. Eventually, this might be referred to as the third stage of struc­
tural change.
Figure 1.2
Import Substitution and Export Expansion of Two Major Industries
40
30 1882-91 Periods
1902-11 1922-31 1956-60 1976-80
% 20 T—'— 1—'—r
Export/
production ratio —i | 1 r 1 1 i i r
Textiles (excluding raw silk)

% 20 -] r 1 1 1 1 1 1 1 1 1 1 f T r—t ••-
Chemicals, metal and machinery
Export/ *
production, ratio jr

J i i_
1874-83 1892-1901 1912-21 1930-39
Periods 1966-70
Note: Overlapping decade averages.
Sources: Calculated by the author from Yamazawa and Yamamoto (1979a: tables 1-13, 1-15, 1-17)
and Shinohara (1972: table 1).

20 Economic Development and International Trade
The import/demand and export/production ratios of such sectors as
processed food and wood products have remained low. These are sectors that
strongly depend on indigenous raw materials and for which domestic demand
absorbs the major portion of domestic production. To a certain extent, this
could also be said of ceramics and miscellaneous products. The trend in the
import/demand and export/production ratios for all industrial goods is the
aggregate of the different trends in these three product categories (light and
heavy industrial goods, food and wood products, etc.), but the trend becomes
ambiguous and is generally lower, a reflection of the overwhelming share of
food-processing.
Import substitution and export expansion did not occur at once over the
entire range of modern industries. Import substitution began with light in­
dustrial products, such as textiles, and was gradually expanded from one in­
dustry to the next to include heavy industries such as chemicals, metals, and
machinery. Successful import substitution was then followed by export ex­
pansion. Fundamentally, the change in Japanese trade can be summarized
as the movement from import substitution to export expansion, first in the
light industries, then in the heavy industries. Thus, changes in industrial
production follow a predictable pattern.
Three stages in Japan's pattern of trade can be identified. The first stage
consists of the export of primary products and the import of light industrial
goods. The second consists of the export of light industrial goods and the
import of heavy industrial goods as well as raw materials. The export of heavy
industrial goods and import of raw materials characterize the third stage.
In the first stage, as a result of primary product exports, import substitu­
tion of light industrial goods could proceed. In the second stage, as light in­
dustrial goods are successfully exported, import substitution in heavy
industrial goods takes place. In the third stage, the export of heavy industrial
goods is achieved. There is some overlap between stages, and in every stage,
a fundamental and discernible pattern of trade emerges.
In attempting to delineate these stages, the point in time marking the tran­
sition between stages one and two could be set at 1895 and that between stages
two and three at 1955. This can be seen in Table 1.4, where in the first half
of the 1890s the import share of light industrial goods and the export share
of primary products was large (see Table 1.3). Then, around 1895, the ex­
port/production ratio of textile products began to exceed their import/de­
mand ratio, delineating the start of stage two. The early 1950s saw the advent
of stage three, when raw material imports exceeded heavy industrial goods
imports and claimed a large share of total imports. During the first half of
the 1950s, the export share of light industrial goods was about equal to that
of heavy industrial goods. However, the export/production ratios of chemi­
cals, metals, and machinery finally exceeded their import/demand ratios in
the second half of the 1950s. 7

Economic Development and Trade Structure 21
INDUSTRIAL EXPORT GROWTH
AND THE TRADE BALANCE
The relationship between changes in the industrial output structure and
changes in the trade structure has been demonstrated, and the mechanism
(the catching-up product cycle) through which these large structural changes
occurred has been highlighted. It has been shown that there is a 30- to 40-year
lag between light and heavy industrialization. To summarize, there were shifts
over time in four variables—output, exports, imports, and the raw material
imports that were inputs to industrial production. Figure 1.3 shows the trends
for these four variables in constant 1934-36 prices. Because the data are plotted
on a semi-log graph, the slopes of the lines can be interpreted as growth rates.
Except for small undulations, the monotonic growth of industrial output
during the period up to the Second World War is obvious. The trend con­
tinues in the post-Second World War period and exhibits an even steeper rate
of increase. In contrast, increases in raw material imports initially exceeded
that of output, and after reaching a tenth of output value in the 1900s, changes
in raw material imports gradually began to synchronize with output. This
trend continued into the post-Second World War period. Industrial imports
also grew faster than output until around the first decade of this century,
when the rate of growth settled into a slow yet steadily rising trend that, ex­
cept for a spurt of growth after the First World War, has been maintained
in the postwar period. Moreover, during the early years industrial exports ini­
tially grew at the same rate as imports. After 1925, industrial exports were
higher than industrial imports. In the periods both before and after the Sec­
ond World War, growth of industrial exports was higher than growth of in­
dustrial output.
The relationship between the long-run trend in industrial production and
raw material imports is important, as the latter constitutes indispensable in­
puts to the former. The elasticity of industrial output with respect to raw
material imports was 2.6 in the period before the First World War. The elastic­
ity dropped to 0.7 during the interwar period and rose to 0.9 in the period
after the Second World War. 8
The abnormally high elasticity for raw material imports in the period
1874-83 may be explained by the shift from domestic to imported raw material
inputs. For instance, in this period the use of domestic cotton in cotton produc­
tion failed, and a shift to dependence on imported raw cotton took place.
This was also the case for iron ore and coal in the decade after the turn of
the century. The low value of imported inputs in the interwar period reflects
the difficulty of importing during the second half of the 1930s, which re­
sulted from the mounting tensions between Japan and the United States. Af­
ter the Second World War elasticity remained close to 1.0; thus, raw material

Figure 1.3
Industrial Output, Exports, Imports, and Raw Material Imports

1870 1880 1890 1900 1910 1920 1930 1950 1960 1970 1980
Year
Noies: In seven-year moving average. Raw material imports are plotted at ten times the scale
of industrial output, industrial imports, and industrial exports.
Sources: Yamazawa and Yamamoto (1979a: tables 1-3, 1-4) and Shinohara (1972: table 2).

Economic Development and Trade Structure 23
imports grew in proportion to the increase in production, which is to be ex­
pected under the assumption of fixed commodity composition and techno­
logical coefficients. In reality, however, large shifts were observed in both
commodity composition and technological coefficients, and their influences
must have offset each other as the trend seems to indicate. After the two oil
shocks of the 1970s, the degree of output responsiveness declined due to the
decrease in the number of raw material-consuming industries and the tech­
nological changes that led to conservation of raw material inputs.
The movements in industrial imports, exports, and raw material imports
relative to industrial output can be analyzed to determine the trend of the
trade balance. Figure 1.4 shows the seven-year moving average of the ratios
of industrial exports to the sum of industrial commodity and raw material
imports, as well as the ratios of current account receipts to payments (a five-
year moving average is used for the post-Second World War period). Since
the former is directly related to the trade balance, it is simply referred to as
the industrial trade balance. When the ratio is equal to 1.0, trade is said to
Figure 1.4
Long-Term Swings in the Balance of Payments
1.5
1.0
0.5
1870 1880 1890 1900 1910 1920
Year 1930 1950 1960 1970 1980
Notes: A represents the ratio of current values of industrial exports to the sum of industrial
imports plus raw material imports. B represents the ratio of total receipts to total payments
in the current accounts. Both series are smoothed by seven-year moving averages (five-year
for post-Second World War years).
Source: Yamazawa (I984).

24 Economic Development and International Trade
be in balance, and when it is greater (less) than 1.0 there is a surplus (deficit).
In the periods both before and after the Second World War, the industrial
trade balance, although exhibiting cyclical behavior, nevertheless showed a
strongly rising trend overall (that is, an improvement in the balance of trade).
This long-run rising trend corresponds to the upward trend in the long
swings of the industrial output growth cycle. Thus, the growth rate of indus­
trial output accelerated as the structure was shifting from light to heavy in­
dustrialization. Meanwhile, the growth in raw material imports, although
undergoing similar undulations, was decelerating. Moreover, the growth in
imports of industrial goods—except for the import rush of the First World
War—was also weakening as import substitution was being achieved in tex­
tiles and in heavy industrial goods. The result was a steady improvement in
the industrial trade balance. From a trade deficit in most of the preceding
period, the trade balance ratio finally exceeded 1.0 around 1930; the same
scenario repeated itself in the post-Second World War period, so that in the
latter half of the 1960s continuing surpluses resulted. Whereas the problem
of international trade deficits plagues the industrialization drives of contem­
porary developing countries, industrial growth seems to have been self-
reinforcing for Japan.
The critical components in Japan's current account balance are exports
and imports of industrial goods and imports of raw materials, although trade
in services and primary goods other than imports of raw materials are also
included. (The industrial goods trade balance outweighs by far the services
trade balance.) Trade balances in other primary goods as well as the trade
in services balances showed surpluses until 1925 and from the Second World
War until 1963, but from 1925 to the Second World War and after 1963 they
posted deficits. Between 1925 and the Second World War, the deficits were
caused by problems in the export of primary goods and increases in food
imports. In the case of the post-1963 deficit, the trade in services deficit was
responsible. These can thus be considered changes in trends rather than cy­
clical changes.
Japan's current account balance shifted from a deficit to a surplus posi­
tion as its economy developed. Its long-term capital balance tended to offset
the current account imbalance, resulting in an equilibrium of its basic balance
(current account balance plus long-term capital balance). The long-term cap­
ital inflows of the period 1895-1914 were partly encouraged by the govern­
ment to finance the persistent current account deficit during the period. But
to continue long-term capital inflows (issues of bonds by local governments
and public enterprises), the government would have to tighten its money sup­
ply (under the gold standard) and decelerate the heavy industrialization. On
the other hand, long-term capital outflows were large (1) during the First World
War, (2) after the end of the 1920s, and (3) after 1965. The first and third

Economic Development and Trade Structure 25
outflows came from large current account surpluses and did not cause any
payment difficulties. The second outflow (investment in Manchuria in the
1920s) exceeded the current account surplus and led to deficits in the basic
balance, but since investment was undertaken within the yen bloc, it did not
lead to balance-of-payments constraints.
CHARACTERISTICS OF JAPANESE ECONOMIC
DEVELOPMENT AND INTERNATIONAL TRADE
With regard to the long-term trends in Japanese economic development
and trade growth, the following characteristics have been observed:
1. Rapid industrial growth lay at the core of Japan's economic development
and was accomplished by an even greater expansion in exports and im­
ports, reflecting Japan's resource endowment conditions.
2. The industrial growth process illustrates the CPC model of development,
wherein several modern industries introduced from abroad first achieved
import substitution and later developed into export industries. This can
be considered a suitable development strategy for later-developing countries.
3. As a result of CPC development in individual industries, remarkable struc­
tural changes in the economy took place in terms of the domestic output
structure and the export and import structures.
4. The sequencing of CPC development from light to heavy industries made
possible the long-term positive trend—that is, the accumulation of
surpluses—in the trade balance, thereby avoiding the balance-of-payments
constraints in the latter stages of the development process.
Chapter 2 examines the mechanism behind each of these characteristics
more closely and presents a Japanese model of economic development and
trade.

I
Two
Interaction Between Trade
and Development: A Japanese Model
The previous chapter overviewed the long-term trends of Japan's economic
development and identified some characteristics of industrial development
and the expansion of international trade. Unlike the one-way cause and ef­
fect often assumed in international economics textbooks, a two-way interac­
tion has been observed between the two—that is, a close association between
changes in the industrial structure and changes in the trade structure, each
promoting the other in the process of economic development.
Long-term industrial development was successfully achieved in Japan
through the interaction of trade and production. This interaction between
trade and production forms the basis of what can be called a Japanese model
of development. However, this interaction is not unique to Japan; thus, the
development strategy adopted by the Japanese may be regarded as a suitable
development strategy for newly industrializing economies (NIEs) today. The
principal task of this book is to describe the Japanese model in detail and
clarify its underlying mechanism. The catching-up product cycle (CPC) model
describes industrial development through the interaction of trade and produc­
tion. The chapters that follow exemplify the CPC model in a few major in­
dustries and clarify the roles played by firms and the government. This chapter
provides a schematic representation of the CPC model and substantiates its
use in the chapters that follow. In the latter part of this chapter, the model
is extended to analyze the interaction of production and trade at the aggregate
level.
CATCHING-UP PRODUCT CYCLE: THE BASIC TYPE
For countries that began their industrialization later than the United King­
dom, France, the United States, Germany and Russia (hereafter referred to

28 Economic Development and International Trade
as late-starting countries), the development of modern industry typically be­
gan with the import of a new product from more advanced countries, fol­
lowed by import-substituting production, and finally progressed to production
for export abroad. The growth sequence of imports, production, and exports
was first highlighted by Kaname Akamatsu (1943 and 1961) in his statistical
study of trade and production of a few modern industries in Japan before
the Second World War. It was Akamatsu who first used the term flock for­
mation of flying wild geese pattern of industrial development to describe the
shape of import, production, and export growth curves and who pointed out
that the pattern was typical of industrial development in a NIE, particularly
Japan.'
This pattern has been empirically substantiated in many major manufac­
turing industries in Japan and has often been cited in studies of Japan's eco­
nomic growth (Kojima 1958; Shinohara 1962; Baba and Tatemoto 1968;
Yamazawa 1972, 1984). Akamatsu's thesis was first translated into English
in 1956 and was introduced to Western Europe by Sautter (1973). Kojima (1973)
renamed the pattern the catching-up product cycle (CPC) after its associa­
tion with the product cycle model of Vernon (1964). The new name conveys
more precisely the mechanism of industrial development that has been adopted
in this study, but the original name has remained popular. Yamazawa (1984),
Pasha (1987), and Chen (1989) have further extended the pattern to include
the transfer of modern industries to East and Southeast Asian countries. The
pattern has also been frequently cited in trade policy discussions on Pacific
economic cooperation (Kojima 1973; Ohkita 1987; Yamazawa 1988b).
CPC—that is, the sequence of imports, production, and exports—may at
first appear to be self-evident and require no further investigation. But there
are many questions about the CPC process that need to be answered. For
example, why does the CPC process start earlier in one industry than in
another? Why do certain industries succeed in import substitution and ex­
port promotion more quickly than others? Why do some industries fail to
follow CPC? Under what conditions is the growth of an industry preceded
by imports and followed by exports, and what factors determine the speed
of import substitution and export expansion? The answers to these questions
are not self-evident and require theoretical explanation with empirical verifi­
cation as is attempted in this book.
Two extensions to Akamatsu's original CPC model are developed to ex­
plain the new problems arising after the Second World War. First, with many
Japanese industries reaching the export stage after the Second World War,
there was concern over what stage should come next. International special­
ization grew beyond the confines of commodity trade to include technology
transfer and direct foreign investment (DFI). Second, the CPC model has
spread to East and Southeast Asian countries, enabling their rapid industri-

Interaction Between Trade and Development 29
alization over the past two decades. This chapter also attempts to incorporate
these new situations into the schematic outline of the CPC model.
Panel A in Figure 2.1 illustrates the CPC model of a modern industrial
product with four growth curves representing imports (M), domestic produc­
tion (S), domestic demand (D), and exports (X). Theoretically, the growth
of domestic demand leads CPC development. Statistically, domestic demand
Figure 2.1
Schematic Diagram of the Catching-Up Product Cycle Development of an
Industry
D = S + M – X Domestic
k production (S)

I II III IV V
Development stages
Five development stages M/D = Import/demand ratio.
I introductory X/S = Export/production ratio.
II import substitution S/D = Production/demand ratio.
III export
IV mature
V reverse import

30 Economic Development and International Trade
is defined to be production plus imports minus exports (D = S + M – X),
but its growth curve determines the basic pattern of development for a par­
ticular industry. Panel B illustrates the change for the two key ratios, the im­
port/demand ratio (M/D) and the export/production ratio (X/S), in the
process of CPC development. The two ratios are often used to measure the
progress of import substitution and export expansion, respectively. In this
schematic diagram, exports begin only after imports cease. In reality, however,
products are differentiated within the same category, and the export of lower-
quality products can occur at the same time that import of sophisticated or
higher-quality ones continues.
Panel C in Figure 2.1 illustrates the CPC with a single growth curve that
represents the production/demand ratio (S/D). Five development stages—
introductory, import substitution, export, mature, and reverse import—are
distinguished by the specific values of S/D, namely, 0.5, 1.0, and, again, the
maximum, 1.0. Although the values are arbitrary, distinguishing the stages
remains useful in describing the life cycle of an industry. The main charac­
teristics of these five stages are described below.
At the introductory stage, a new product is introduced via imports from
advanced countries, and domestic consumption of the product increases
gradually. Domestic production begins through imitation or borrowed tech­
nology, but the domestic product cannot compete with the imported product
because of the inferior quality and high production costs of the domestically
produced good.
In the import substitution stage, domestic consumption increases rapidly,
which encourages production to expand at a faster rate than demand, there­
by decreasing the share of imports in the domestic market. Production tech­
nology is standardized, and large-scale production becomes possible, with the
domestic product gradually replacing the imported one. Product quality im­
proves and the price falls below the price of the imported product.
In the export stage, the domestically produced good begins to be exported.
The growth of domestic demand slows down, but with the increase in export
growth, the increases in production can be maintained.
In the mature stage, both domestic demand and exports slowly decrease,
preventing further expansion of production. Exports begin to decrease when
the domestic product fails to compete with similar products from late-starting
countries.
Finally, in the reverse import stage, products of late-starting countries, which
are cheaper and of no less inferior quality, begin to be imported and gradu­
ally replace domestic products in the domestic market, which contributes to
the accelerating decline of domestic production.
The five stages describe the life cycle of an industry's development, but
what mechanism underlies the shift from one stage to the next? At the in-

Interaction Between Trade and Development 31
troductory stage, the learning-by-doing effect in both consumption and
production plays an important role. As consumption of the new product grows,
market conditions are right for domestic entrepreneurs to begin production.
As production experience accumulates, domestic producers gradually improve
quality and reduce costs. Factors common to both the import substitution
and export expansion stages help to make the shift from one to the other
a continuous process. In particular, the growth of domestic and foreign de­
mand for the good enables substitution of the domestic product for foreign
products in the domestic market (import substitution) initially, and then in
foreign markets (export expansion). Decreasing unit costs are later realized
through operating on a larger scale, adoption of better technology, and ac­
cumulated experience in both labor and management, all of which are made
possible through the expansion of production and increasing capacity in­
vestment.
Entrepreneurship and the accumulation of managerial resources specific
to an industry underlie capacity investment and decreasing unit costs in the
process of CPC development. Managerial resources include not only capital
for capacity investment, but also production and marketing technologies
specific to the industry, as well as managers, engineers, foremen, and other
skilled workers embodying industry-specific technology. Domestic produc­
tion is initiated by an entrepreneur who, with foresight, observes the increase
in demand for the new product in the domestic market and realizes the profita­
bility of import substitution. He may be handicapped by the inadequate avail­
ability of managerial resources but will gradually accumulate them as he
produces and sells his product. The learning-by-doing effect in production
and the increased competitiveness of domestic products both reflect the
achievement of accumulated managerial resources.
The shift from the export to the mature stage is caused by stagnant de­
mand growth, which discourages capacity investment and further cost reduc­
tions. As long as export growth steadily increases, it will offset stagnant
domestic demand and lead to further expansion of domestic production, thus
postponing the mature stage. However, producers often lack knowledge of
market conditions abroad, and their decisions to invest in capacity expan­
sion tends to be influenced by domestic demand. It is akin to the emphasis
Linder (1961) places on the domestic market. On the other hand, as import
substitution occurs in late-starting countries, the domestic demand of these
late-starting countries begins to be met by their own production and the ex­
porting country will have to decrease its production of the good. The process
of decreasing exports will continue to the point when exports cease altogether.
Imports will begin again, initiating the reverse import stage. These last two
stages are continuous and are moved by a common factor, which is a decrease
in competitiveness resulting from the entry of products from late-starting coun­
tries into the domestic markets of exporting countries.

32 Economic Development and International Trade
How do government policies affect CPC development? They can either ac­
celerate or decelerate CPC at each stage of its development. During the in­
troductory stage, governments can encourage local or foreign firms to initiate
new industries by means of subsidies or tax exemptions. A government may
even establish a state enterprise to begin a new industry and finance losses
for a fixed period of time, as is often the case in strategic industries such as
iron and steel. During the two import stages, import substitution and reverse
import, a government can support domestic production by affecting the com­
petition between the domestic product and the imported one. The aims of
government policies in the two different import stages, however, are not the
same. In the import substitution stage, import substitution is promoted
through government policies that restrict imports or subsidize domestic
producers. This is typically done to protect infant industries. On the other
hand, during the reverse import stage, a government assists domestic produc­
tion by slowing the inflow of or restricting imports or by subsidizing produc­
tion. In this case, protection is usually implemented to avoid the high social
and economic costs of adjustment in declining industries. However, the in­
crease in imports should not be restricted too long.
During both the export and mature stages, a government's ability to affect
the CPC is reduced. Some governments promote the initiation of new product
exports with subsidies, but continuing export subsidies indefinitely is costly.
Thus, the main mechanisms for these stages in CPC development are market
forces such as growth in demand and decreasing costs.
It has already been confirmed that the major modern industries of Japan
followed CPC development. It is never easy to match production and trade
data for well-defined homogeneous products. For broadly defined products,
the difference in product mixes among production, imports, and exports
should be considered and price indexing should be employed to obtain series
at constant prices. 2
It is not sufficient to construct CPC curves for individual industries; anal­
ysis of the development of selected industries in the chapters that follow also
considers the other relevant variables. Chapters 4 and 5 analyze the CPC of
two major industries, textiles and steel. Chapter 5 presents a simple econo­
metric model that incorporates such variables as domestic and foreign in­
come and relative domestic/foreign prices. The model is extended to simulate
the alternating performance of import substitution and export expansion un­
der some counter-factual assumptions. The two industries illustrate success­
ful CPCs with and without government protection. Chapters 7 to 9 analyze
policy aspects of CPC development, and Chapter 10 looks at the relevance
of the CPC model to contemporary developing countries. 1

Interaction Between Trade and Development 33
Variation 1: Linkage to Other Industries
Akamatsu (1943) observes CPC development in a succession of industries.
Diagrammatically, the CPC development of other industries can be illustrated
by superimposing import, production, and export curves (as in Panel A of
Figure 2.1) as well as other industry life-cycle curves (S/D curves [as in Panel
C of Figure 2.1]), over the observed set of curves of the First industry (with
some time lags) along the same time scale (overlap is illustrated in Figures
5.3 and 10.1). Akamatsu describes the overlap of curves as a variation of the
CPC model. The resulting diagram bears a resemblance to a flock formation
of flying wild geese, but the relationship between the first CPC and succeed­
ing CPCs has deeper economic meaning than its mere resemblance in shape.
The linkage between the industry curves reflects the mechanism underlying
the diversification of the production structure in the process of industriali­
zation.
Four identifiable forces affect the diversification of industrial structure.
The first force is the increased availability of capital. The factor proportion
theory (or Heckscher-Ohlin theory) of international economics is applied to
the process of structural change in production and trade as industrialization
proceeds. Industrial production is mobile and can be located in any country
by importing material inputs unavailable domestically. The final output is
produced by combining labor and capital with materials. Individual indus­
tries differ in their capital and labor composition; light manufacturing in­
dustries require relatively more labor, whereas heavy manufacturing industries
require relatively more capital. The two-commodity, two-factor model of the
factor proportion theory determines the output composition of two indus­
tries, so that all available capital and labor are fully employed. In Japan, where
labor was abundant at the early stage of industrialization, light industry was
relatively larger than heavy industry, but as Japan accumulated capital the
share of heavy industry increased.
The statistical classification of light and heavy industry does not coincide
exactly with the distinction between labor- and capital-intensity in the factor
proportion theory. Although metal refining and petrochemicals production,
which are classified as heavy industries, require relatively few laborers to main­
tain and operate capital equipment, the assembly of machinery, which is also
considered a heavy industry, requires many, especially skilled laborers. On
the other hand, light industry requires relatively more unskilled laborers and
is characterized by small-scale operations and technology that is easily ab­
sorbed even in modern factories. H. G. Johnson's (1968) capital-theoretical
approach resolved this discrepancy between theory and statistics. In addition
to physical capital such as machinery and equipment, both the skill and tech-

34 Economic Development and International Trade
nological know-how embodied in laborers and engineers are considered to
be forms of capital. If only unskilled labor is defined as labor and all forms
of capital are broadly defined as capital (including managerial resources),
then the factor proportion theory serves well to explain heavy industrializa­
tion in the process of capital accumulation.
Thus, the two-factor, two-commodity model explains how capital-intensive
production expands—that is, how the industrial structure is diversified—so
that accumulated capital is fully employed. If the industries are disaggregated,
the different forms of capital and technology specific to individual indus­
tries have to be distinguished in order to explain the actual process of indus­
trialization; this forms the basis of the multi-factor, multi-commodity model.
It is well known that there is no unique allocation of the many factors among
the many industries in such a model (Caves and Jones 1973: chaps. 7 and
8). That is, changes in production and the trade structure cannot be deter­
mined by full employment alone. Additional factors have been introduced
in order to determine the diversification of the industrial structure (see the
sectoral models of Kojima [1958: chap. 7], Inada et al. [1972], and the nuclear
sector model of Bensusan-Butt [1954]). The accumulation of capital provides,
then, the first explanation for the broad diversification of the industrial struc­
ture, but it must be supplemented by additional factors affecting actual diver­
sification at a more disaggregated level. The CPC model meets this need.
Entrepreneurship is viewed as the force behind the decision to begin a new
industry. The basic CPC model reveals a few important determinants for the
expansion of production in a newly introduced industry: growth of domestic
demand, realization of decreasing unit costs as production expands, and pro­
tection provided by government. The entrepreneur's decision to begin a new
industry is affected by the expected value of these variables but these varia­
bles do not explain the situation fully. Educational background and business
acumen, for example, help shape decisions (Yamazawa and Tambunlertchai
1985). Chapter 6 analyzes the CPC model for trading companies and a ser­
vice industry handling foreign transactions; Japanese trading companies that
have helped manufacturers determine advantageous shifts in comparative ad­
vantage are also discussed.
Forward and backward linkage effects assist the transfer of CPC from the
first industry to the others. In forward linkages, industries relying on the in­
put of the first industry's output are assisted by following the CPC model,
and in backward linkages, industries supplying inputs to the first industry
benefit from increases in demand. Chapter 4 provides two econometric ex­
amples of backward linkages in the cotton textile industry, and Figure 5.5
shows an overlap of CPC curves, which reflects industrial diversification.
Chapter 3 analyzes the export growth of the silk industry, which, although
it was the first modern manufacturing industry established and the biggest

Interaction Between Trade and Development 35
earner of foreign exchange in its first few decades, neither followed CPC de­
velopment nor benefited from linkage effects.
The learning-by-doing effect helped the transfer of one industry's CPC to
the other. CPC development is generally achieved first in goods for which
production is standardized and then in goods requiring more sophisticated
production processes. This process has been observed most clearly in the
production of low-count and high-count cotton yarns, and black-and-white
and color televisions. However, owing to elements common to both produc­
tion and marketing technology for each pair, the CPC of sophisticated
products may be cut short because the know-how gained in the production
of less sophisticated products can reduce the need for the trial and error
method in both the introductory and import substitution stages. Production
can then be geared to the export stage from the beginning.
Variation 2: International Transfer of CPC,
Direct Foreign Investment, and Technology Transfer
The transfer of CPC from one country to another can result in variations
of the CPC model; the overlap of an industry's life-cycle curves among coun­
tries at different development stages illustrates variations of flock formations
of flying wild geese (see Figure 10.1). This can be explained by the basic type
of CPC in a late-starting country, which is also affected significantly by the
transfer of managerial resources from an early- to a late-starting country.
Managerial resources can be imported from abroad rather than generated and
accumulated locally to hasten CPC development.
In many industries in Japan, indigenous firms accumulated their own
managerial resources through trial and error, using domestic capital and relying
on imported machines, equipment, and materials. In a few industries, however,
CPC was hastened by introducing managerial resources from abroad. New
technology was introduced under licensing agreements (technology transfer)
or, if foreign suppliers of technology insisted, through investment in kind under
joint-venture agreements with the foreign firms (direct foreign investment—
DFI).
Inward DFI and technology transfer have been observed in the introduc­
tory and import substitution stages, but in the export and mature stages out­
ward DFI and technology transfer can occur and tend to accelerate CPC
development in the investing country. Outward DFI and technology transfer
occur only after enough managerial expertise has been acquired. Outward
DFI is further distinguished between the export and mature stages. Export
stage DFI primarily supports export marketing and aims to secure a share
of export markets. Thus exports and DFI increase in a parallel manner. On
the other hand, during the mature stage, DFI can be used to relocate domes-

36 Economic Development and International TYade
tic production abroad to take advantage of cheaper input prices or further
market expansion: in this case, DFI is a substitute for domestic exports.
DFI and technology transfer accelerate import substitution in host, late-
starting countries, as has been frequently observed in East and Southeast Asian
countries for the past two decades. Both DFI and technology transfer have
enabled rapid catching-up industrialization in these countries. Host coun­
tries and domestic policies on DFI and technology transfer may encourage
or restrict them. However, such policies are often influenced by policy objec­
tives other than trying to deliberately control CPC, including correcting the
balance of payments or providing employment.
Domestic production sometimes begins entirely with foreign entrepreneurs
who control 100 percent of the equity shares and with imported managerial
resources, which is consistent with the product cycle model proposed by Ver­
non (1964). In this situation, the trial and error method can be avoided, much
of the learning-by-doing process in the product cycle model can be shortened,
and the foreign firm can begin to export immediately, thus cutting short the
CPC. However, only a minimum of managerial resources is usually trans­
ferred to the host country, and the industry is never able to produce new
products or employ new processes on its own. Every new product or process
has to be introduced from abroad—a situation that does not fit the defini­
tion of a transfer of new industry.
The transfer of new products and new technology from advanced coun­
tries in one form or another is indispensable for the start of CPC in a late-
starting country, but the manner in which this transfer occurs significantly
affects the later performance of CPC. The manner of technology transfer
is determined by the entrepreneurial skills of local firms and the negotiations
between local and foreign firms. The last section of Chapter 7 explains how
Japanese electrical appliance firms began as joint ventures with American
and European firms but quickly evolved into independent enterprises in the
1920s. Tran (1988) analyzes the similar experience of Korean synthetic fiber
producers in the 1970s. Chen (1989) extends the CPC model to promote the
transfer of the electronics industry among the Asian NIEs and ASEAN mem­
bers in such a way that industrial adjustment is coordinated in the region.
EFFECT OF TERMS OF TRADE
ON CPC DEVELOPMENT
A necessary condition for successful CPC is the realization of decreasing
costs in the expansion of production, which makes domestic products cheaper
relative to foreign products both in the domestic market (through import sub­
stitution) and in foreign markets (through export expansion). In order to ex­
amine this mechanism, it is necessary to investigate changes in prices. In Japan,

Interaction Between Trade and Development 37
it was, of course, the declining relative prices of domestic to foreign products
that promoted substitution, while all absolute prices, both domestically and
in foreign markets, moved parallel with one another and showed a common
trend, rising slowly in the period before the First World War, rising steeply
during the war, and steadily declining throughout the 1920s (Yamazawa and
Yamamoto 1979a: chap. 1).
Figure 2.2 shows four types of relative price changes. The commodity terms
of trade (TT in Figure 2.2) is defined as the total export price index divided
by the total import price. Since this ratio gives the amount of imports that
could be exchanged per unit of exports, it is often used as an indicator of
changes in the gains from trade. The effect of the terms of trade on Japan's
export expansion in the pre-Second World War period has been widely de­
bated. Over the entire prewar period, Japan's terms of trade decreased. Ac­
cording to Shinohara (1962: chap. 10), this decline enabled the expansion of
Japan's exports, but Kojima (1958: chap. 3) claims that the decline in the terms
of trade contributed to export growth mainly during the interwar period. A
new estimate by Yamazawa and Yamamoto (1979a), shown as TT in Figure
2.2, seems to support Kojima. The declining trend in the terms of trade is
clear and is, without doubt, related to the export growth during the periods
1908-18 and 1924-36, but not during the period of rapid export expansion
before 1900. On the other hand, during the high-growth period after 1945,
the terms of trade were stable, declining steeply only during the period after
1973 when the price of oil and other raw materials increased dramatically.
The factoral terms of trade (FTT) are derived by introducing labor produc­
tivity in the export industries (e.g., manufacturing) into the terms of trade
calculation. FTT measures the volume of imports that could be exchanged
per man-hour of labor input in the export industry and complements the com­
modity terms of trade figure. FTT increased even before 1908 and rose steeply
in the 1920s and after the Second World War. The sharp increase in the last
two periods was the result of a remarkable increase in labor productivity.
The other two price indices in Figure 2.2 offer a better means to measure
real price effects because they compare prices of Japan's exports and domes­
tic prices with comparable world prices. The relative export price (REP) com­
pares manufactured export prices in Japan (P x in yen) with the corresponding
world market prices (P w in US. dollars), adjusted by the exchange rate (R
= yen/US$). Represented as [P x R)/P Wt this equation measures the com­
petitiveness of Japan's exports abroad. The other index, the relative import
price (RIP), compares domestic and import prices (P and P m in yen) of
manufactures in the domestic market {P m/P). Extending back to 1898 and
including the whole post-Second World War period, the REP shows a longer
declining trend than do the terms of trade. The RIP, on the other hand, ex­
hibits more variability with an overall declining trend that is offset by increases
during the periods 1906-17 and 1922-36.

Figure 2.2
Terms of Trade, and Relative Export and Import Prices: 1885-1980

1890 1900 1910 i • i
1920 1930
Year 1951 1960 1970 1980
Notes:
TT: Commodity terms of trade (total export price divided by total import price).
FTT Factoral terms of trade (terms of trade multiplied by productivity index in the export sector).
REP: Relative export price ((Py/?)/P w: Japan's export price of manufactures, P x in yen, divided
by corresponding world market price, P w in U.S. dollars, adjusted by the exchange rate,
R. yen/US$).
RIP: Relative import price (P m/P: Japan's domestic price of manufactures in yen divided by
the import price of manufactures in yen).
The base period is 1934-36 for all four series before the Second World War, but 1965 for the
REP and RIP series and 1975 for the TT and FTT series after the Second World War.
Source: Yamazawa and Yamamoto (1979a: chap. 2).

Interaction Between Trade and Development 39
The lack of an obvious trend of the RIP partly results from the combined
effects of diverse price movements in different manufacturing industries.
Against the overall trend of rising prices, there emerges a clear pattern that
shows modern industries had only small price increases, whereas in traditional
industries such as food processing and wood products prices increased stead­
ily. Because traditional industries generally did not have declining unit costs
as did import substituting industries, and because traditional industries oc­
cupied a greater share of manufacturing, the overall trend of the RIP is ob­
scured. On the other Hand, within modern industries such as textiles, metals
and chemicals, which chose import substitution and export expansion as the
means to growth, domestic prices declined relative to import prices, thereby
increasing the RIP. 4
Despite the ambiguities of the movement of aggregate relative prices, it is
worthwhile to investigate their effects on the CPC development of the Japanese
manufacturing industry as a whole. Table 2.1 shows the estimation of export
and import equations for manufactures for three separate periods. Both equa­
tions follow the usual specification and include income and price variables.
The RIP is multiplied by the sum (1 + the average tariff rate) in order to
include the effects of tariff protection. The First World War years are excluded
from the estimation of exports because of the unavailability of statistics for
world trade volume and prices. These years are, however, included in the esti­
mation of imports. A First World War dummy variable has been included
so that the limited availability of manufactured imports during the war can
be taken into account. AH coefficients but one (the price coefficient in the
import equation after the Second World War) are significant and consistent
with theoretical expectations. These findings support the theory that chang­
ing relative prices associated with the decreasing costs from import substitu­
tion and export expansion have a significant impact on industrial development.
It should be noted that the price effect prevails over the income effect in the
export equation in the second and third periods. The actual price effect, which
is measured by multiplying the coefficient by the relative price change, was
more significant in export expansion during the interwar period when world
trade stagnated.
On the other hand, raw material imports (equation 3) are related to
manufacturing production, although with decreasing impact in later years,
as is evidenced by a coefficient less than unity. Any increase in manufactur­
ing production would therefore lead to a less-than-proportionate increase in
raw materials imports. This tendency partly contributed to the balance-of-
trade surplus at later stages of industrial development.
Equations 4 and 5 support the preceding discussion of the effects of domes­
tic and export prices of manufactures. Domestic prices were affected posi­
tively by the wage rate and raw materials prices, but were affected negatively
by productivity increases. Export prices increased less than domestic prices

40 Economic Development and International Trade
in the period of overall rising prices (with elasticity less than unity before
1913 and after the Second World War). Furthermore, the export prices tended
to lag behind domestic prices, as is evidenced by the negative time-trend coeffi­
cient, which implies higher productivity increases and decreasing costs in ex­
port production than in the rest of manufacturing production.
LONG SWINGS AND STRUCTURAL CHANGES
The export and import equations for total manufacturing only provide em­
pirical verification of the importance of changing relative prices at best, be­
cause they represent the aggregate effects of CPC development of individual
manufacturing industries. The composition of industries within total manufac­
turing changed as individual industry CPCs proceeded at different times. The
changes were not smooth but were characterized by long swings associated
Table 2.1
Estimation of Equations for Export, Import, and Related Variables
1. Export of Manufactures
World Relative
Period Constant trade volume export price R 1 (DW)
1885-1913 7.07 2.73 (20.4)
1921-38 7.73 1.18 (3.4)
1953-70 7.92 1.50 (15.3) 0.98 (3.0)
1.79 (10.1)
2.13 (5.4) 0.971 (1.59)
0.925 (0.69)
0.997 (1.06)
2. Import of Manufactures
Period Constant Japan's GDP Relative
import price
including
tariff First World
War dummy R 1 (DW)
1885-1913 – 12.52 2.08 (9.1)
1914-38 -4.04 1.15 (8.7)
1953-70 – 10.58 1.70 (14.5) -1.83 (3.6)
-1.43 (7.3)
-0.48 (0.5) -0.17 (2.1) 0.958 (1.49)
0.968 (1.94)
0.979 (0.43)
3. Import of Raw Material
Manu­
facturing
Period Constant output Relative
import price
of raw
material First World
War dummy R 1 (DW)
1885-1913 -15.03 2.57 (15.4)
1914-38 0.92 0.66 (10.1)
1953-70 -1.87 0.91 (44.9) -1.96 (4.3)
-0.19 (2.9) 0.979 (0.84)
0.961 (1.51)
0.996 (1.84)

Interaction Between Trade and Development 41
with fluctuations in industrial activity. Figure 2.3 shows the growth cycle of
Japan's manufacturing output (A) and the corresponding moving average of
the coefficients of structural change in manufacturing output (B). The out­
put growth cycle shows a long swing of around 20 years' duration and a clear
upward trend through swings (connecting one peak with the next, and one
trough with the next) until the early 1970s when it declined abruptly (Oh-
kawa and Rosovsky 1973).
A relationship between curves A and B can be observed by comparing the
peaks in curve A with the peaks in curve B at least until 1970. If an indus­
try's share increases steadily by say 2 percent every year, curve B will be flat.
But if the industry's share increases rapidly for several years and then ceases
for another several years, curve B will peak, which indicates a boom in in­
dustrial activity. During industrial booms (peaks of the growth cycle), all in­
dustries do not expand simultaneously. Some industries expand more than
Table 2.1 Continued.
4. Domestic Price of Manufactures (P)
Labor Material
Period Constant Wage productivity price R 1 (DW)
1885-1913 0.75 0.77 (10.2) -0.16 (1.2) 0.985 (0.83)
1914-38 0.33 0.16 (2.6) -0.24 (3.0) 0.63 (7.6) 0.924 (1.29)
1953-70 0.08 0.29 (5.2) -0.09 (1.8) 0.35 (5.5) 0.986 (1.84)
5. Export Price of Manufactures
Domestic price
Period Constant manufactures Time-trend R 1 (DW)
1885-1913 0.56 0.90 (3-5) 0.011 (1.1) 0.902 (0.63)
1914-38 1.31 1.06 (11.3) 0.026 (7.9) 0.982 (0.90)
1953-70 1.03 0.71 (4.9) 0.015 (7.2) 0.873 (1.52)
Notes: The relative export price is Japan's price of manufactured exports divided by the world
price of manufactures and adjusted by exchange rate changes. The relative import price of
manufactures and that of raw materials are Japan's import price of manufactures and raw
materials, respectively, divided by the domestic price of manufactures.
All equations are estimated using the ordinary least squares method in natural log-linear
form. Thus the estimates give the contribution of the rate of changes in individual explana­
tory variables on the right-hand side to the rate of changes in explained variables on the
left-hand side despite different units of measurement of individual variables. Figures in paren­
theses after coefficients are /-statistics. R 2 is the coefficient of determination adjusted for
degrees of freedom, and DW is the Durbin-Watson statistic.
Sources: Ohkawa, Takamatsu, and Yamamoto (1974); Shinohara (1972); and Yamazawa and
Yamamoto (1979a). See Shionoya and Yamazawa (1973) for further technical details.

42 Economic Development and International Trade
Figure 2.3
Long Swings in Manufacturing Output: Growth and Structural Change
20
15
10
% T 1—r
A T 1 1 1 1 1 1 r-Nn 1 1 1 1 p
5 I I i ' ' ' i 1 i I l I uJs,J i I i I u
1880 1890 1900 1910 1920 1930 1950 1960 1970 1980
1890 1900 1910 1920 1930 1955 1960
Year 1970 1980
(A) Growth cycle of manufacturing output: seven-year moving average of the growth rate of
manufacturing production to previous years calculated from 1934-36 constant price series.
Five-year moving average after the Second World War.
(B) Seven-year moving average of the coefficient of structural change as defined below. Five-
year moving average after the Second World War.
v(t) = Z\s(t,j) – s(t-lj)\ t
j
s = the percentage share of industry j in total manufacturing.
Sources: Yamazawa and Yamamoto (1979a: chaps. 1 and 4).
others, thereby enlarging their shares, and this causes curve B to peak.
Figure 2.3 shows the long swings in manufacturing output, which can be
related to the structural changes of industrial composition. The 1889 peak
corresponds to the expansion of textile production while food processing and
other sectors decreased; the 1915 peak corresponds to the expansion of metal,

Interaction Between Trade and Development 43
chemical and machinery manufacturing, as well as textiles, while food process­
ing and other sectors decreased further; and peaks in 1936 and around 1960
correspond to the expansion of metal, chemical and machinery manufactur­
ing at the expense of textiles and food processing production.
The correspondence of the growth cycles of trade to the output growth cy­
cle reveals the interaction between trade and production. A clear relationship
between changes in the production and imports of both raw materials and
manufactures has been observed, which is consistent with the estimation of
significant income effects in import equations 2 and 3 of Table 2.1. That is,
the manufacture of industrial products and the growth of GDP strongly in­
duced imports of raw materials, machinery, and intermediate goods.
On the other hand, the cause and effect relationship between production
and exports is reversed in theory, which argues that exports, an important
element of aggregate demand, induce output growth. The data generally sup­
port the theory, showing, for the most part, steady export growth relative to
the output growth cycle, although the relationship was obscured during the
First World War and in the early 1930s.
The correspondence between the growth cycle and structural change coef­
ficients has been observed in manufactured exports and imports as well
(Yamazawa and Yamamoto 1979b). Moreover, the correspondence between
structural change coefficients of both exports and imports and output cycles
provides further evidence of a peculiar characteristic of structural changes
in Japan's production and trade; that is, the same change, beginning with
light industrialization and moving to heavy industrialization, first occurred
with imports, then with production, and ended with exports. These all show
that the CPC development of textiles, metals, chemicals, and machinery
production were the underlying mechanisms behind the expansion of and
interaction between production and trade in Japan.
BALANCE-OF-PAYMENTS CONSTRAINT
ON INDUSTRIAL GROWTH
Why did the process of light and heavy industrialization not continue its
steady growth? Why was this process interrupted by more than one indus­
trial boom? The most important constraint to steady growth was a deteriora­
tion of the balance of payments. The relationships between output cycles and
trade described above affect the balance-of-payments situation, which, in turn,
either constrains or encourages output growth.
The effect of an industrial boom depends on whether it is led by growth
of domestic demand or by growth of exports. In the rising phase of long swings
that are initiated by some domestic incentive to increase output, import growth
tends to exceed export growth, thereby decreasing the export/import ratio.
The resulting balance-of-payments deficit triggers monetary contraction,

44 Economic Development and International Trade
which restrains investment and other economic activities, and thus slows out­
put growth. In the declining phase of long swings, import growth falls short
of export growth, thereby increasing the export/import ratio. The improved
balance of payments helps ease credit, and investment resumes to increase
output, initiating a new rising phase in the long swing. Figure 2.4 (A) shows
the interaction of production, trade, and the trade balance over the rising and
declining phases of a long swing. A solid line added to the export/import
ratio represents the critical trade balance ratio. The long swings of the ex­
port/import ratios lead those of manufacturing production by about a quarter
cycle, as shown in (A). This is essentially what happened in the pre-First World
War and post-Second World War long swings.
During the export-led booms of the First World War and the early 1930s,
export growth induced output expansion directly. Since export growth exceeded
import growth, the export/import ratio increased and the balance of pay­
ments ran a surplus. Output expansion stopped only after the export booms
were ended by exogenous factors, including the end of the First World War
and the worldwide prevalence of bloc economies in the 1930s. As a result,
import growth then exceeded export growth and caused a deterioration in the
balance of payments. In this case, long swings in the export/import ratio
Figure 2.4
Schematic Diagram of Alternative Correspondence among
Production (S), Trade (X and M), and the Trade Balance (X/M)
A B
Production
growth < ^T'ow ^^^-^-s^ Production
j f | growth
Export [
growth *f ! I i ' ' 1 1
""r*\ 1 ! • ! ' Import
! X 1 ! \s 1 V' ! 9 rowth
Import f\
growth 1 i 'X^ j _s * 1 X i Export
J ! ' i i ! growth
Critical trade ^'^**\,
balance ratio / \ i 1 i ' I ' i
1 [ i ] J —-f-^w 1 Critical trade
^! i [y , l^i^^'Xl balance ratio
Export/ j
import ratio i J ! ! ! ! ! Export/
iii i i i i import ratio
iii i i i i
(A) Domestic demand-led growth.
(B) Export-led growth.
Source: Yamazawa and Yamamoto (1979b: figs. 7.3, 7.4).

Interaction Between Trade and Development 45
tended to lag behind exports and manufacturing production by about a quarter
cycle, as shown in Figure 2.4 (B).
Figure 2.4 is an oversimplification of reality, and the correspondence be­
tween the growth cycle and balance of payments (export/import ratio) is ac­
tually more complicated. In long swings, the actual, observed rising and falling
phases are not of the same length, and the degree of monetary contraction
brought about by the decreasing export/import ratio will vary depending on
the prospects for capital inflows and with different exchange rate systems.
Comparing (A) in Figure 2.3 with Figure 1.4 (keeping in mind the schematic
correspondence of Figure 2.4) is worthwhile. Long swings in the export/import
ratio appear to have led the growth cycle of manufacturing production by
two to three years before the beginning of the First World War and after the
Second World War, while it was synchronous or tended to lag behind during
the interwar period. Decelerating trends of imports of materials and manufac­
tures relative to the high, steady growth of exports produced a strong upward
trend in the export/import ratio, and this resulted in a persistent surplus dur­
ing the 1930s and the late 1960s. Long swings of accelerated growth of
manufacturing production corresponded to strong improvement in the long
swings of the balance of payments (steep upward trend of curve B in Figure
2.4). s
The international monetary system played a key role in determining how
quickly a deteriorating balance of payments would negatively affect indus­
trial growth. Under the gold standard between 1897 and the First World War,
and under the pegged exchange rate system in the 1950s and 1960s, the de­
terioration in the balance of payments triggered monetary contraction, which
restrained investment and other activities, and slowed industrial growth.
However, as long as foreign capital inflows were available, the effect of a deficit
in the balance of payments was delayed and structural change could continue
uninterrupted. But, when the yen depreciated in 1932, a balance-of-payments
deficit resulted; nevertheless, exports expanded, producing an industrial boom
in the early 1930s.
CONTRIBUTION OF FOREIGN CAPITAL
Another characteristic of the Japanese model is that domestic saving
financed a major portion of Japan's domestic investment before the Second
World War. During a few periods, foreign capital made significant contribu­
tions to domestic investment and eased the balance-of-payments constraint
to industrial growth (Yamamoto 1969; Key 1970). Table 2.2, which shows the
balance of payments of Japan, breaks down the inflow of foreign capital by
flow type and provides three alternative measures of the net flow of foreign
capital. The first measure is the most broadly defined and includes transfer

Table 2.2
Japan's Balance of Payments
1886-93 1894-1903 1904-13 1914-19 1920-31 1932-36
1. Commodity trade balance -24.8 -399.1 -552.6 1,333.7 -4,737.3 -818.9
2. Service trade balance 8.5 -14.3 -445.6 1,595.9 2,092.5 -118.6
3. Net transfer payment 17.7 436.4 75.1 150.9 242.9 416.7
4. Net long-term capital balance -6.6 156.4 1,254.5 -1,668.5 – 1,720.2 -2,850.0
4.1. Increase in external debt — 170.0 1,600.0 – 246.0 524.0 —
4.2. Direct investment — — 70-100 — 145.0 —
5. Net short-term capital balance 24.9 – 165.7 -97.2 343.3 2,726.5 1,712.5
6. Specie movement and changes in specie held abroad -19.7 -13.7 -234.2 – 1,775.3 1,395.6 1,658.3
7. Current surplus/GDCF (<7o) a 1.5 12.0 15.2 -29.1 8.0 3.3
8. Current surplus/GDS (°7o) a 1.5 13.4 20.2 -24.0 9.6 4.5
9. Transfer and capital balance/GDCF 3.3 12.4 18.7 -11.6 3.8 -2.6
10. Transfer and long-term capital balance/GDCF 1.0 17.2 20.2 -15.1 -4.5 -8.1
Notes: The sum of lines 1 to 6 is 0.
Line 7 = -{(1) + (2)}/GDCF.
Line 8 = -{(1) + (2)}/GDS.
Line 9 = -j(3) + (4) + (5)}/GDCF.
Line 10 = {(3) + (4)}/GDCF.
a. Reparation payments are not included in the current account because they served as capital inflows during these periods in Japan.
Sources: Based on Yamamoto's estimate in Yamazawa and Yamamoto (1979a); GDCF and GDSare from Ohkawa, Takamatsu, and Yamamoto (1974).

Interaction Between Trade and Development 47
payments of grants from abroad and reparation payments, short- and long-
term capital flows, and exports of specie—the sum of lines 3 to. 6.
The second measure is the sum of lines 3 to 5. It excludes specie movement
and changes in specie held abroad. A substantial share of Japan's foreign
exchange reserve was held in the form of specie deposits at banks in London
during the years 1900-1930. Annual capital flow data (line 4) clearly show
a change from a net inflow over a period of consecutive years to a net out­
flow over the next period, followed by another period of net inflows and then
outflows. These flows correspond to the periodization in the table.
The third measure has the narrowest definition and includes only the flow
of funds of a long-term nature—lines 3 and 4. They are divided by either
gross domestic capital formation (GDCF) or gross domestic saving (GDS)
to indicate their relative contributions.
Significant capital inflows were recorded for two periods, 1894-1903 and
1904-13. In terms of the third measure, it contributed 17 and 20 percent to
the GDCF, respectively. Whereas the capital inflow in the first period was
from the reparations paid by China (1896-98) after the Sino-Japanese War,
in the second period it was from loans floated in London and New York.
Japan adopted the gold standard from 1897 to 1917, during which time per­
sistent trade deficits reduced the specie held abroad, thereby setting the stage
for monetary contraction and depressed industrial activity. On the other hand,
a huge amount of specie was accumulated in London banks during the ex­
port boom of the First World War. After the First World War (1921-31), capi­
tal flows into Japan came from both foreign portfolio and direct investment.
But the inflow of capital was largely offset by Japanese investment in Man­
churia, and for this reason, the third measure shows a net outflow.
Incidentally, after the Second World War only a small percentage of net
inflow was recorded. The equivalent to line 10 in Table 2.2 for 1962-64 was
2.2 percent, showing that the rapid growth was mainly financed by domestic
savings.
The model as discussed thus far shows how major modern manufacturing
industries followed CPC development and how both the industrial and trade
structures changed drastically as CPC development was accomplished. In mac-
roeconomic terms, the accelerated trend of industrial growth was accompa­
nied by both a steady expansion of manufactured exports and a deceleration
of imported raw materials and intermediate manufactured goods, thereby im­
proving the industrial trade balance over time and achieving persistent trade
surpluses in the 1930s and again after the mid-1960s. Japan's industrial growth
was continuously severely constrained by balance-of-payment deficits for the
first 50 to 70 years of industrial development but these were corrected even­
tually.
Japanese industries have outgrown the CPC model as a means for develop-

48 Economic Development and International Trade
ment. The successful CPC development of major industries has resulted in
an industrial complex with a high degree of self-sufficiency except in its needs
for raw materials. This success now generates persistent trade surpluses which
have become a major source of friction with Japan's major trading partners.
Some industries are at the forefront of technology development and are no
longer in a position of catching up; therefore, the development and growth
of these industries is more consistent with the product cycle model. However,
the CPC model still remains useful in explaining the industrialization of East
and Southeast Asian countries. While the CPC model was developed from
Japan's special situation in the late nineteenth and early twentieth centuries,
its extension to late-starting countries under severe resource constraints as
a strategy of industrial development should be carefully considered. This is
discussed further in Chapters 9 and 10.

Part II
Industry Growth
and International Trade

Three
Raw-Silk Exports and
Japanese Economic Development
There are numerous instances of countries in which foreign trade generated
a surge in economic growth. Canada is a well-known example of a country
whose growth experience can be represented by the ' 'staple theory of growth.''
In the case of Japan, it has been argued that raw-silk exports spurred the
country's economic takeoff. However, closer examination shows that Japan's
experience with growth due to raw-silk exports differed from the staple the­
ory in several ways. Moreover, the expansion of export-oriented raw-silk
production exhibits the characteristic features of Japanese industrial develop­
ment by which Japan has become known as the pioneer of the so-called export-
led growth strategy, now the model for small, resource-poor, newly industri­
alizing economies (NIEs). Thus, an analysis of Japan's raw-silk export growth
is relevant to current economic events. However, because of inherent limits
in domestic demand expansion, scale economies, and technological progress,
the silk industry did not become the core of Japan's economic growth. Rather,
sustained industrialization only became possible with the type of CPC growth
that characterized the cotton industry.
STAPLE GOODS EXPORTS AND
THE SURGE IN ECONOMIC DEVELOPMENT
In the nineteenth century, the United States and Canada expanded their
exports to Europe with such goods as lumber and wheat, which led to the
inflow of additional capital and labor to the United States and Canada. As
demand for export production and consumption goods expanded, and as the
population increased, production of intermediate and capital goods was also
stimulated. Thus, according to the staple theory of growth, foreign demand

52 Industry Growth and International Trade
stimulated expansion in the production of staple goods and ultimately brought
about economic growth in the sparsely populated, newly opened countries
of North America.
However, production of staple exports in response to foreign demand has
not always led to economic growth. For example, during the same period,
the Southeast Asian countries' exports of rice, rubber, and tin did not result
in a similar development process. The abundance of natural resources in
Southeast Asia that were suitable for production of staple goods were used
merely to meet the small domestic demand, and many resources were left un-
exploited. With the opening of foreign trade, a demand for exports was gener­
ated and capital and labor were channeled to export production. However,
this growth in the production of staples merely led to the utilization of hereto­
fore idle resources, but did not bring about sustained economic growth. In
this sense, exports became merely a surplus outlet (vent for surplus theory).
The Southeast Asian experience points out that the resource endowments of
a region and the characteristics of the goods themselves determine whether
staple goods exports will lead to sustained economic growth. 1
In Japan, too, export-oriented domestic production began with the export
of staples such as raw silk, coal, and copper and grew with the opening of
foreign trade in 1858. Raw silk was considered an important commercial item,
both in terms of its export value as well as its prospects for long-term growth.
For this reason, Baba and Tatemoto (1968) and Henmi (1969) view raw-silk
exports as Japan's version of a staple export that induced the country's sus­
tained economic growth.
Indeed, this view is supported by the following considerations:
1. Raw silk was one of the specialty goods that was indigenous to resource-
poor Japan.
2. Because of the limited domestic market for silk, which is a luxury good,
most of Japan's production of raw silk was destined for export markets.
At the time, export expansion was enhanced by favorable conditions such
as the development of the U.S. weaving industry, which increased demand
for raw silk, and the silkworm disease epidemic in Europe, which reduced
the worldwide supply of raw silk.
3. Exports of raw silk became an important source of foreign exchange for
Japan and made possible the Financing of essential imports required in
the initial stages of economic growth. During the first three decades fol­
lowing 1858, raw-silk exports made up a third of Japan's total exports.
However, Japan's silk exports did not conform precisely to the schema of
the vent for surplus theory. In response to the growth of the silk-reeling in­
dustry, sericulture also grew but did not absorb all of the idle land and labor
resources. Wheat production competed for the land needed for mulberry trees
and for the labor required in silkworm breeding.

Raw-Silk Exports and Economic Development 53
In addition, closer examination of the three points raised above shows that
raw-silk exports did not perfectly conform to the staple theory of growth:
1. Export growth did not come about merely as a result of the growth in for­
eign demand for Japanese silk. Even within the growing U.S. market,
Japanese raw-silk exports were able to take away an increasing share of
the market from Italian and Chinese exporters because of lower prices.
2. At the time there was little left in the way of unutilized resources in dense­
ly populated Japan, and the sericulture industry had to vie with the more
developed areas of wheat agriculture for resources. This resource limita­
tion was overcome by technological progress and increased labor produc­
tivity in the sericulture and silk-reeling industries, leading to further export
expansion.
3. Though the effect on foreign exchange earnings was undoubtedly great,
because silk was a luxury consumer good, its scale of production remained
small. Furthermore, since production of silk relied in part on traditional
technology, its ability to induce the development of other industries or to
generate a surge of modern economic growth was limited.
Because of the expansion of demand for silk resulting from the export trade
and the subsequent growth in production, technological progress and produc­
tivity improvements were realized. This demonstrates the dynamic effect of
trade, which is strongly evident in the case of Japan's raw-silk export growth.
Such effects were subsequently realized in the development process of the silk-
weaving and textile industries. Moreover, if one considers the considerable
effect on foreign exchange earnings as well as the peculiarities of the indus­
try in terms of productivity growth and technological progress under condi­
tions of resource constraints, then the industry may be considered the
forerunner of the export-oriented industrialization strategy.
Although Japanese silk exports continued to grow—even to the extent of
gaining a monopoly position in the U.S. market in the 1920s—their role in
Japan's economic growth began to decline after 1910. The analysis presented
here therefore considers silk exports only until 1910.
THE PROCESS OF RAW-SILK EXPORT GROWTH
Japan began exporting raw silk in response to strong demand from the Eu­
ropean silk-weaving industries that were plagued by an insufficient supply
of raw silk due to the silkworm disease that had ravaged the sericulture in­
dustry in Europe. The composition of Japanese silk exports changed con­
siderably over time. In the second half of the 1860s, 40 percent of silk exports
were in the form of silkworm eggs and cocoons, but the share of these ex­
ports declined thereafter to a minimal share by 1885. Excluding this initial
period, silk exports consisted mainly of raw silk, and between 1870 and 1910

54 Industry Growth and International Trade
raw-silk exports continued to grow at an annual rate of 8.7 percent (as mea­
sured in yen). During this period, the export price of raw silk rose at an an­
nual rate of 2.2 percent while export volumes expanded at about 6.5 percent.
In the second half of the 1870s, the United States became a major importer
of raw silk due to the rapid development of its domestic silk-weaving indus­
try. At the time, only a small share of Japan's raw-silk exports was destined
for the U.S. market. However, the U.S. share of raw-silk exports from Japan
rose to 42 percent during the first half of the 1880s, to 57 percent in the latter
half of the 1880s, and eventually reached 71 percent in 1901 (see Table 3.1).
In contrast, exports of silk fabrics to the United States, which expanded rapidly
in the second half of the 1880s, at best represented only a third of the total
value of Japan's raw-silk exports. (As silk yarn export volume was insignifi­
cant, it is not discussed here.) This was partly due to the US. tariff structure,
which caused the volume of silk yarn and silk fabric exports to remain small
relative to that of raw-silk exports to the United States. Under the Tariff Law
of 1883, raw-silk imports could enter duty-free, but silk fabrics and silk yarn
could only be imported at a general ad valorem tariff rate of 50 percent and
30 percent, respectively (Mason 1910).
From 1870 to 1908, a change occurred in the composition of the US. silk
import market with regard to the composition of market shares held by the
principal exporting countries (see Table 3.2). The opening of the transcon­
tinental railway in 1879 in the United States lowered the export freight rates
of Asian raw-silk exports to the silk-weaving industries of the East Coast,
leading to market share increases for China and Japan. In addition, during
Table 3.1
Direction of Japan's Raw-Silk Exports (%) a
Years United
States France United
Kingdom Others Total
1873-75 1.7 42.0 41.4 14.9 100.0
1876-80 19.1 47.5 27.5 5.9 100.0
1881-85 42.2 46.4 10.7 0.7 100.0
1886-90 57.7 36.6 3.5 2.2 100.0
1891-95 56.0 37.6 1.5 4.9 100.0
1896-1900 59.2 32.5 0.6 7.7 100.0
1901-05 65.4 19.8 0.2 14.6 100.0
1906-10 71.1 18.7 0.1 11.1 100.0
Note:
a. In terms of value of exports.
Source: Computed from Ministry of Finance (Okurasho), Dainihon Gaikoku Boeki Nempyo
(Foreign Trade Annual of the Empire of Japan), various issues.

Raw-Silk Exports and Economic Development 55
the period between 1870 and 1880, the share of Japan's exports grew rapidly
at the expense of China's share, mainly as a result of changes in the pattern
of U.S. demand. In the United States, technological progress had occurred
in the reeling and weaving stages, including a shift from hand to power looms.
In addition, consumer demand shifted to broad fabrics, which called for raw
silk of a better, standardized quality. Although European steam filature was
best suited to meet these requirements, an increase in the European export
supply would not have been possible without a price increase since the indus­
try had to compete for scarce labor. As a result, the growth of European coun­
tries' share in the US. market was minimal.
At the same time, rapid technological innovations in the sericulture and
silk-reeling industries in Japan helped to increase the competitiveness of
Japan's raw-silk exports. Technological innovations, such as those allowing
for silkworm breeding in the fall as well as in the spring and for the artificial­
ly controlled environments appropriate for silkworm breeding, began to be
devised in the late 1870s and spread rapidly. Thus, by the early 1880s the co­
coon requirement per unit of filature—one measure of raw-silk export costs—
was reduced by about 30 percent. 1
Other improvements contributed to the increased productivity of Japan's
silk industry. In the silk-reeling industry, the traditional hand and sedentary
machines gave way to threading machines. Moreover, steam-filature technol­
ogy was introduced from Europe and assimilated into the Japanese mode of
production, quickly surpassing hand filature by the end of the 1890s. The
share of steam filature in total production in 1890 was 40 percent but rose
to 70 percent by the middle of the next decade. Since steam-filature output
was, for the most part, destined for the export market, its share of Japan's
total raw-silk exports was 70 percent in the mid-1890s, rising to 90 percent
in the early 1900s. In contrast, Chinese producers continued to rely on tradi­
tional hand-loom technology and were late in introducing the steam-filature
Table 3.2
Sources of U.S. Silk Imports (%) a
Years Japan China Italy France
1870-71 11.0 28.5 — 7.1
1879-81 28.6 54.7 0.1 9.6
1889-91 53.9 22.1 16.5 4.4
1906-1908 56.5 18.7 21.8 3.0
Note:
a. In terms of volume.
Source:. Mason (1910).

56 Industry Growth and International Trade
technique. As a result, the share of steam filature in China's raw-silk exports
only rose from 35 percent in 1898 to 42 percent in 1908.
A CONSTANT MARKET SHARE (CMS) ANALYSIS
OF EXPORT GROWTH
In the following sections, the growth mechanism of Japan's silk exports
to the United States (discussed above) is analyzed. By disaggregating Japan's
total volume of raw-silk exports to the United States (X) into hand-filature
exports (A',) and steam-filature exports {X 2) and denoting their shares as s p
s2, respectively, in total U.S. silk import volume (M ]t M 2), the following
equation can be derived:
X=X^X 2
=s lM l + SJMJ
Given the change in exports between time 0 and 1 (that is, AX = X l -X°),
the above equation can be rewritten as follows:
2 2 2
AX= £5° AM ( + EWJAJ, + LAM. As f (3.1)
1=1 /=i 1=1
2 2 2
= s°AM + (L s° AMj – s°AM) + E M? As,- + £ AM; As,-
/= 1 / = 1 1 = 1
where AM and As are changes in import volume and import shares, respec­
tively, and where M-M x +M 2 and X=sM.
The first term on the right-hand side of equation 3.1 (s°AM) represents the
market-demand-growth effect—that is, under the assumption of constant
Japanese shares, exports increase as a result of an increase in total U.S.
imports of a given commodity. The second term (Ej^sjAMj – s°AM)
represents the commodity-composition effect, where even assuming constant
shares of hand and steam filatures in the US. market, an increase in import
demand for silk is greater than demand for total imports, thereby inducing
an increase in Japan's silk exports. The third term (Ef =1 M^As^ shows the
competition effect—that is, in the absence of any increase in US. imports
of the product, an increase in the share of Japan's exports will increase total
silk exports. The fourth term (£^ =1 AM ( As, ) represents the interaction
term—that is, the effect of the combination of changes in Japan's share and
changes in U.S. import demand. However, the fourth term, like the third term,
may be considered a competition effect, since its value increases whenever
an increase in Japan's share accompanies a large increase in U.S. market
demand.

Raw-Silk Exports and Economic Development 57
Using equation 3.1, the growth of Japan's raw-silk exports to the US. mar­
ket can be disaggregated into the market-demand-growth effect, the commod­
ity-composition effect, and the competition effect. Since the effects are
differentiated according to whether market shares are assumed to be constant
or variable, this approach is referred to as the constant market share (CMS)
analysis. 3
Table 3.3 reveals certain trends in the five-year average of Japan's silk ex­
ports to the United States and of total U.S. silk imports from 1876 to 1910.
In analyzing these fluctuations the focus is long-term trends; short-term
changes are disregarded. Notwithstanding the arbitrary nature of the five-
year divisions, the sources of export growth can be identified. To do this, as­
sumptions had to be made with regard to the published data on total silk
exports when disaggregating them into hand- and steam-filature exports. 4
Beginning in the second half of the 1870s and continuing through the first
half of the 1880s, U.S. silk imports grew at a rapid rate of 12.5 percent per
year. In subsequent years, growth of U.S. silk imports stabilized at 6 to 8 per­
cent annually. However, during the same time period the share of steam fila­
ture rose from 14 to 86 percent of total silk imports, indicating that
hand-filature imports were being rapidly crowded out of the market.
During this period, Japan's share of U.S. silk imports rose from 29 to 56
percent. Furthermore, in response to the dramatic change in U.S. import de­
mand (which was mentioned above), Japan's steam-filature export share
climbed from 18 to 95 percent. Still, Japan's export growth did not proceed
smoothly. Until the second half of the 1880s, annual export growth rates as
high as 15 to 19 percent were posted, but in the following ten-year period
(1886-95), annual export growth declined to 4 to 5 percent. During the peri­
od 1896-1900, the annual growth of exports climbed back to 8 percent. There
were several factors behind these fluctuations in the growth of exports to the
US. market.
Applying the data in Table 3.4 to equation 3.1, a CMS analysis was con­
ducted for each period. It is evident that over the entire period from 1876
to 1910, the market-demand-growth effect was large and represented about
half of Japan's export increase. This means that the growth in U.S. import
demand induced a major part of the growth of Japan's raw-silk exports to
the United States during that period. Thus, the slowdown in growth of the
US. import market in the 1890s was one of the causes of the sluggishness
of Japan's export growth. There were, however, other forces at work.
The commodity-composition effect became positive and remained that way
from the first half of the 1880s to the 1890s. Since the growth in U.S. silk
imports mainly reflected the growth of steam-filature imports, this effect was
positive during the period when Japan's share of steam filature exceeded that
of hand-filature imports. In contrast, the commodity-composition effect was
negative in the 1880s when the share of hand-Filature imports rose. After 1890,

Table 3.3
Japan's Raw-Silk Export Growth (L '000, five-year average) 3
Japanese raw-silk Japan's share
U.S. raw-silk imports export growth in U.S. imports (Vo)
Hand Steam Hand Steam Hand Steam
Years Total filature filature Total filature filature Total filature filature
1876-80 1,874 1,607 267 547 450 97 29.2 28.0 36.3
(12.5) (19.0)
1881-85 3,507 2,198 1,309 1,418 991 427 40.4 45.1 32.6
(7.9) (14.4)
1886-90 5,193 2,324 2,869 2,918 1,310 1,608 56.2 56.4 56.1
(6.5) (4.4)
1891-95 7,175 2,368 4,807 3,628 814 2,814 50.6 34.4 58.6
(5.4) (4.5)
1896-1900 9,383 2,590 6,793 4,541 831 3,710 48.4 32.1 54.6
(8.3) (8-9)
1901-05 14,241 2,831 11,410 7,089 1,003 6,086 49.8 35.4 53.3
1906-10 19,645 2,644 17,000 10,906 564 10,342 55.5 21.3 60.8
Note:
a. Figures in parentheses show percentage rate of increase to the next five-year period.
Source: Niyomka (1975: table A-8).

Raw-Silk Exports and Economic Development 59
when only steam-filature exports showed an increase, the commodity-
composition effect again showed a positive contribution to the growth of
Japan's exports.
Interpretation of the competition effect, however, poses some problems.
Although largely positive in most periods, this effect was negative from the
second half of the 1880s to the beginning of the twentieth century. In the
1880s, productivity and quality improvements were realized particularly in
the hand-filature industry, whereas technological progress and productivity
increases were realized in steam filature in the first decade of the 1900s. How
then could the competition effect be negative during these periods? One pos­
sible explanation is that Japan's filature exports might have been put at a
relatively disadvantageous position vis-a-vis China's filature exports as a result
of Japan's adoption of the gold standard in 1897. However, it can also be
argued that during the preceding decade the silver depreciation that took place
made Japan's exports more competitive. The effect of this depreciation on
Japan's export growth performance in the United States is discussed in the
next section.
Table 3.4
Constant Market Share Analysis of Japan's Silk Exports
to the United States (t '000) a
Increase in
Japan's silk Commodity-
exports Demand- compo­ Compe­ Inter­
to the growth sition tition action
United States effect effect effect effect
Periods (1) (2) (3) (4) (5)
1876-80 to 871 477 67 265 62
1881-85 (54.8) (7-7) (30.4) (7.1)
1881-85 to: 1,500 681 -116 556 379
1886-90 (45.4) (-7.7) (37.1) (25.2)
1886-90 to 710 1,114 -2 -440 38
1891-95 (156.9) (-0.3) (-62.0) (5-4)
1891-95 to 913 1,117 123 -253 -74
1896-1900 (122.3) (13.4) (-27.7) (-8.0)
1896-1900 to 2,548 2,351 247 -3 -57
1901-05 (92.3) (9.7) (-0.1) (-1.9)
1901-05 to 3,817 2,691 222 467 437
1906-10 (70.5) (5.8) (12.2) (115)
Note:
a. Increase of five-year averages in column 1 is decomposed to individual contributions in columns
2 to 5 with percentage shares in parentheses.
Source: Computed by applying equation 3.1 to the figures in Table 3.3.

60 Industry Growth and International Trade
THE EFFECT OF SILVER DEPRECIATION
Until 1897, when Japan changed to the gold standard, silver was used for
all overseas transactions by Japan. When the United States and the major
European countries adopted the gold standard in the 1870s (changing from
a bimetallic standard in many cases), the value of silver fell relative to gold,
as did the Japanese yen. During the decade before the adoption of the gold
standard in Japan, the exchange rate of the yen relative to the dollar had
declined by 33.2 percent, and this is thought to have encouraged the growth
of Japanese exports. 1 As is often observed in contemporary developing coun­
tries, an overvalued currency tends to restrain export growth. In the case of
Japan prior to 1897, the rapid decline in the value of the yen was the result
of changes in the international Financial system rather than a planned govern­
ment policy.
How did the depreciation of the yen affect the growth of Japanese silk ex­
ports? Compared to Japan's European competitors in the U.S. market, such
as Italy and France, which were already on the gold standard, Japan was in
a more favorable position. However, Japan's main competitor, China, re­
mained on the silver standard even after Japan had adopted the gold stan­
dard. For this reason, Japan's exports did not gain from the silver depre­
ciation. 6 As was also demonstrated in the earlier CMS analysis, there was no
significant increase in Japan's share of the U.S. market at the time of the
depreciation of the yen. In order to explain this apparent contradiction, the
following hypothesis is presented. The silver depreciation, rather than mere­
ly favoring Japan's competitive position through a larger share of the US.
market, lowered the Japanese and Chinese raw-silk export price (in dollar
terms) and hence also the price of raw silk in the US. market. This, in turn,
reduced the price of silk fabric, encouraged an increase in its demand, and
thereby ultimately induced an increase in imports of silk Filature. It can there­
fore be said that the impact of silver depreciation showed up more in the
market-demand-growth effect than in the competition or increase-in-share
effect. This hypothesis can be verified empirically using import and export
functions:
M=a 0 + a lY+a 2(P s/P) (3.2)'
X=b Q + b xM+b 2{Pj/fy (3.3)
X/M=c 0 + Cl(P/P s) (3-4)
where M refers to the volume of U.S. raw-silk imports; X is the volume of
Japan's raw-silk exports to the United States; Y is US. income, which is a
determinant of the demand for silk fabrics; P s is the price of raw silk in the

Raw-Silk Exports and Economic Development 61
United States; P- is the price of Japanese raw silk in the United States; and
P is the general price level in the United States.
Equation 3.2 shows that US. imports of raw silk are determined by the
price of raw silk relative to other commodities in the U.S. market and by US.
income levels. Note that a more accurate formulation would have as the de­
pendent variable silk fabric consumption rather than imports of raw silk.
However, the demand for raw-silk imports is a valid proxy since the U.S.
processed imported raw silk into silk fabrics. With a statistically significant
price coefficient, a 2> tne effect of the silver depreciation through the change
in /Jean be shown. In addition, it is possible to determine the relative im­
portance of the silver depreciation as a source of growth by computing the
income and price effects. Equations 3.3 and 3.4 are Japanese export func­
tions to the United States. Equation 3.3 expresses Japanese raw-silk exports
to the United States as a function of the total import volume of raw silk of
the United States and Japan's export price relative to the average U.S. price
for silk. Equation 3.4 hypothesizes that Japan's share of the U.S. raw-silk
import market is a function of Japan's export price of silk.relative to its price
in the United States.
Table 3.5
Estimates of US. Import and Japanese Export Functions of Raw Silk 3
M Constant Y P5/P R 2 DW
Eq. 3.2
Eq. 3.2* 0.9347
11.0703 0.9268
(20.60)
0.8405
(21.56) -1.1445
(3.72)
-2.1219
(6.68) 0.9498
0.9774 1.49
0.61
X Constant M R 2 DW
Eq. 3.3 0.0007 1.7003
(11.24) 0.5138
(0.92) 0.8273 0.67
X/M Constant P/P s R 2 DW
Eq. 3.4 0.3516 1.6467
(2.61) 0.1434 0.76
Note:
a. Equation 3.2 was estimated based on annual data for 1869-1910; equation 3.2* was estimat­
ed using the three-year moving average for the same period; and equations 3.3 and 3.4 were
estimated using annual data for the period 1875-1910. All equations were estimated using
ordinary least squares regression in log-linear form with /-values in parentheses.

62 Industry Growth and International Trade
Table 3.5 shows the results of regression analysis for the parameters of equa­
tions 3.2 through 3.4. Equation 3.2 was estimated using annual data from
1869 to 1910 and a three-year moving average of the data over the same peri­
od. The results are reported in Table 3.5 as equations 3.2 and 3.2*, respective­
ly. As proxies for Y and P, the volume of U.S. silk fabric production and the
US. wholesale price index were used. An interesting result of the analysis is
that the price elasticities doubled when the three-year moving average data
were employed; this implies that the price effect is more pronounced in the
long run. On the other hand, the price elasticity in equation 3.3 is positive
but not statistically significant, whereas it is both positive and significant in
equation 3.4. The incorrect signs of the price elasticities in equations 3.3 and
3.4 may be attributed to the fact that the unit import price used is a weighted
average of hand- and steam-filature prices, and therefore the relative price
variable may not be directly linked to the changes in shares. While the use
of disaggregated hand- and steam-filature price data may have yielded sig­
nificantly negative price elasticities, the lack of reliable data on a disaggregated
basis precludes such reestimation. Furthermore, disaggregation is meaning­
less in the case of equation 3.2.
However, the significantly negative price effects observed in equation 3.2
support the hypothesis that the silver depreciation ultimately led to an in­
crease in demand for silk fabric and silk filature. Moreover, the differentia­
tion of equation 3.2 with respect to time yields the following relationship
between the growth rates of the individual variables:
M = 0.9268 Y- 1.145 {P s'/P),
Substitution of the average annual rate of change during the period 1887-97
into Y and P s/P (that is, 4.582 and -2.91%, respectively) shows the contri­
bution of the income effect (4.24%) and the price effect (3.33%) in stimulat­
ing the increase in total U.S. imports of raw silk during the period.
In addition, during the ten years preceding the adoption of the gold stan­
dard, the yen rate declined from 77 to 49 dollars per 100 yen, an average annual
decrease of 4.52 percent. Assuming that the Japanese raw-silk export price
could be used as a proxy for the U.S. import price, Japan's export price (in
yen) would have risen by 1.61 percent relative to the U.S. general price level
(in US. dollars). Thus, without the silver depreciation the relative price of
silk in the US. market would have risen by 1.61 percent and would have re­
strained as much as a third of the actual increase in U.S. imports of Japanese
raw silk.
In the foregoing analysis, it was assumed that changes in Japan's raw-silk
export price were exogenous. However, since changes in supply conditions
also determine prices, it is necessary to employ a demand-supply-equilibrium
model to properly evaluate such changes.

Raw-Silk Exports and Economic Development 63
Figure 3.1
Mechanism of Silk Export Growth

F 0 F, F 2 G 0 G 2 G 3
Volume
Figure 3.1 illustrates the influence of several factors determining the in­
crease in Japan's silk exports to the United States. On the left-hand side of
the figure, S A and S B represent the long-run export supply curves of two com­
peting exporter countries, countries A and B, which when summed up horizon­
tally give the total export supply curve S. Given the short-run demand curve
D 0, the point of intersection E ] determines the initial equilibrium in the im­
port market. An increase in income causes D 0 to shift to D V moving the
equilibrium point to E 2. Both countries' exports increase as a result, but the
magnitudes of the increase (F 0FLT F 0F2) differ depending on the elasticities
of their long-run supply curves. For instance, relative to that of European
exporting countries, Japan's long-run supply curve is more elastic, causing
its share of the expanding U.S. market to rise. Country A's and B's export
supply prices equalize at E 2.
On the right-hand side of Figure 3.1, the effects of the yen depreciation
can be examined. Assuming that exporting countries A and B are both on
the silver standard while the importing country is on the gold standard,
depreciation of silver would tend to lower the exchange rate of the exporting
countries and shift the total supply curve to S'. The equilibrium point would
then move to E Y The increase in imports from C 0 to G 3 can be broken down

64 Industry Growth and International Trade
into the income effect (G 0-*G 2) and the price effect (G 2-*G 3). Given the
previous estimate of the import function, in the absence of an exchange
depreciation, there would have been no shift in the supply curve, and at the
equilibrium point E 2 the negative price effect (G 2-*G,) would have cancelled
part of the potential income effect.
To summarize, there were several factors affecting the export growth of
Japan's raw silk. Essentially, the increase in U.S. demand for silk fabrics
brought about an increase in the demand for raw silk, and hence, for Japanese
raw-silk exports. However, compared with other exporting countries, such as
China and Italy, the expansion in Japan's share of the U.S. import market
for raw silk was due to the high long-run supply price elasticity for Japanese
raw silk. This resulted from the growth in raw-silk production capacity brought
about by the technological progress that had taken place in the industry. In
addition, although the price of raw silk (in terms of yen) rose under the in­
flationary conditions of the Japanese economy at the time, this potential
dampening effect on export growth was more than offset by the fall in mar­
ket prices for exports caused by the depreciation of the silver-based yen, which
followed the fall in the value of silver relative to that of gold. Thus, Japan's
competitive position in the U.S. market improved when compared to com­
peting gold-standard countries, such as Italy and France. However, since its
major competitor, China, also used a silver-backed currency, the deprecia­
tion of silver stimulated Japan's silk export growth not so much in terms of
share expansion but more through a lowering of the market price, which
prompted an overall increase in import demand for raw silk.
THE CONTRIBUTION OF SILK EXPORTS
TO JAPANESE ECONOMIC DEVELOPMENT
This chapter concludes with explanations of five identified effects of the
growth of raw-silk exports on Japan's economic development.
Foreign-Exchange-Earnings Effect
The most evident economic contribution of growth in raw-silk exports is
its role as a major source of foreign exchange earnings (see Table 3.6). Be­
tween 1868 and 1881, raw-silk exports earned approximately 107 million yen,
which is about 35 percent of total export earnings. From 1881 to 1893, the
value and share of Silk exports increased to 253 million yen and 36 percent,
respectively. Import demand during the initial period of economic growth
was large. During the first period in particular, total Japanese imports reached
a value of 473 million yen, and huge deficits were recorded. Although the
deficits were fully financed through the outflow of gold and silver coins that

Raw-Silk Exports and Economic Development 65
had been saved rather than used to pay for imports during the Edo period
(1601-1867) when trade was minimal, the government encouraged exports to
compensate for the huge current account deficits, and raw silk was always
considered an important commercial item to be promoted. Thus, the contri­
bution of raw-silk exports, which represented some 35 percent of total for­
eign exchange earnings from exports, was substantial.
Income-and-Output-Generation Effect
Although it is impossible to estimate the income generated by raw-silk
production and exports, an analysis of the data in Table 3.7 reveals the extent
of the industry's contribution to increases in total production in both the
agricultural and manufacturing sectors. The growth of raw-silk exports caused
a rapid expansion of the sericulture and silk-reeling industries. The share of
the sericulture industry in agricultural production was 8 to 9 percent and con­
tributed 11 to 16 percent of the income from agricultural production. However,
as noted earlier, sericulture competed with wheat production for available
resources so that the more resources devoted to sericulture implied less
resources would be devoted to the production of wheat. If the decline in wheat
production that results from the growth of sericulture is subtracted, the final
contribution of sericulture is 4 percent lower.
In the.manufacturing sector, the shares of the silk-reeling and -weaving in­
dustries, as well as their rates of contribution to the growth of manufactur­
ing production, were even larger than sericulture's contribution to agricultural
production. From the 1880s to the second half of the 1890s, these two indus­
tries were responsible for approximately 21 to 25 percent of the increase in
total manufacturing output.
It should be noted that the distribution of production from sericulture, silk-
reeling, or silk-weaving in the country or even in the cities was not uniform.
Producers were unevenly concentrated in certain prefectures such as Nagano,
Table 3.6
Foreign Exchange Earned Through Silk Exports (¥ *000) a
Years Exports (f.o.b.) Silk exports Imports (c.i.f.) Trade balance
1868-81 302,243 107,093 472,927 (170,684)
(35.4)
1882-93 699,006 253,209 721,810 (22,804)
(36.2)
Note:
a. Figures for the trade deficit and silk exports' share of total exports are in parentheses.
Sources: Yamazawa and Yamamoto (1979a: table 2-2); Toyo Keizai Shin posh a (1935).

66 Industry Growth and International Trade
Gifu, and Fukushima. Efforts to introduce sericulture were widespread in
mountainous regions that were unsuitable for rice cultivation. For logistical
reasons, silk-reeling factories were set up near sericulture areas. Since there
was no other major agricultural or industrial production in these regions, the
contribution of the sericulture and reeling industries to regional agricultural
and industrial production would have greatly exceeded the national statistics
cited earlier. The regional economic-development-inducement effect must have
been overwhelming.
Employment Effect
In the labor-surplus economies of contemporary developing countries, the
labor-absorption effect of the modern industrial sector is one of the consider­
ations in evaluating the effects of industrialization. Since the filature indus­
try is highly labor-intensive, it generated a great deal of employment relative
to other industries. In the first decade of the twentieth century, employment
Table 3.7
Production Effects of Silk Exports (°7o) a
Contribution of sericulture Contribution of silk reeling
to agricultural production to manufacturing production
Periods S/A S/A A AS'/AA b R/M AR/AM
1876-80 to
1881-85 7.5 16.6 7.1 9.2 6.0
1881-85 to
1886-90 8.7 c c 8.5 24.9
1886-90 to
1891-95 8.6 13.6 11.4 11.9 21.6
1891-95 to
1896-1900 9.9 9.7 7.5 15.3 15.9
1896-1900 to
1901-05 9.8 11.4 9.2 15.6 3.6
1901-05 to
1906-10 10.1 15.1 12.3 13.6 13.5
Average 9.1 13.3 9.5 11.5 14.4
Notes:
a. S, A, /?, and M denote sericulture, agriculture, silk reeling, and manufacturing, respectively.
A denotes the increase from the average over the first five years to that over the last five years
for individual periods.
b. AS' is AS adjusted for the hypothetical decrease of wheat production.
c. — refers to a decrease in both agricultural and sericultural production.
Source: Agricultural production data are from Henmi (1969), and manufacturing production
data are from Niyomka (1975).

Raw-Silk Exports and Economic Development 67
in this industry made up some 27 to 29 percent of all manufacturing sector
employment. In addition, since the filature industries were not normally lo­
cated in the large cities, their capacity to absorb rural labor, which enjoyed
few employment opportunities, was great. Although rural cottage industries
did not create many jobs in such prefectures as Gunma and Fukushima, where
hand-filature production was concentrated, a significant amount of employ­
ment was provided by the steam-filature factbries of Nagano and Gifu prefec­
tures (Kajinishi 1964).
Capital-Accumulation Effect
The mechanism of industrial development depends on capital accumula­
tion for investment, which in turn generates profits for investment in the de­
velopment of other industries. In the silk-reeling industry, whose exports
earned a third of the nation's total export earnings, active and profitable bus­
iness enterprises were set up. Were these profits reinvested in other industries
that in turn contributed to sustained industrialization?
Initially, since steam-filature firms were small enterprises that often lacked
funds for both equipment investment and working capital (for example, for
the purchase of cocoons), they did not generate surplus capital for invest­
ment in other industries. Rather, raw-silk merchants, export traders, and lo­
cal banks were more likely to provide the working capital for the silk-reeling
industry. Regional banks, such as the Fourteenth National Bank in Matsumoto
and the First National Bank in Ueda (both of Nagano Prefecture), success­
fully collected funds from local raw-silk traders. Between these regional banks
and the banks in Yokohama, such as the Yokohama Foreign Exchange Com­
pany and the Second National Bank, a flow of raw-silk exports was estab­
lished and was matched by a reverse flow of abundant financing. Thus,
although this phenomenon is difficult to demonstrate empirically, raw-silk
exports did, in a sense, contribute to the development of modern banking. 1
It can be further inferred that at least some of the funds used to establish
new industries became available as an indirect result of these developments.
Linkage Effect
As observed in the foregoing sections, although raw-silk exports had a large
foreign-exchange-earnings effect and had contributed greatly to the growth
of agricultural and manufacturing production, the linkage effect (in terms
of inducing the development of other industries) was limited. Since silk was
a luxury item, there was not much room for growth in demand. And, apart
from sericulture, there was no need for large investments in a related indus­
try. Moreover, because production was based on traditional production tech-

68 Industry Growth and International Trade
niques with their limited scale economies, the possibilities for technological
progress and productivity improvements were constrained accordingly.
Although government-managed factories introduced steam-filature tech­
nology in the early 1870s without much adaptation, its application was not
widely used because of the high price of imported machinery. In both the
raw-silk and silk-weaving industries, imported machinery and technology be­
gan to spread to privately owned firms only when the imports were adapted
to Japan's factor endowment conditions in terms of labor intensity, small-
scale production, and the already available indigenous technology (Ono 1968).
In contrast, it was the cotton industry, which from the start applied modern
production techniques and realized scale economies, that would play the key
role in Japan's light industrialization.

Four
Catching-Up Product Cycle
Development in the Textile Industry
The textile industry is usually the first modern industry to be introduced by
late-industrializing countries. In general, domestic demand of a late-industrial­
izing country is initially met by production based on existing indigenous tech­
nology. Once quality imported goods made with modern production
technology are introduced and gain wide acceptance, demand shifts from tradi­
tional to modern goods. Through imitation, domestic production of modern
goods begins. In the textile industry, domestic production of modern goods
is easily assumed because indigenous technology exists in most late-
industrializing countries. Technology absorption is relatively easy, and be­
cause the industry is characterized by labor-intensive production, low-wage,
late-industrializing countries can easily achieve comparative advantage. The
case of Japan was no exception, and after the modern cotton-spinning in­
dustry was introduced in the 1860s, the development pattern followed the typi­
cal path from import substitution to export expansion.
This chapter analyzes the development of the cotton industry in Japan and
explains the role of the catching-up product cycle (CPC) model of develop­
ment in the textile industry. It includes a discussion of the industry's diver­
sification process from woven fabrics to rayon and synthetic textiles, and the
induced development of such related industries as textile machinery and syn­
thetic dyestuffs. It also explains the diversification process. Finally, the CPC
theory is considered not only as it applies to commodity trade, but also in
explaining the relationship between overseas investment and technology trans­
fer and changes in commodity trade.

70 Industry Growth and International Trade
THE STRATEGIC ROLE
OF THE COTTON TEXTILE INDUSTRY
The textile industry played a leading role in Japan's industrial growth be­
fore the Second World War. Even if raw silk were excluded, textiles made up
a third of Japan's total industrial production and half of industrial exports
during the period 1890-1930. In addition, raw material imports of the textile
industry made up 70 percent of the country's total imports of raw materials
(see Table 4.1). Although the textile industry included the wool and rayon in­
dustries, its core was the cotton industry; during the entire prewar period be­
fore 1940 the woolen fabrics trade showed net imports, and in the 1920s the
rayon industry was just beginning to grow.
The cotton industry was the first modern industry based on large-scale
production techniques. Although its production technology was perfected
around the middle of the nineteenth century, the geographical areas involved
in this technological revolution were limited. Nevertheless, with its low-wage,
high-quality labor force, Japan was able to achieve comparative advantage
in this labor-intensive industry and overtake Great Britain in the 1920s. In
the second half of the 1960s, however, when wages rose and a labor shortage
set in, Japan lost that comparative advantage to other late-developing coun­
tries. As a result, from 1970 onward, imports of textiles exceeded exports of
textiles. Thus, the Japanese cotton industry has passed through the complete
life cycle of an industry.
Table 4.1
The Textile Industry's Share of Japanese Manufacturing Activity (%)
Period Share in
manufactured
exports a Share of textile
materials in raw
material imports* Share of textile
output in
manufacturing
output b
1882-1891 35.2 27.8 42.4
1892-1901 52.0 34.7 78.5
1902-11 50.0 26.8 79.2
1912-21 51.3 30.1 80.8
1922-31 55.5 30.1 71.8
1930-39 43.7 23.7 60.7
1951-55 41.7 18.3 45.5
1966-70 14.0 8.2 12.1
Notes:
a. Raw silk is excluded from both the denominator and numerator.
b. Raw material imports include nonedible agricultural products, forestry products, and minerals.
Sources: Yamazawa and Yamamoto (1979a: tables 1 and 2); Shinohara (1972: table 1).

Catching-Up Product Cycle Development 71
Figures 4.1 and 4.2 chart the CPC development of cotton yarn and cotton
fabrics, using the values of domestic production, exports, and imports in con­
stant 1934-36 prices. Even within the cotton yarn and cotton fabrics groups,
there are great differences among products in terms of quality and price per
unit (for example, per tonne of cotton yarn or per square meter of cotton
fabric). Before the. Second World War in particular, the composition of
products shifted from low- to high-quality goods. Disregarding such quality
and price differentials but still avoiding the simple summation of quantita­
tive data, a composite weighted average index (in 1934-36 prices) based on
unit price indices of individual products was constructed and used to deflate
the total value. However, for the post-Second World War period, values were
obtained by using a simple summation of quantitative data multiplied by
1934-36 prices.
COTTON INDUSTRY DEVELOPMENT
In 1859, with the opening of Japan to foreign trade, cotton yarn and cot­
ton fabrics produced with modern production technology began to be im­
ported. Imports rapidly grew because domestically produced fabrics based
on indigenous technology were of inferior quality. By 1870, imported cotton
yarn and fabrics constituted 36 percent of Japan's total imports. Because of
foreign exchange constraints, rapid import substitution based on modern
production technology was encouraged. By 1886, domestic import substitu­
tion of cotton yarn was achieved when half of total domestic demand was
being met by domestic production (see Figure 4.1). Import substitution of
cotton fabrics (1878) preceded that of cotton yarn (1886). In terms of the im­
port/export ratio, however, the opposite was true. Whereas cotton yarn ex­
ports matched imports in 1895, this feat was not achieved by cotton fabrics
until 1909. The cotton yarn industry reached its mature stage in 1960, and
the cotton fabric industry in 1963, as indicated by the slowdown in domestic
production growth. The reverse import stage began in 1969 for cotton yarn
and in 1972 for cotton fabrics.
As mentioned previously, drastic changes occurred in the product compo­
sition of the cotton industry. Cotton yarn production began with the easily
absorbed low-count technology and gradually expanded to high-count tech­
nology. Similarly, the production of cotton fabrics diversified from gray cot­
ton fabrics to colored and processed cotton fabrics. Export growth, like import
substitution, followed the same pattern. Initially, however, the production of
cotton yarn and cotton fabric was not integrated and thus the two industries
followed separate paths of CPC development. Cotton fabric production in
each of the country's traditional weaving areas was easily able to expand as
a cottage industry based on traditional technology and imported yarn. Be-

Figure 4.1
Catching-Up Product Cycle Development of Cotton Yarn in Japan

1B80 1B90 1900 1910 1920 1930 1950 1960 1970 1980
Year
Notes: Seven-year moving averages before the Second World War and five-year moving
averages after the Second World War.
Source: Estimated by the author. See Yamazawa (1984) for details of the estimation.

Figure 4.2
Catching-Up Product Cycle Development of the Cotton Textile Industry
in Japan
1,000
100
7 'io
0.1 i • i i i i i i i i
Production •Ni-i 1 r T r
Domestic
demand

Exports •
J I I u J I L
1880 1890 1900 1910 1920 1930 1950 1960 1970 1980
Year
Note: Seven-year moving averages of domestic demand, production, exports and imports of
cotton fabrics before the Second World War. Five-year moving averages after the Second
World War.
Source: Estimated by the author. See Yamazawa (1984) for details of the estimation.

74 Industry Growth and International Trade
ginning in 1880, cotton yarn produced in large spinning firms began to com­
pete with imported yarn; these firms were able to rapidly achieve export ex­
pansion. The development of Japan's cotton industry is clearly revealed by
the changes in the activity of large-scale spinning firms. 1
From Initial Introduction to Import
Substitution in Cotton Yarn: 1883-95
Private firms first began producing cotton yarn in the mid-1870s when the
government sold its model factories to the private sector. But because of their
small size (2,000 spindles), these firms could not compete with imported cot­
ton yarn. In 1883, Osaka Boseki began spinning production using steam en­
gines and 10,500 spindles. Its success encouraged the entry of an additional
20 large-scale spinning firms into the industry in the following decade. Total
spinning capacity rapidly grew from 80,000 spindles in 1886 to 350,000 spin­
dles in 1890, thus achieving import substitution in low-count cotton yarn.
Exports to Korea and China, however, met stiff competition from Indian yarn.
And since cotton yarn of more than 20-count could not be produced domes­
tically, imports continued. In 1892, imports of better-quality Indian raw cot­
ton began replacing domestic raw cotton as well as raw cotton imported from
China. With the decline in domestic cotton cultivation, cotton production
based on indigenous technology rapidly disappeared.
Cotton Yarn Export Expansion and
Cotton Fabric Import Substitution: 1895-1914
By 1897, domestic spinning capacity reached 1,014,000 spindles. The large
spinning mills were able to produce high-count yarn from American raw cot­
ton, and the high-count yarn replaced American and English cotton yarn im­
ports. Most of the spinning mills, however, continued to produce low-count
yarn. As much as 80 percent of the cotton yarn produced was low-count yarn,
whose exports spread from China to Southeast Asia and directly competed
with Indian cotton yarn.
The large spinning firms began to produce fabrics using their own cotton
yarn. Such integrated production of yarn and fabrics continued to grow until
the 1900s, catering not only to the domestic market but also to China. By
1909, the entire cotton fabric industry had reached the export stage, and in
1913, domestic spinning capacity reached 24 million spindles.
Cotton Fabric Export Expansion: The Interwar Period
Cotton yarn exports began to decline and cotton fabric exports began to
expand with the outbreak of the First World War. In 1913, the large spinning

Catching-Up Product Cycle Development 75
mills began to move to China to take advantage of the cheap labor. In many
ways, the war favored the development of the Japanese cotton industry and
helped the industry reach its mature stage during the two decades after the
war. During this period Japanese cotton fabrics replaced English cotton in
the Chinese market. The share of Japanese cotton rose from 20.2 percent in
1913 to 66.9 percent in 1926, while the share of English cotton declined from
53.3 percent to 24.0 percent. The war also induced an export boom. Thus
small and medium-sized weavers from the traditional weaving regions shifted
production from domestic to export markets. In addition, regional produc­
tion of gray cotton fabrics as well as bleached cotton and colored cotton cloth
expanded, and as production expanded, the goods were then exported to
China. Because the war halted imports of European dyestuffs and textile
machinery, import substitution of synthetic dyes and textile machinery was
necessary. Although these import substitutes met stiff competition from im­
ported European goods, complete import substitution was nevertheless
achieved in the 1920s and 1930s. By the end of the 1920s, Japan had emerged
as the world's major exporter of cheap cotton fabrics and, as a consequence,
had to face discriminatory import barriers, particularly in British Common­
wealth countries.
Slowdown in Growth and the Reverse
Import Stage: The Postwar Period
Cotton fabric production and exports climbed back to their pre-Second
World War levels around 1955 and peaked at the start of the 1960s, after which
they slowed down. In the 1960s exports to import-substituting countries be­
gan to decline, and Japanese imports began to grow. This occurred because
the Japanese economy shifted from a labor-surplus economy to a labor-scarcity
one. As wages rose, comparative advantage was lost in labor-intensive produc­
tion. Accordingly, textile production shifted toward more capital-intensive
production lines. Concurrently, domestic firms also began to establish sub­
sidiary factories overseas to take advantage of low-wage labor, particularly
in Southeast Asia. In turn, this encouraged import substitution and export
expansion in the developing countries. Thus Japanese imports expanded in
the 1970s as the industry entered the reverse import stage of development,
which is the final stage in the life cycle of the Japanese cotton industry.
The Mechanism of CPC Development
What mechanism underlay the successful CPC development of the Japanese
textile industry?
First, the existence and growth of sizeable domestic demand certainly con­
tributed to the start and rapid increase of domestic production. It is estimated

76 Industry Growth and International Trade
that domestic consumption was as high as 40,000 tons in 1880 and increased
annually by an average of 2.5 percent during the period 1880-1940. Rapid
export growth added to domestic consumption. Compositional changes in
cotton fabric exports every 10 years (see Table 4.2) illustrate the expansion
of the export market, which shifted from Korea and China to Southeast Asia
and then to the rest of the world as Nawa (1937) and Kojima (1958) point out.
Second, there was a long-run cost reduction as indicated by the declining
prices of domestic products compared to foreign ones, and by product qual­
ity improvement and diversification. Because production cost is composed
of the costs of raw materials, capital, and labor, Japanese producers were
at first handicapped by the high cost of imported raw cotton and machinery.
But these costs were offset by the productivity increase realized through im­
proved technology, enabling the mixing of raw cotton imported from various
parts of the world and the institution of a 24-hour, two-shift operating sched­
ule. Labor productivity increased faster than wages increased, which implies
a reduction in labor costs. Fujino, Fujino, and Ono (1979:10) estimate that
labor productivity in cotton textiles increased by an annual average rate of
2.56 percent for the period 1894-97; within cotton textiles the increase was
much higher for cotton fabrics than for spinning, reflecting a shift in Japan's
comparative advantage from cotton yarn to fabrics. Increased wage costs,
however, exceeded productivity increases after the Second World War, which
promoted the shift to the mature and reverse import stages.
Third, a group of large private firms played a leading role, while the govern­
ment's role was quite limited. Large private firms led the development of
Japan's cotton textile industry without much support from the government.
Table 4.2
Changes in the Export Market Structure of Japanese Cotton Fabrics (Vo)
Taiwan
Period and
Korea China 3 Southeast
Asia b Rest of
world Total
1890-1892 14.3 68.9 7.5 9.3 100.0
1900-1902 60.5 37.9 0.8 0.8 100.0
1911-13 33.5 58.4 3.9 4.2 100.0
1924-26 12.6 46.7 26.3 14.4 100.0
1934-36 9.8 15.4 30.8 44.0 100.0
Notes:
a. China includes Hong Kong, Manchuria, and Kwantung provinces.
b. Southeast Asia includes India, the Philippines, and Thailand.
Source: Ministry of Finance (Okurasho), Dainihon Gaikoku Boeki Nempyo (Foreign Trade
Annual of the Empire of Japan), various years.

Catching-Up Product Cycle Development 11
Import substitution was achieved under tariffs of 5 percent or less before the
recovery of tariff autonomy. Japanese producers were even handicapped by
an import duty on raw cotton and an export duty on cotton yarn, though
both tariffs were eliminated in 1895. Low conventional tariffs of 7 percent
on cotton yarn and 10 percent on cotton fabrics continued to be applied to
imports from Great Britain even after the recovery of tariff autonomy in 1899. 2
It was only after the Second World War that Japanese cotton textiles received
adjustment assistance from the government (see discussion in Chapter 9).
Before the Second World War, the Greater Japanese Association of Spin­
ning Firms (Dainihon Boseki Rengokai) took bold steps to protect the indus­
try's development (Nihon Boseki Kyokai 1982). A typical example was the
Indian Cotton Transportation Arrangement in 1891. The association boycot­
ted the English shipping firm Peninsular and Oriental Steam Navigation Com­
pany, which had monopolized the market, because of that firm's increase in
shipping costs for raw cotton brought to Japan from India. It supported in­
stead the initiation of a Bombay-Yokohama route by the Japanese shipping
company Nihon Yusen Kaisha, by assuring that a given minimum amount
of raw cotton would be loaded. In this way the association succeeded in stop­
ping the shipping fare increase (Nihon Boseki Kyokai 1982: Pt. 1, chap. 4).
The spinning firm association was also active in arranging production cur­
tailment and export promotion. During the 1890 recession, spinning firms
suffered from surplus capacity for the first time, which resulted in a serious
management crisis. The association's members all stopped spinning opera­
tions for eight full days in June and July in an attempt to maintain market
prices of yarn. This production curtailment was repeated ten times before
1937 to secure an orderly expansion of total production. Another initiative
of the spinning firm association was the attempt to increase exports by sell­
ing products abroad at cheaper prices. The group was also able to have the
export duty on cotton yarn and the import duty on raw cotton eliminated
(Nihon Boseki Kyokai 1982: Pt. 1, chap. 2).
The organization of cotton fabric production and its export structure merits
mentioning. There was an established division of labor in fabric production:
small and medium weaving firms produced narrow cotton fabrics for domestic
consumption, whereas large spinning firms produced broad cotton fabrics
for export using their own yarns. However, due to the wider usage of power
looms and the gradual decline in sales of traditional narrow cotton cloth,
the small and medium firms began to shift to production of broad cotton
fabrics. The production shift began around 1910 and accelerated during the
export booms of the First World War and the 1930s. By 1937, small and me­
dium weaving firms were supplying 45.3 percent of the broad fabrics produced
for export (Kajinishi 1964: Pt. 4, chaps. 1 and 2).
Unlike the integrated weaving and finishing production of large spinning

78 Industry Growth and International Trade
firms, the small and medium weavers organized cooperative groups in rural
cities. Within each group, each firm specialized in a nonintegrated process
of weaving, dyeing, and finishing. Through on-the-job experience, firms ac­
cumulated technical expertise and developed an efficient group of weavers
who could supply a variety of quality fabrics in small lots, a large proportion
of which were exported. Big trading companies, as well as their affiliated lo­
cal wholesalers, provided the weavers with yarns, financing, design, and mar­
keting information and sold their products in Japan and abroad.
Quickly foreseeing a change in the competitiveness of British and Ameri­
can products, Japanese general trading companies promoted the upgrading
and standardization of their cotton fabric, exports. Unfinished gray fabrics
were quickly replaced by finished dyed ones, and cotton weaving firms soon
met the higher standards for exports required by trading companies (Kajinishi
1964: Pt. 4, chap. 4). New markets in Southeast Asia, Oceania, Latin Amer­
ica, and Africa were explored, and branch networks were extended. A similar
pattern of export development was repeated for rayon fabrics in the 1930s
and for synthetic fabrics in the 1960s.
Another export-promotion activity of trading firms.was the formation of
export cartels. Mitsui Bussan encouraged large spinning firms to organize car­
tels for the expansion of exports to specific markets. In 1906 Sanei Mempu
Yushutsu Kumiai was formed for exports to Korea, and in 1913 the Nihon
Mempu Yushutsu Kumiai was formed for exports to Manchuria. At first, Mit­
sui Bussan provided the cartels with trading services at no charge, but later
set low commission rates, with export credit at preferential rates.
The achievements of the cartels were impressive in both the Korean and
Manchurian markets. In 1901, British cotton fabrics made up 70 percent of
the Korean market, and American cotton fabrics made up 99 percent of the
Manchurian market. Because Mitsui supplied Japanese fabrics at prices 20
to 25 percent cheaper than those of its British and American rivals, Japanese
fabrics were able to quickly replace the British and American fabrics in both
markets. By 1910, Japanese fabrics made up over 90 percent of both markets
(MITI 1982). This exemplifies the success of the concerted efforts of manufac­
turers and distributors in export expansion.
TEXTILE INDUSTRY DIVERSIFICATION
The CPC development of one industry tended to induce CPC development
in related industries, thereby causing the diversification in the production and
export structures of Japan. Kojima (1958: chap. 10) called this phenomenon
the diversification pattern of the CPC and attributed it to a change in the
comparative advantage structure of the importing country as a result of cap­
ital accumulation. The following section emphasizes two aspects of this diver-

Catching-Up Product Cycle Development 79
sification development pattern in industries directly related to the textile in­
dustry. The diversification development pattern of several industries is dis­
cussed in detail in Chapter 5.
Diversification of Textile Raw Materials
Among the raw materials used in textiles—which include natural cotton,
chemical yarns (rayon and acetate), and synthetic yarns (nylon, polyester, acryl­
ic)—synthetic yarn is the raw material that is now produced and consumed
in the greatest quantity. The shares of raw materials used in textile produc­
tion (based on yarn tonnage for 1965) are as follows: cotton (36%), chemical
fibers (26.1%), and synthetic fibers (24.7%). Chemical fibers were first devel­
oped as a substitute for silk and became inexpensive as the use of wider fabrics
became popular. Synthetic fibers were also substituted for materials such as
silk, cotton, and wool because of their uniform quality, a characteristic not
found in natural fibers. Moreover, due to other chemical properties, synthetic
fibers were the raw material of choice in the manufacture of fabrics for house­
hold and industrial uses other than clothing. Cotton, chemical, and synthetic
fibers do not always blend together perfectly, but the technologies used in
the weaving through the sewing stages for all three fibers are similar. Because
of this similarity, all three fibers have come to be considered substitute raw
materials by midstream (e.g., spinning firms) and downstream (e.g., garment
manufacturers) firms. This substitutability has led to a drastic change in the
industry's production structure.
Figure 4.3 traces the movements of the import/demand ratio (M/D) and
the export/production ratio (X/S) for each of the three textile raw materials,
following the schema depicted in Figure 2.1. The interrelationships of the ra­
tios depicted in the figure reveal three important findings: (1) the time lag
in CPC development for chemical and synthetic fibers reflects the time lag
in their technological development; (2) for both cotton and chemical Fibers,
import substitution (decline in M/D) preceded export expansion (increase in
X/S), although in the case of synthetic fibers export expansion commenced
without import substitution; (3) in the post-Second World War period the
export/production ratio of both cotton and synthetic fibers declined as their
import/demand ratios rose. However, net imports of cotton began, whereas
for chemical and synthetic fibers large net exports were still evident. The three
findings, which provide additional insight into Japanese CPC development,
are discussed below in their historical context.
Finding one. The time lag in chemical and synthetic fiber development
reflects the time lag in production technology development in the United States
and Europe. However, the time interval between technology development and
the start of domestic production was reduced in the case of Japan for the

80 Industry Growth and International Trade
Figure 4.3
Diversification in Textile Industry Development

60
40
20
0
60
40
20 -I 1 1 1 1 1 1 1—71 1 1 1 1 1 1 1 1 r
Chemical textiles
M/D
J i L • • • . V- i i i i i i i ir"?^
t > 1 1 1—
Synthetic textiles i—i 1 1 1—r
j i 1 i
1890 1900 1910 1920 1930 1940 1950 1960 1970 1980
Year
Notes: M/D = import/demand ratio.
X/S = export/production ratio.
Figures for cotton are smoothed by a seven-year moving average before 1940 and a five-
year moving average after 1950.
Source: Estimated by the author. See Yamazawa (1984) for details of the estimation.
following reasons. In 1892, the production technique for viscose rayon was
developed in Europe. Although Azuma Kogyo (which became Teijin) suc­
ceeded in developing its own technology in 1913, commercial production did
not begin until continuous spinning machinery was imported in 1923. After­
ward, throughout the rest of the 1920s, Asahi Kenshoku, Toyo Rayon, and
the big cotton-spinning Firms began production with imported technology
and machinery.
The manufacture of synthetic fibers began in 1938 with the acquisition of
basic patents for nylon 66 from Dupont. In 1950, Japan developed on its own
the technology for nylon 66, and in 1951 the technology for vinylon, and in
general made considerable progress in closing the technology gap. Domestic

Catching-Up Product Cycle Development 81
production of polyester and acrylic began in the second half of the 1950s.
Although major chemical fiber firms were already actively engaged in au­
tonomous technology development for the preparatory stages, they had to
purchase patents from American and European firms that owned the basic
patents for production. 1
Finding two. Although the raw materials for chemical and synthetic fibers
were new commodities, it was possible to adapt the technology for cotton
and silk weaving (after the yarn-producing stage) to accommodate the shift
to the new raw materials relatively easily. Even for chemical fibers, only yarn
had to be imported, and thanks to the use of the accumulated technology
in silk weaving, the woven fabrics produced were competitive from the begin­
ning. Existing trade channels were used for marketing, and the synthetic tex­
tile industry also made use of existing weaving technology and marketing
channels (except for yarn). Practically no imports were needed and export
expansion was possible. The active shift from silk and cotton to chemical and
synthetic fibers took place in every weaving region.
Competition with European imports was severe during the import substi­
tution stage for chemical fibers. From 1919 to 1926, the world price of rayon
fell by about 33 percent, and between 1929 and 1931 it dropped further by
50 percent. As a result, the pioneering rayon firms urged the government to
raise tariffs by 25 percent. But the increase in tariffs merely induced the en­
try of other firms, ultimately leading to more severe competition in the domes­
tic market. Although labor productivity rose by 23 percent between 1926 and
1930, most of this was a result of cost reduction from scale expansion (Ni­
hon Kagaku Sen-i Kyokai 1974).
To provide an outlet for surplus production, the spinning firm association
resorted to a compulsory export quota of 5 percent in 1929. This generally
took the form of rayon fabric exports, which increased rapidly, with Korea,
India, Indonesia, Australia, and China as the principal markets. Except in
Korea, Japan was able to wrench away a share of the market from Italy, the
United Kingdom, France, and the Netherlands by taking advantage of im­
proved production technology, the low prices resulting from economies of
scale, and the low wages reinforced by the exchange rate depreciation of 1932.
Japan's rayon fabric exports encountered import restrictions in these mar­
kets after 1931.
Finding three. After the Second World War there was a marked increase
in the production and consumption of chemical and synthetic fabrics as op­
posed to natural fiber fabrics. The share of chemical and synthetic fibers in
total fabric raw materials on a yarn basis rose from 25.4 percent in 1952 to
50.8 percent in 1965. By 1982, the share was 68.7 percent. This was largely
due to the rapid rise in usage of synthetic fibers, which were developed after
the war. The share of synthetic fibers (excluding chemicals) in total fabric

82 Industry Growth and International Trade
raw materials was 0.5 percent, 24.7 percent, and 57.9 percent for 1952, 1965,
and 1982, respectively. Such rapid growth in production was due to the cor­
responding demand growth that occurred as a result of the wide acceptance
of the new synthetic fibers, as well as the price decline that resulted from
expansion in the scale of production.
The competitive edge resulted from the shift of traditional cotton- and silk-
producing regions to the use of the new synthetic materials. Given such ac­
quired price competitiveness, 45 percent of synthetic textiles were exported
in the form of fabrics. About 38 to 39 percent of exports were in the form
of yarn, whereas a much smaller percentage, 16 to 17 percent, was accounted
for by finished forms of textiles. Although developing countries had caught
up with Japan in the production of chemical and synthetic fibers (to the ex­
tent that the import/demand ratio had risen in Japan), the increase in im­
ports was concentrated in finished products. Today, by contrast, synthetic
textiles as a whole still exhibit large export surpluses. Recently, a few firms
in the synthetic fabric industry have begun to use a new type of yarn, and
the resulting difference in fabrics has called for a special processing technol­
ogy. These new fabrics were exported mainly to the American and Middle
Eastern markets from 1979 to 1984 in response to the sophisticated demand
at higher income levels. This export boom contributed in part to the revital-
ization of the Japanese textile industry.
Linkage Effects of Cotton Industry Development
CPC development in the textile industry has induced similar growth pat­
terns in its two supplying industries (known as the backward linkage effect):
(1) the textile machinery industry (spinning preparation, spinning, weaving,
dyeing, and knitting), and (2) the dyestuffs industry (alizarin, aniline, and
other dyes). Expansion in the cotton yarn and cotton fabric industries called
for increased investments in textile machinery and synthetic dyestuffs; however,
since there was no domestic production of these goods at the time, they be­
gan to be imported in 1883. With continued growth in domestic demand,
domestic production was attempted, but unlike in the textile industry where
indigenous technology made absorption of new technologies easier, technol­
ogy absorption in the machinery and dyestuffs industries was more difficult
and came about only after a considerable time lag. Domestic production of
textile machinery came before dyestuff production because of the accumu­
lated production technology know-how derived from traditional weaving and
spinning machinery. By 1889, there were 100 machines that had been produced
domestically compared to 3,386 units that had been imported. By 1897, Sakichi
Toyoda had already built the Toyoda weaving loom (made of wood), but due
to the large gap between modern and traditional spinning technologies, domes-

Catching-Up Product Cycle Development 83
tic production of modern spinning machinery did not come about until 1925.
Although import substitution occurred continuously, imports reflected the
equipment-investment cycle, which peaked in 1920 and rapidly declined there­
after. By 1925, the level of imports was less than domestic production and
by 1933 even less than exports.
In contrast, production of synthetic dyestuffs was an entirely new chemi­
cal industry. Before the First World War all dyestuffs had to be imported,
but in 1912, the government-run industrial laboratory succeeded in synthesizing
alizarin. By 1914, commercial production of dyestuffs began in a factory at­
tached to the privately owned mine Mitsui Kozan. With the onset of the First
World War, import volume during the period 1915-17 fell to less than one-
seventh of the volume during the period 1911-12. In 1915, however, a bill was
introduced and enacted to promote the production of dyestuffs and medical
goods (Senryo-Iyakuhin-Seizo-Shorei-ho). As a result, a dyestuff manufac­
turing firm (Nippon Senryo Kaisha) was set up and by 1918 domestic produc­
tion had caught up with import volume. Although imports recovered during
the period 1922-24 due to dumping, they later declined and fell below ex­
ports in 1930. Steady export expansion of both textile machinery and dyestuffs
to China and Korea began only in the 1930s.
It is difficult to empirically measure the magnitude of the backward link­
age effects. The relationship between textile machinery and synthetic dyestuffs
(domestic demand in volume terms) and cotton production is estimated us­
ing ordinary least squares method techniques as follows:
Textile machinery (1883-1936)
In D 2 = 1.5575 + 0.8294 In 5,
(0.08) (7.33)
R 2 = 0.8262, DW = 1.0212
Synthetic dyestuffs (1883-1937)
In Dj = -96.2934 + 0.4810 In 5, + 0.0515/
(6.31) (5.72) (6.09)
R 2 = 0.9233, DW = 0.5826
where D 2 and D s are domestic demand for textile machinery and synthetic
dyestuffs, respectively; S, refers to production volume of cotton textiles; t is
the time-trend variable; and the figures in parentheses are the /-values. In both
equations, statistically significant, positive coefficients for cotton production
endorse the backward linkage effects on supplying industries over the whole
period of the latter group's CPC development. However, such factors as tech­
nology gaps and the occurrence of wars exert their influences on domestic
demand expansion (which is the basis for the CPC development), so that it
is not possible to specify exactly the scale of domestic demand needed in order
for domestic production to begin. – 0.0011/
(0.10)

84 Industry Growth and International TYade
FOREIGN INVESTMENT AND TRADE CHANGES
As discussed in Chapter 2, foreign investment plays an important role in
the last stage of CPC development. It hastens the "catching-up" process of
host (late-developing) countries and the decrease in production growth of in­
vesting (developed) countries, leading to the latter's reverse import stage. In
fact, direct foreign investment (DFI) was undertaken by the Japanese textile
industry during the period of export expansion and the subsequent slowdown
in production. In turn, this contributed to the textile industry's development
in East Asia and the ASEAN member countries, while encouraging the growth
of reverse imports in Japan.
Direct foreign investment by the Japanese textile industry had been under­
taken by spinning Firms during the First World War in the Shanghai, Qing-
dao, and Tian Jing districts of the Chinese mainland. After the Second World
War, Toyo Boseki set up a spinning mill in Brazil. Table 4.3 shows the amount
of DFI by the Japanese textile industry after the Second World War and plots
its structure. Between 1955 and 1976 there were a total of 377 DFI projects.
The 317 DFI projects during the two periods 1965-69 and 1970-74 represent
a concentration of some 84 percent of total DFI. Before 1964, Japan suffered
from a scarcity of funds to invest overseas, and since 1975 DFI activity has
virtually come to a standstill. The large amount of overseas investment in
the period 1965-74 is characteristic of the Japanese textile industry when com­
pared with the relatively small amount of DFI by American and European
textile firms (OECD 1981:64). Since DFI implies the transfer of a package
of managerial skills, capital, and technology, the active DFI undertaken by
the Japanese textile firms reflects the industry's abundance of human resources
possessing such skills and technology.
Table 4.3 also shows that DFI by the Japanese textile industry in developed
countries was rare; about 92 percent of the industry's total DFI was in de­
veloping countries. Of this, East Asia and the ASEAN member countries had
shares of 40 and 28 percent of Japan's total DFI, respectively, and became
the principal host countries for Japanese investment.
Of the 258 DFI projects in East Asia and the ASEAN member countries,
238 are almost equally divided between the midstream and downstream stages
of production. Investments have been concentrated in the downstream stage
(clothing and other finished products) in East Asia, whereas in ASEAN, in­
vestments in the midstream stage (spinning, weaving, dyeing, finishing) are
almost double those in the downstream stage. The upstream stage, which sup­
plies raw materials to the midstream and downstream stages of production,
consists wholly of chemical and synthetic fiber making. But in both East Asia
and ASEAN, however, investments in this stage are small.
Much of the DFI undertaken by the Japanese textile industry has been in
chemical and synthetic textiles. In contrast to the cotton textile industry, which

Catching-Up Product Cycle Development 85
had generally been established by local firms, the chemical and synthetic tex­
tile industry, being a new industry, was often encouraged by host countries
through promotion policies, including high import tariffs and exemption from
corporate and business taxes. A number of investments took the form of joint
ventures with local entrepreneurs, and the share of wholly owned Japanese
firms ranged from 62 to 100 percent in the upstream stage in each country
except Taiwan. In contrast, at the midstream stage, majority-owned Japanese
firms constituted a smaller share; for Malaysia's spinning industry the share
was 45 to 57 percent, for Thailand's spinning industry the share was 20 per­
cent, and for Indonesia's spinning industry it was 15 percent (all in 1978
values). At the downstream stage, Japanese firms represented an even smaller
share (Tran 1985).
The motives for DFI by Japan's textile industry include: (1) DFI has al­
lowed the industry to preserve its market in the host country by circumvent-
Table 4.3
Direct Foreign Investment by the Japanese Textile Industry (number of
projects)
Other
Production East developing Developed
Period stage Asia ASEAN countries countries Total
1955-64 U 0 1
M 4 4 20 1 38
D 2 6
1965-69 U 5 2
M 17 18 24 4 106
D 31 5
1970-74 U 3 9
M 29 38 29 23 211
D 51 19
1975-78 U 0 0
M 2 5 5 3 22
D 7 0
Total upstream 8 12
Total midstream 52 65
Total downstream 91 30
Total 151 107 88 31 377
Note: U, M, and D denote the upstream, midstream, and downstream stages of synthetic fiber
textile production.
Source: Tran (1985).

86 Industry Growth and International Trade
ing the import barriers set up to encourage domestic production of the im­
port substitute, and (2) DFI has allowed the industry to avoid the rising trend
in wages in Japan by transferring production to low-wage developing coun­
tries. While Japan's DFI has helped Japanese industry, it has also promoted
the growth of the East Asian and ASEAN textile industries. This has result­
ed in a drastic change in the direction of the textile goods trade in the region.
In the early 1950s, Japan was virtually the only country with a modern tex­
tile industry in the East and Southeast Asian regions and was competitive
in exports. Cotton textile production began in the early 1950s in Taiwan and
Hong Kong and in the mid-1950s in Korea. However, until the 1960s only
Hong Kong was competitive in export markets. Since the ASEAN member
countries had just begun cotton production, Southeast Asia was Japan's major
export market for cotton products in the 1950s. In the 1960s, however, im­
port substitution in cotton products grew rapidly in the ASEAN member coun­
tries. During this period, East Asian exports of cotton products to the United
States swelled and displaced Japanese exports. It was thus in the U.S. market
that trade competition between Japanese and late-developing East Asian coun­
tries began.
Chemical and synthetic yarn production began in the second half of the
1960s in both the East Asian and the ASEAN member countries. By the early
1970s, Taiwan and Korea had rapidly achieved import substitution and ex­
ports were growing. The ASEAN member countries reached the export stage
in the second half of the 1970s. As discussed earlier, Japan's DFI played a
leading role in the industrialization of the chemical and synthetic textile in­
dustries in these countries. Rapid localization of production was achieved in
the East Asian countries that had already reached the export stage in cotton
textile manufacturing, and in the ASEAN member countries in which im­
port substitution in cotton and synthetic textiles had taken place. Japanese
firms made significant contributions even in the development of the midstream
stages of these countries' textile industries. Today, even in ASEAN's slowest-
developing country (Indonesia), cotton and synthetic textile exports have be­
gun to expand. As a whole, the East and Southeast Asian region has had
great capacity to supply textile exports. Indeed, in the 1970s, East Asian ex­
port goods began leading the penetration of the Japanese market. This
"boomerang effect," as Shinohara (1982: chap. 9) calls it, is the major cause
of the Japanese textile industry's entering the reverse import stage. Chapter
9 discusses in greater detail the corresponding structural adjustment in Japan's
domestic market.

Five
CPC Development
of the Iron and Steel Industry
The development of the iron and steel industry followed a pattern similar
to that advanced in the previous chapter. Here, however, a simulation analy­
sis is conducted using a simple econometric model based on the hypothe­
sized development pattern. The summary that follows outlines the CPC
development for the chemicals and other heavy industries and attempts to
identify the mechanisms behind the diversification of Japan's industrial
structure.
DEVELOPMENT OF THE IRON AND STEEL INDUSTRY
The iron and steel industry has three production processes: the iron-making
process in which pig iron is refined from iron ore in a blast furnace fired by
coking coal; the steel-making process through which pig iron is changed into
steel ingots in open-hearth furnaces or converters; and finally, the milling
and casting processes in which ingots are fabricated into various types of steel
products such as bars, rods, plates, pipes, and rails. The output of ingots in
the steel-making process is often used to measure the growth of the industry
in international comparisons. But the output of ingots is not a good measure
of the impact of the Japanese steel industry in the international market be­
cause Japan's ingot trade has been almost negligible relative to its ingot
production. Industry trade has been in pig iron and steel products, and domes­
tic ingot output by and large goes into the domestic production of steel
products.
This situation can be partly accounted for by technical explanations, but
it also reflects unbalanced growth between the iron-making process on the
one hand, and the steel-making and milling processes on the other. This im-

88 Industry Growth and International Trade
balance is partly explained by the historical development of the industry. Be­
fore the Second World War, the integrated operation of the three processes
occurred only in the government-owned Yawata Iron Works and one private
firm; the remaining firms in the industry, including major zaibatsu firms,
were steel-making and milling firms that were dependent on imported pig
iron.' Because production of iron was handicapped by high-cost imports of
iron ore, and because large investments of capital were necessary to reach
optimum-scale production, the private firms were forced to specialize in the
steel-making and milling processes. Therefore, in the following analysis, steel
products represent the industry as a whole. For comparison, however, frequent
reference will be made to the growth pattern of pig iron production.
CPC Development of the Steel Industry
Figure 5.1 shows the growth of domestic production (5), imports (A/), ex­
ports (X) t and (apparent) domestic demand (D) for steel products over the
past one hundred years. Domestic demand for steel products is defined as
D = S + M – X and represents the demand by the Japanese for steel
products, both domestic and imported. Seven-year moving average values are
plotted to show the relationships between the growth trends of the four vari­
ables. Figure 5.1 exhibits the typical pattern of growth, with successful im­
port substitution followed by export expansion. The industry's steady growth
began in 1901 when the government-owned Yawata Iron Works commenced
operations. At that time, there already existed a fairly large domestic market
for imported steel products. Yawata was equipped with integrated facilities
for iron and steel making at the time of its establishment, but nine years passed
before the firm was able to operate at a profit. Both the deficits and capacity
expansion over this period were financed by the government. Until 1910, the
major zaibatsu firms followed Yawata's lead, but they were equipped with
only open-hearth furnaces and milling facilities and were dependent on either
Yawata or foreign suppliers for their pig iron inputs. Domestic steel produc­
tion expanded rapidly during this period.
The First World War spurred the industry to expand rapidly once again.
Price hikes due to the stoppage of imports from Europe and tax exemptions
under the Iron-Manufacturing Encouragement Act (1917) increased profits
and encouraged the entry of small and medium-scale firms into the industry.
Production grew rapidly and exceeded imports in 1923; imports then decreased
and by 1934, exports exceeded imports. Rapid growth in both production and
exports resumed following the recovery from the destruction of the Second
World War and even accelerated in the postwar period of heavy investment
and technology transfer. Exports of steel products in the early years of the
interwar period seem to have been mainly reexports, and before the Second

Figure 5.1
CPC Development of Steel Products

Source: Yamazawa (1984).

90 Industry Growth and International Trade
World War exports were mostly to Japan's colonies (Korea, Taiwan, Man­
churia, and the rest of China). Only after the Second World War were Japan's
steel products sold in the world market at competitive prices. Table 5.1 shows
the average annual growth rates of domestic demand, production, imports,
and exports for the periods 1900-1914 to 1952-66.
It is interesting to view the interwar period from the perspective of indus­
try growth. Having suffered both from overcapacity and severe competition
from cheap imports after the First World War, the industry undertook ra­
tionalization and cost reduction and achieved complete import substitution.
The industry also benefited from government support, which included tariff
hikes (1921, 1926, 1932), tax exemptions to steel-making firms, subsidization
of steel production for shipbuilding (1921), the formation of cartels, and fi­
nally the establishment of the Nippon Steel Company (1934) (see Table 5.2).
In addition, a 43 percent depreciation in the exchange rate in early 1932 was
effectively equivalent to a rise in tariffs on imports and a subsidy to exports.
Cartels were formed according to categories of steel products; in each
category, Yawata and the major private firms collaborated to maintain prices
of domestic products as low as those of imports. Price-reducing competition
could thus be avoided and the combined share of Japanese firms in the domes­
tic market increased, although some inefficient firms had to sell at a loss at
cartel prices, but the cartels themselves were eventually terminated. In 1934,
Yawata and six zaibatsu steel mills merged to form Nippon Steel Company,
which produced 95.2 percent of the pig iron and 43.9 percent of all steel
products in Japan. Elimination of duplicate capacity, rationalization through
economies of scale, and the strengthening of international competitiveness
were put forward as reasons for the merger, but the merger also helped some
inefficient private mills. Import dependence (M/D) declined rapidly from 60
percent to less than 10 percent in the interwar period.
Table 5.1
Average Annual Growth Rates of Steel Products (%)
1900-14 1914-23 1923-38 1952-66
Domestic demand 10.1 8.9 7.6 15.9
Production 26.3 11.0 12.9 16.5
Imports 6.5 7.6 -5.7 -0.3
Exports 23.7 13.4 15.1 18.9
Note: Rates are calculated from the seven-year moving averages for each of the four variables
in 1900, 1914, 1923, 1938, 1952, and 1966.
Source: Nihon Tekko Renmei, Tekko Tokei Yoran (Iron and Steel Statistics Annual) (1970).

CPC Development of the Iron and Steel Industry 91
Unbalanced Growth between Iron and Steel
The growth of pig iron production was much slower than that of steel. The
first success in the production of pig iron in a blast furnace came in 1887,
and almost half of Japan's needs were supplied domestically by 1900. But
Table 5.2
Government Policies Protecting the Iron and Steel Industry
Period Policy
Before 1900 Experimental operations in government-owned small-scale
foundries
1901 Yawata (government-owned) Iron Works commenced operations
1907-09 First expansion plan for Yawata
1911 General Revision of Customs Tariff Act (1.67 yen per tonne of
pig iron and 15% ad valorem equivalent rate on steel products)
1911-16 Second expansion plan for Yawata
1917 Iron-Manufacturing Encouragement Act (tax exemption for iron-
making firms)
1921 First amendment to Iron-Manufacturing Encouragement Act (en­
larged tax exemption). Change in tariffs on steel products from
specific duties to ad valorem ones (15% ad valorem rate).
1926 Second amendment to Iron-Manufacturing Encouragement Act
(subsidy to pig iron production; 3 to 6 yen per tonne). Rise in
customs tariff on steel products (18% ad valorem rate).
1926-31 Formation of cartels by category of steel products with govern­
ment support
1932 Rise in customs tariff on pig iron (6 yen per tonne)
1934 Establishment of the Nippon Steel Company (the merger of
Yawata with six zaibatsu firms. Eighty-two percent of its shares
were held by the finance minister. It produced 96 percent of the
pig iron and 44 percent of the steel products in Japan).
1937 Iron-Manufacturing Enterprise Act (support for integrated iron-
and steel-making firms; a change to production expansion policy
for wartime economy)
1937-41 Five-year expansion plan for steel production
1947-50 Reconstruction of steel production under Priority Production
Project
1951-55 First rationalization project for steel
1956-60 Second rationalization project for steel

92 Industry Growth and International Trade
pig iron imports increased at a pace that almost equalled growth of domestic
production, and as a result, import dependence declined by only 10 percent
during the interwar period. During the period 1900-1936, domestic demand
for pig iron increased at an average annual rate of 9.7 percent, the same rate
as the demand for steel products, whereas imports and production of pig iron
increased by 8.6 and 10.7 percent, respectively. Exports of pig iron remained
negligible, even in recent years. Figure 5.2 illustrates the decreasing depen­
dence on imports of both pig iron and steel products, and clearly contrasts
the import-substitution process occurring in both. The dependence on steel
imports declined rapidly from 94 percent in 1900 to 9 percent in 1936, except
for the slack period during the First World War. The dependence on pig iron
imports, however, declined slowly from 55 percent to 30 percent during the
same period. The differing patterns of growth between the pig iron and steel
products industries can be partly explained by the fact that during the First
World War, zaibatsu firms began producing pig iron near deposits of iron
ore and coke in Manchuria and Korea, and the pig iron was imported to Japan
for further fabrication. Dependence on pig iron produced in Manchuria and
Korea amounted to 18-20 percent of domestic demand during the First World
War and in the 1930s. Slow import substitution of pig iron is also explained
by Japan's lack of iron ore and coke, the continued competition from cheap
Indian imports during the 1920s, and the existence in Japan of a large group
of steel-making Firms largely dependent on imported pig iron. These firms
resisted an increase in the tariff on pig iron until 1932.
Throughout the 1930s, national security arguments promoted self-
sufficiency in pig iron and further expansion of integrated iron and steel
production. The rise in pig iron tariffs (1932), the dominance of pig iron
production by Nippon Steel (1934), and the Iron-Manufacturing Enterprise
Act (1937) were all in accordance with the promotion of integrated iron and
steel production. A five-year expansion plan for steel production (1937-41)
was designed to increase japan's self-sufficiency in pig iron to 89 percent.
In addition, the substitution of Manchurian pig iron for Indian pig iron was
further encouraged. With the help of such government policies, 80.5 percent
self-sufficiency in pig iron was achieved in 1940; in addition, two-thirds of
all imported pig iron came from Manchuria. The substitution of integrated
iron and steel production for imports was thus accomplished immediately
before the Second World War.
AN ECONOMETRIC MODEL OF CPC DEVELOPMENT
Mechanism of CPC Development
The review of the steel industry's development in the previous section gives
a clue as to the mechanism behind the successful import substitution and

Figure 5.2
Dependence on Pig Iron and Sleel Product Imports
60 I 1 | 1 1 1 1 1 1 1 1 1 1 1 1 r

100 I 1 ' 1 ' 1 ' 1 1 ' 1 • r

1900 1910 1920 1930 1940 1950 1960 1970 1980
Year
Notes: M/D = import/demand, X/S = export/production, and M i = imports from Korea
and Manchuria, all calculated from the seven-year moving average series.
Source: Yamazawa (1984).

94 Industry Growth and International Trade
export expansion of an industry. The factors behind this mechanism may be
summarized as follows:
• Steady reduction in production costs as the industry's production expands
is required. Reduction in costs tends to lower the price of the domestic
product relative to import and world prices, thereby making possible im­
port substitution and export expansion.
• Growth of domestic demand is also required to induce investment and out­
put expansion in the industry. Foreign trade is also important as export
expansion occurs and as domestic demand increases through import sub­
stitution.
• Government protection accelerates the growth of demand, which leads to
output expansion, cost reduction, and finally import substitution and ex­
port expansion for the industry. Industry protection may include such poli­
cies as tariffs and subsidies, and more direct but less apparent instruments
such as tax exemptions and preferential purchasing policies. Furthermore,
in the case of the steel industry, the government-owned Yawata Iron Works
initiated and led the growth of the industry.
For Japan, both cost reduction and increased domestic demand were in­
dispensable factors behind the successful import substitution and export ex­
pansion strategies; government protection was also important for most of
Japan's heavy manufacturing industries. The following econometric model
traces both the long-run growth trend of the industry and its short-term fluc­
tuations. 1
Four behavioral equations and an equilibrium condition are needed to de­
termine the equilibrium values of five endogenous variables for a given set
of values of exogenous variables in each year (see Table 5.3). The level of
domestic demand for steel is a function of domestic income and the demon­
stration effect (equation 1, Table 5.3). Domestic income is represented by the
total output of industries using steel products as inputs. This is different than
the indicator used for income in the case of consumer goods, which is repre­
sented by per capita gross domestic expenditure or consumption expenditure,
because demand for steel, an intermediate good, will be determined by the
income of steel-using firms. The demonstration effect reflects the changing
pattern of demand over time for a new product in the domestic market. It
can be formulated as a monotonically increasing function of time beginning
from the introduction of the product to the domestic market. The function
can be represented by a growth curve that rises first at an increasing rate, then
at a decreasing rate, and finally approaches a fixed level. The time necessary
for the demonstration effect to reach this fixed level differs from one com­
modity to the next. It will be shorter, for example, for synthetic products,
as a domestic market has already been developed for their natural substitutes.

CPC Development of the Iron and Steel Industry 95
The level of imports is determined in equation 2 in Table 5.3 by the ratio
of the import price and the domestic price for a given level of domestic de­
mand. Steady reduction of the domestic price relative to the import price pro­
motes import substitution, and imports will decline in absolute volume when
the price effect more than offsets the income effect (that is, the increase in"
domestic demand).
When the domestic price of a product declines significantly relative to the
world price, export of the domestic product begins. The growth of exports
will depend on both the growth of foreign income and the decline in the
domestic price relative to the world price (equation 3, Table 5.3).
The price equation, equation 4 in Table 5.3, represents the relationship be­
tween the domestic price (cost) and the total production of an industry, and,
therefore, the basic relationship of the model. Equation 4 hypothesizes a nega­
tive correlation between the domestic price and the total production of the
industry. This correlation has been widely observed in the growth processes
Table 5.3
Model of CPC Development
Domestic demand D = D(Y, L)
> 0, D[ > 0 (1)
Imports M = M(D, P, PM)
M' D > 0, M p > 0, M' pM < 0 (2)
Exports X = X(Z, P, PZ)
X± > 0, X'p < 0, x pz > 0 (3)
Price P = P{S, PO, W)
ps < °' pk> > °- pw > 0 (4)
Equilibrium condition D = S + M – X (5)
Note:
Symbols
D: Domestic demand
S: Production
X: Exports
M: Imports
Y: Domestic income variable
Z: Foreign income variable
L: Demonstration effect (monotonically increasing function of time)
P: Price of steel products
PM: Import price of steel products (including tariff)
PZ: World price of steel products
PO: Import price of iron ore
W: Wage rate of metal workers

96 Industry Growth and International Trade
of principal industries in Japan and implies long-run decreasing costs in these
industries. As one example, except during the First World War and the late
1930s, there was a negative correlation between price and total production
of steel products.* Because this relationship has been observed in the develop­
ment of several industries, it was introduced into the model. A theoretical
explanation is, however, also needed and is given below.
Long-run Decreasing Costs in an Industry
Long-run decreasing costs in an industry seem to stem from the complex
interaction of various factors. Among these various factors, the following seem
most significant:
(1) The expansion of an industry's total production is generally accompa­
nied by an enlarged scale of production (as measured by productive capac­
ity). In the process of industry expansion, new firms enter the industry and
new plants are established by existing firms, thus increasing the total number
of plants. But more important is the fact that the newer the plant, the larger
is its optimum scale of production, and the larger the optimum scale, the
lower the unit cost of production.
It may be that new small or medium-scale plants are established during
the years of persistent excess demand and price increase. But as the excess
demand slackens and the price decreases to its previous level, the inefficient
plants are eliminated through competition. This is in fact what happened in
the steel industry during and after the First World War. If the market is com­
petitive enough, the unit production cost in the most efficient new plant tends
to dominate the market price in the long run, and inefficient old plants are
scrapped and replaced by more efficient and larger plants. Cost reduction
due to large-scale production seems to be one of the main factors for an in­
dustry's decreasing costs over the long run. A new large-scale plant makes
it possible to reduce production costs in various ways. Although there is little
room for saving on material costs, since raw materials are used in proportion
to the level of output, a large-scale plant reduces the costs of construction,
energy, and labor per unit of production. Production capacity of blast fur­
naces increased steadily from 160 tonnes per day in 1901 to 500 tonnes in
1930 and 1,000 tonnes in 1937, and to more than 3,000 tonnes in 1965. This
is also the case for the production capacity of open-hearth furnaces in steel
manufacturing. Larger-scale production also makes possible the automatic
and continuous operation of production processes, thereby leading to reduc­
tion of costs per unit of production. A good example is the "hot strip" mill
process in steel making, which has replaced the old-fashioned "pull-over"
mill process because of its profitability in larger-scale production.
(2) Long-run decreasing costs can be partly explained by improvements in
the production process—including improvements in production equipment

CPC Development of the Iron and Steel Industry 97
and in management systems—that are a result of the accumulation of produc­
tion experiences and improvement in skills of production workers. In order
to expand a newly introduced industry, foreign technology must be quickly
learned, and the cost reduction that occurs with the learning effect is impor­
tant in the early stages of industry growth. This learning effect may not neces­
sarily increase in proportion to the level of production but may rise in
proportion to the accumulated total output. For example, it took several years
for Japanese iron- and steel-making firms to adapt foreign technology to the
types of iron ore and coke that were available in Japan.
(3) Cost reduction is also accelerated by external economies, that is, by the
effect of one industry on related industries. Machinery production in Japan
benefited a great deal from the growth of the steel industry, which in turn
gained from successful domestic production of steel-making machinery in
the 1930s. The steel industry has also benefited from the development of the
automatic control system in the post-Second World War years.
The relative contributions of these three factors, which seem to have brought
about a steady cost reduction in the case of steel products, differ considera­
bly in different industries and at different stages of industry growth. In the
case of the steel industry, the learning effect was most important in the in­
dustry's early years, and economies of scale and external effects gained im­
portance in the 1920s and the post-Second World War years.
It is not easy to quantitatively analyze long-run decreasing costs because
of the unavailability of reliable data on capacities and input-output relation­
ships in steel making. It is possible, however, to introduce this observed rela­
tionship as the combined effect of various factors. Increases in costs of
materials and labor (equation 4) tend to offset the reduction in costs men­
tioned earlier.
The dynamics of the model can be demonstrated. Industry growth is first
generated on the demand side, primarily with the growth of domestic income
and assisted by the learning effect in the domestic market and the growth
of foreign income. Steady growth of domestic demand induces output ex­
pansion, which is accompanied by the reduction in unit cost and market price.
This, in turn, leads to import substitution and export promotion for a given
level of import prices, world price of the product, and input prices, thereby
adding to the growth of domestic demand. 4 Exogenous changes in these prices
can affect this process. Government protection of an industry can lead to
a change in prices. Tariffs on imports, for example, raise import prices and
accelerate import substitution of the product.
One theoretical question remains. What assurance is there that the indus­
try always moves along the long-run decreasing cost curve? Movement down
the curve depends on two conditions being met. The First condition is that
firms accurately forecast any increase in demand and then expand produc­
tion accordingly without a time lag. The second condition is that the produc-

98 Industry Growth and International Trade
tive capacity be divisible so that firms can produce any level of production
at minimum cost. The actual growth process of steel products is character­
ized by cyclical fluctuations along a steady growth trend; this observation ap­
pears to attest to the accuracy of the assumptions underlying the model.
However, except for the First World War years and the late 1930s, which were
unusual periods for Japan because of persistent excess demand for steel
products, it may be assumed that the industry moved along the curve with
only some random divergences. Government support for the industry through
government-owned firms and preferential purchasing policies helped the in­
dustry meet the First condition.
ECONOMETRIC ANALYSIS
OF PRE-SECOND WORLD WAR DEVELOPMENT
The hypothesis and model put forward in the previous section need to be
tested by empirical data. Thus far, the model has been applied to actual data
for the pre-Second World War period excluding the years of persistent excess
demand, namely 1900-1914 and 1921-36, and for the post-Second World War
period (1952-68). Simulations of the growth process under counter-factual
policy assumptions were done for each period. The results are described in
this and the following section.
Four behavioral equations were estimated in log-linear form by the two-
stage least squares method for the periods 1900-1914 and 1921-36. The results
are summarized in Table 5.4.
The growth of domestic demand is explained fairly well by the domestic
income variable and the demonstration effect. Several alternative indicators
for income were used, including manufacturing production and the sum of
the net products of mining, manufacturing, and construction. However, the
production index of machinery gives a better estimate than either of the al­
ternatives. The machinery and metal goods industries make up the bulk of
the demand for steel products, and the sum of the two would improve the
estimates, but no data are available for this alternative. The demonstration
effect is estimated by a truncated exponential function of time. Alternative
values are applied to the time coefficient, and the value 0.1 gives the best es­
timate.
• The import function seems to capture fluctuations completely. The elasticity
of domestic demand for imports is estimated to be .45, and the domestic and
import price elasticities are 1.4 and -1.4, respectively. The low value of the
elasticity of domestic demand for imports is worth noting. This means that
when domestic demand for steel products increases by 1 percent, demand for
imported steel products increases by less than .5 percent. This low elasticity
of demand can be explained in part by the fact that a large part of domestic

Table 5.4
CPC Development of Steel Products: 1900-14, 1921-36
Estimates*
Domestic demand
log D = 6.6967 + .8511 log Y + 1.0804 log L
(12.02) (1.72)
s = .1610 R2 = .9741 DW = 1.57
Imports
log M = 4.4806 + .4494 log D + 1.3779 log PP – 1.4317 log PPM
(3.37) (2.30) (5.87)
s = .2936 R J = .7152 DW = 1.80 (1)
(2)
Exports
log X
Price (3)
-1.0181 + 1.7015 log Z – 6.4112 log PP + 2.0700 log PPZ
(3.47) (6.29) (3.36)
.6904 R 2 = .8701 DW = 1.12
(4)
log PP – 1.1994 – .0910 log 5 + .1800 log PPO
(4-79) (1.23)
s = .1332 R 1 = .7011 DW = .96
Final Tests D M X S PP
Root Mean Square 3729.5 2317.75 576.41 J070.90 0:11
Von-Neuman Square 0.45 0.57 1.10 0.34 0.87
Nofes:
Symbols
D:
S:
X:
M:
Y:
Z:
L:
PP:
PPM:
PPZ:
PPO: \- of steel products (hundred tonnes) Domestic demand
Production
Export
Import
Production index of machinery
World production of steel ingots
Demonstration effect (= I – exp(
Price of steel products
Import price of steel production including tariff
World price of steel products
Import price of iron ore 0.1(/ – 1890)1); '•• year, / = 1 for 1900
J- deflated by GDE deflator
a. Figures in parentheses show /-values of estimates.

100 Industry Growth and International Trade
demand for steel products came from the government for military produc­
tion and railroad construction. Therefore, there existed a strong preferential
purchasing policy by the government for domestically produced steel products.
This explanation is supported when the elasticity of domestic demand for
imports of steel is compared with that for pig iron (1900-1936).
log M = -.09 + .9093 log D + .1800 log PM/P
(45.54) (1.11)
R J = .9836 DW = 1.68
Domestic demand elasticity for pig iron imports is twice as large as that for
steel products. As mentioned previously, a large group of steel-making firms
depended on imported pig iron, and the import demand for pig iron increased
almost in proportion to the growth of domestic demand.
The export function explains the rapid growth of steel exports before the
Second World War. World production of steel ingots, a proxy variable for
world consumption of steel products, grew annually at 2.9 percent for the
period 1900-1936, and it induced a 4.9 (= 2.9 x 1.7) percent growth rate
in Japan's exports. The rest of export growth is attributed to price effects,
that is, the decline in the domestic price and the rise in the world price. Of
course, it cannot be denied that other factors not included in the export equa­
tion, such as preferential purchasing in Japan's domain, contributed to some
part of this export growth.
The price equation gives a statistically significant negative relationship be­
tween price and production, but the estimated coefficient for material cost
is less significant. The wage variable has been eliminated since its coefficient
is statistically insignificant and has the wrong sign.
A partial test of the estimation (standard error of residual (s), coefficient
of determination (R 1), and Durbin-Watson ratio (DW)) shows a relatively good
fit of the model to actual data. But it may be worthwhile to see whether the
model generates the process of import substitution and export expansion over
the whole period when only exogenous variables and initial values of en­
dogenous variables are given. The results can be seen in the final tests in Ta­
ble 5.4. The final tests do not follow every fluctuation in the observed data,
but they do trace the rapid growth of domestic production, import substitu­
tion and export expansion, and the steady decline in the domestic price. 9
The effect of government policy on the process of industry growth and trade
has already been mentioned. But which policy had a greater effect on import
substitution of steel products in Japan: import tariffs or the preferential pur­
chasing policy? Simulation 1 traces the values of imports and production in
the case where import tariffs are not levied over the whole period. Simula­
tion 2 traces those values in the absence of a preferential purchasing policy.

CPC Development of the Iron and Steel Industry 101
In simulation 1, the import price is lowered by the absence of tariffs; thus
imports shift upward and domestic production downward over the whole pe­
riod. In simulation 2, imports increase in proportion to domestic demand,
and domestic production shifts downward by the corresponding amounts.
Thus the effect of a preferential purchasing policy is almost three times as
large as that of import tariffs. In either case, however, the value of imports
follows a pattern similar to that in the estimations; growing slowly at first,
then turning downward in the mid-1920s, which demonstrates import substi­
tution through price effects. Thus the production values in both cases ap­
proach those of the estimations because of the rapid decline of imports.
DEVELOPMENT OF THE POSTWAR
IRON AND STEEL INDUSTRY
By the end of the Second World War, production in the iron and steel in­
dustry had fallen to 1.5 million tonnes, or one-fourth of the production level
of 1939. The industry's postwar development led initially to a recovery to
the prewar production level, followed by industry reorganization and increased
international competitiveness through technological innovation. This post­
war period was characterized by the expansion of domestic production based
on domestic- and export-oriented growth. Since the industry's integrated
production base had already been established, imports of both pig iron and
steel products during periods of domestic supply imbalances played only a
supplementary role. The following section reviews a number of phases in the
industry's pattern of production expansion.
Overview of Postwar Development 6
The first phase of postwar development consisted of the recovery of the
iron and steel industries under the Priority Production Policy (Keisha Seisan
Hoshiki) during the period 1947-50. Three major industries—coal, iron and
steel, and electric power—were given priority access to scarce raw materials
through the government's fiscal, financial, and foreign exchange control poli­
cies. The government's intent was for the development of these key indus­
tries to spur the overall development of the Japanese economy.
In order to reactivate the iron and steel industry's production equipment,
investment funds were made available through the Reconstruction Finance
Corporation (Fukko Kinyu Koko). The industry was also given priority ac­
cess to the supply of domestically produced coal, as well as to scarce foreign
exchange for imports of iron ore, coal, pig iron, and other raw materials and
equipment. Moreover, in order to be able to provide low-priced steel products
to other industries, the difference between the production cost and the product

102 Industry Growth and International Trade
price was subsidized by the government. As a result, by 1953, pig iron and
steel production had reached their prewar levels, and the industry was. able
to supply other industries with steel products at low prices.
The heavy government subsidization of these industries contributed to in­
flation. Moreover, the concentration on the reactivation of machinery and
equipment whose use dated to before the Second World War did not neces­
sarily lead to a sound remodeling of the industry.
The second phase of postwar development consisted of the adoption of
an industry rationalization policy. Between 1951 and 1955 under the first re­
organization program (Gorika Keikaku), the price control and subsidization
systems were dismantled. 7 Instead, low-interest financing as well as acceler­
ated depreciation policies were adopted to encourage the modernization of
equipment, particularly in the rolling-mill sector. To this end, the foreign ex­
change rationing mechanism was used to give the industry priority in import­
ing American and European technology and equipment that were also
exempted from import tariffs. This was followed by the second rationaliza­
tion program (1956-60), which concentrated on the construction of the most
modern integrated mills and the modernization of cargo-handling and trans­
port equipment for the industry's raw materials and products. During this
period (1951-60), more than half of the needed investment funds were financed
internally, although loans from the World Bank were also used.
In the process of industry rationalization, there was an increasing con­
centration of production within the integrated firms. The affiliation of the
semi-integrated and nonintegrated firms with the large integrated firms was
strengthened, and the present industrial structure, which is dominated by six
large-scale firms, was firmly established. The modernization of production
equipment reduced raw material utilization per unit of production, leading
to cost reductions and the strengthening of international competitiveness. The
volume of steel production doubled during the first reorganization period
and grew 240 percent during the second period. During this period, the rela­
tionship between production expansion and price decline is clear.
From 1951 to 1960, domestic demand (including indirect export demand
in the form of machinery and ships) exceeded prewar levels by 16.3 percent.
This high level of domestic demand led to an expansion of domestic produc­
tion, since pig iron and steel product imports were controlled by the govern­
ment's foreign exchange allocation policy. As the prices of domestic steel
products were higher than those of imports from the United States and Eu­
rope in the early years of the rationalization program, the import control sys­
tem effectively restricted steel imports. Except for the 9 percent increase in
imports during the severe supply scarcity of 1957, imports represented only
1 to 2 percent of the domestic demand for steel products. In 1960, pig iron
imports were fully liberalized, and in 1961 steel product imports were liberal-

CPC Development of the Iron and Steel Industry 103
ized. During the period 1956-69, Japan registered a 6 to 12 percent depen­
dence on imports of iron and steel (Tecson 1985).
Boosted by the growth of world demand and the special demand condi­
tions of the Korean War, exports began to expand in the early 1950s mainly
to the Southeast Asian countries. In order to maintain the rate of capacity
utilization during periods of domestic slowdown, each firm endeavored to
sharpen its export competitiveness. With export expansion further cost reduc­
tions were achieved, leading to intensified export competition among domes­
tic firms. Firms were thus driven to diversify their export markets, and the
American market became Japan's largest market for steel exports. From an
average export volume of 840,000 tonnes from 1950 to 1952, exports rapidly
grew to 2.2 million tonnes from 1960 to 1962, and to 18.9 million tonnes from
1970 to 1972. The process of export expansion was actively supported by the
government through various export promotion policies, such as exemption
from the export income tax (1953-64) and special accelerated depreciation
of the fixed assets of overseas branches (1953-58). The contribution of Japan's
general trading companies to foreign market development and export expan­
sion must also be acknowledged (cf. Chapter 6, pp. 132-136).
Econometric Analysis of Postwar Development
The equations for domestic demand, exports, and prices in Table 5.5 are
estimated using the same method as the estimates in Table 5.3 and cover the
years 1952-68, a period of rapid export growth. Import equation 4 was omit­
ted in this case since imports were negligible during the period. Estimates
of the domestic demand function closely resemble those of the prewar pe­
riod. The learning effect can be excluded from the estimates because the value
of log L is close to zero. The income elasticities and constant terms hardly
differ from those of the prewar estimates. This implies that the domestic de­
mand function during the two time periods remained stable.
Different estimates were obtained for the export function between the prewar
and postwar periods. Income elasticity increased and price elasticity decreased
in the latter period. World consumption of steel products grew faster in the
postwar period, and Japan's exports grew more than twice as fast, without
even considering the favorable price effect. The markets for Japan's steel
products changed, however, a great deal after the war. Southeast Asia and
North America became Japan's principal markets, replacing East Asia. Ex­
ports were concentrated in steel products such as plates and sheets, for which
world demand was rapidly increasing, while the supply capacities of Euro­
pean and American competitors were limited. These factors seem to explain
the high income elasticity previously mentioned.
Japan's steel industry also gained a share in the world market in this pe-

104 Industry Growth and International Trade
riod as a result of a decline in cost (price) relative to its European and Ameri­
can competitors. This is tested in the simulation based on the model of Table
5.5.
The price equation is less successful because of volatile price fluctuations
in the mid-1950s, but price elasticity with respect to production is four times
higher than before the war. This reflects the heavy investment since the late
1950s in large-scale plants equipped with newly developed technology. Addi­
tionally, the coefficient of the price of materials turned out to be insignifi­
cant and of the wrong sign, whereas the wage level coefficient has the correct
sign but is also statistically insignificant.
Table 5.5
CPC Development of Steel Products: 1952-68
Estimates*
Domestic Demand (1)
log D = 6.2531 + .8137 log Y
(40.19)
s = .0755 R 1 = .9902 DW = 1.97
Exports (2)
log X = .0289 + 2.2711 log X – 1.4337 log PP/PPZ
(3.68) (1.64)
s = .4070 R 2 = .8376 DW = .69
Price (3)
log PP = 5.9651 – .4043 log S + .3437 log PW
(1.43) (.38)
s = .1731 R 1 = .6298 DW = 1.86
Final Tests D X S PP
Root Mean Square 19051.01 16518.72 18044.54 0.22
Von-Neuman Square 1.72 0.48 1.00 1.25
Note:
Symbols
D: Domestic demand
S: Production |- of steel products (hundred tonnes)
X: Export
Y: Production index of machinery
PP: Price of steel products —]
PPZ: World price of steel products |- deflated by GDE deflator
PW: Wage rate of metal workers J
a. Figures in parentheses show r-values of estimates.

CPC Development of the Iron and Steel Industry 105
Final tests show that, apart from fluctuations, the model succeeded in trac­
ing the growth trends for the major variables of steel products during the
postwar period. A simulation was attempted for the mechanism of export
growth in the postwar development of the steel industry. As mentioned previ­
ously, rapid export growth is one of the main features of the industry's post­
war development and is partly explained by high income elasticity and partly
by relative price decline. Simulations are computed on the assumption that
the world price fell to a level comparable with that of Japan so that the rela­
tive price remained constant. Simulations trace export and production values
in the absence of relative price declines. Although the growth of exports slows
to 11.0 percent, the strong growth trend of production continues.
The growth of Japan's iron and steel industry continued after 1968, reach­
ing a production level of 100 million tonnes in 1973-74, after which the rapid
growth in domestic demand and exports decelerated. However, even before
this took place, toward the end of the 1960s, voluntary export restraints had
begun to be imposed on Japanese steel exports to the United States, so that
Japanese exporters were overtaken by new steel-producing countries such as
South Korea and Brazil. This period's export mechanism thus differs from
that of the high-growth period and is discussed in greater detail in Chapter 9.
DIVERSIFICATION OF THE INDUSTRIAL STRUCTURE
In the foregoing sections, CPC development in the textile and steel indus­
tries was analyzed. Japan's other manufacturing industries also followed a
similar pattern of development though with varying time lags, leading to a
diversification of the entire industrial structure as well as an upgrading of
the industries as they became more capital- and technology-intensive.
Sequence of CPC Development
In order to compare the development of several industries, graphs employ­
ing the S/D curve of Figure 2.1 are used. Since exports and imports of
homogeneous products can be netted, it is possible to show each industry's
CPC development by the movements in the curve denoting the ratio of domes­
tic production to domestic demand (also referred to as the self-sufficiency
ratio when production falls short of demand). The process begins in the year
domestic production begins. Import substitution takes place as domestic
production overtakes imports and S/D equals 0.5. The export stage begins
in the year S/D equals 1.0. In the stages of export growth and export deceler­
ation, S/D is greater than 1.0, and the stage of reverse imports sets in when
S/D becomes less than 1.0.
Figure 5.3 shows the S/D curve of four industries—textiles, steel products,

106 Industry Growth and International Trade
Figure 5.3
CPC Development of Major Industries

1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980
Year
Note: Dashed lines in cotton textiles and automobiles indicate figures are not available.
Source: Yamazawa (1984).
automobiles, and industrial machinery. The development process of textile
products is representative of labor-intensive light industries, whereas the de­
velopment pattern of steel products is similar to that of artificial dyestuffs,
chemical fertilizers, and textile industry machinery. Automobiles and indus­
trial machinery are representative of postwar export-oriented industries. By
looking at Figure 5.3 from left to right, it is possible to see which industries
have reached the export stage and which industries are in the import substi­
tution stage at any point in time. The industrial and export structures be­
come diversified. Moreover, as some industries reach the export stage, a highly
self-sufficient industrial structure is created when reverse imports do not take
place.
At this point, some explanation for the growth processes in the automo­
bile and the industrial machinery industries is needed. Although domestic
production of passenger cars was begun in 1914 by Kaishin-sha, S/D began
to rise in 1924-25 when Ford and General Motors started production in wholly
owned subsidiary firms. In the 1920s, the government's subsidization policy
encouraged production in domestic firms, while a rise in tariff rates on im­
ports of parts for knock-down production promoted import substitution. Un­
assembled imports were banned in 1939, and imports of finished cars ended
in 1941. Domestically owned firms achieved complete import substitution un­
der the strong quantitative controls on imports, high tariffs, and foreign ex­
change controls of the postwar period (cf. Chapter 7, pages 159-164).
Expansion in production volume led to cost reductions, and ultimately to ex­
port growth and export market diversification.

CPC Development of the Iron and Steel Industry 107
Industrial machinery includes machinery and tools, excluding electrical
machinery and transportation equipment, and thus represents all types of
industrial and precision machinery. Many of these industries began produc­
tion even before the war. Cameras and watches had reached the export stage
by the early postwar years, whereas machine tools, and industrial robots only
began to be exported in recent years. Disregarding the many different product
types at every development stage of the industry, production value in 1980
was 17.6 trillion yen. The ratio of imports to domestic demand was 4.7 per­
cent, and the export/production ratio reached 19.1 percent.
The time span from the start of production to import substitution and then
exports (depicted by the slope of the S/D curve) differs from industry to in­
dustry and depends on the influence of many factors. The faster the growth
of domestic demand and the greater the possibility for the achievement of
economies of scale, the shorter is the time lag involved. The more difficult
the technology to be absorbed is, however, the longer the process will take
to reach its final stage (Yamazawa 1972). These observations conform with
the basic mechanism of CPC development presented in Figure 4.1.
Mechanism of Industrial Diversification
What patterns can be discerned in the CPC development of different in­
dustries? Within a given industry, there is a tendency for cruder manufac­
tured products to develop earlier than the more sophisticated ones (for
example, low-count as opposed to high-count cotton yarn or black-and-white
television sets as opposed to color televisions). As discussed earlier, however,
such a pattern also depends on the degree of ease with which the technology
is absorbed as well as on the relative speed of domestic demand growth.
The second pattern is that of backward and forward linkage effects. Since
import substitution in consumer goods and finished products requires inter­
mediate goods, parts, and capital goods inputs, it initially induces imports
of these goods. Later, the demand for these goods leads to the start of new
industries and is known as the backward linkage effect. Examples are the de­
velopment of the chemical dyestuffs and textile machinery industries, which
was induced by the growth of the textile industry (cf. Chapter 4). The for­
ward linkage effect consists of the development of industries supplying the
intermediate goods or raw materials at low prices, such as the development
of a cotton yarn industry for cloth production or the development of the iron
and steel industry for the production of automobiles and ships.
The diversification of CPC development into new industries depends on
the relative factor endowments of the country, particularly on its capital ac­
cumulation and the scale of its domestic market. Due to constraints in the
degree of capital accumulation, labor-intensive industries such as the textile

108 Industry Growth and International TYade
industry are introduced in the early stages of industrialization. Where the
size of the domestic market is small, however, it is difficult to achieve econo­
mies of scale. Even more important than these conditions is the ability of
entrepreneurs to correctly identify the possibility of domestic demand growth
as well as the existence of potential comparative advantage so as to begin
the process of import substitution production (this topic is covered in Chap­
ter 6, which focuses on entrepreneurial ability, as exemplified by trading com­
panies, and its relationship to the opening of trade and new industries).
Finally, there should be some mention of the development of construction
industries and public utility industries such as communications and electric
power. Since these industries produce nontradable goods (services), the CPC
development process is not evident. However, they have greatly contributed
to the development of other manufacturing industries. Excluding the 1920s,
the growth of the public utility industries during the pre-Second World War
period occurred at the high annual rate of 9 percent, while that of the con­
struction industry grew at 8.5 percent and included the construction of the
transport-communications and electric power service network. It would be
interesting to see how the CPC development in Figure 5.3 corresponds to the
development of the construction and public utility industries, with the peri­
od when each of these industries achieved its nationwide organization denoted
(that is, 1873-83 for railways, 1873-83 for the postal service system, 1896-1900
for the telephone system).
The proportion of factories equipped with electric motors was 28.2 per­
cent in 1909. By 1929 the proportion had risen to 87.0 percent. This infra­
structure was an important factor favoring the effective development of
modern manufacturing industries. Thus, although the government's protec­
tive and industrial nurturing policy certainly contributed to the promotion
of CPC development, except in the case of strategic industries such as iron
and steel and automobiles, a high-handed nurturing policy was rare. Rather,
the Japanese government offered indirect support, including the development
of necessary infrastructure.

Six
Trading Companies and
the Expansion of Foreign Trade
This chapter analyzes the role of trading companies in the expansion of
Japan's foreign trade before the Second World War. Only a few articles have
been devoted to this subject. The theoretical literature on foreign trade has
concentrated on such aspects as income growth, capital accumulation, and
technical progress, and the players involved in Japanese foreign trade trans­
actions have seldom been mentioned.
Recently, however, there has been increasing interest in trading companies,
especially general trading companies (GTCs), which are peculiar to Japan.
Several articles on these companies have been published by business histori­
ans. Nakagawa (1967) and Yamamura (1976), for example, examine the ori­
gin of GTCs and try to explain why they developed. Because trading companies
were one of the forces promoting the CPC development of Japanese indus­
tries, this chapter concentrates on the interrelationship between the develop­
ment of trading companies and Japan's foreign trade expansion.
FOREIGN TRADE TRANSACTIONS IN THE MEIJI PERIOD
Merchant House Trade and Direct Trade Policy
Japan's foreign trade began at merchant houses in foreign settlements that
were established at major port cities in Japan. 1 Japanese sellers brought such
products as raw silk, tea, crafts, and artifacts from inland areas to these mer­
chant houses while Japanese buyers bought cotton cloth and sundry Western
goods to be sold in domestic markets. Western merchants and their Chinese
interpreters set the prices and typically dictated the terms of the transactions.
Japanese merchants knew neither how much would be paid for their products
nor how much their purchases cost abroad.

110 Industry Growth and International Trade
Initially, the Japanese merchants could not compete on an equal footing
with foreign merchants. After 250 years of seclusion, the Japanese had no
experience in the business of foreign trade and no knowledge of currency ex­
change, marine insurance, and ocean shipping. Furthermore, they were poorly
equipped financially, and some Japanese sellers were paid by foreign mer­
chants in advance for their commodities.
In 1875, after a futile attempt to foster semiofficial trading companies, the
government announced it would subsidize private trading companies in or­
der to promote direct exports. 1 A score of small trading companies (includ­
ing Mitsui) responded and attempted the direct export of raw silk, tea, and
other Japanese specialties. They were also encouraged by the nationwide move­
ment demanding the abolition of extraterritoriality. 3 Mitsui and Company
opened five branches in Europe and the United States (London, Paris, Lyons,
Milan, and New York), and by 1878 more than 70 merchants had gone to
Europe and North America with samples in an attempt to establish direct
contacts with customers.
Most Japanese efforts failed due to the lack of experience of the Japanese
in foreign trade and their lack of access to trade-promoting services. By 1882,
Mitsui and Company had closed all overseas branches, except one branch in
London, and the raw-silk trade continued to be dominated by foreign mer­
chants. Japanese merchants received unfavorable treatment from foreign
banks, marine insurance companies, and shipping companies. 4 Direct trade
began to expand steadily only after the establishment of trade-promoting ser­
vice agencies such as the Yokohama Specie Bank in 1880, the Tokyo Marine
Insurance Company in 1879, and Nihon Yusen Kaisha for ocean shipping
in 1885, which were either financed or heavily subsidized by the government.
Expansion of Direct Trade
The expansion of direct exports and imports is shown in Figure 6.1. Statis­
tics for Japan Proper are not available for the years 1901-36, but the trend
during these years can be estimated from statistics on record at the ports of
Kobe and Yokohama. The direct export ratio rose to between 13 and 15 per­
cent after the government's promotional policy began. It remained stagnant
until 1892, when it again began to rise, reaching 35 percent by 1900. The direct
import ratio, which began to rise after the early 1880s, lagged behind the direct
export ratio but reached 40 percent by 1900 and 90 percent by 1920.
The period 1884-1920 is characterized by rapid expansion of the total
volume of foreign trade. In the first 16 years of this period, from 1885 to
1900, both exports and imports expanded rapidly (7.7% and 8.9%, respec­
tively, in terms of the annual compound rate). Exports and imports continued
to expand during the rest of the period, albeit at slightly lower rates (by 6.2%
and 4.3%, respectively).

Trading Companies and Foreign Trade Expansion 111
Figure 6.1
Japan's Direct Export and Import Ratios: 1874-1960

1870 1880 1890 1900 1910 1920 1930 1940 1950 1960
Year
Note: All Japan (Proper), and Yokohama or Kobe indicate ratios are calculated from figures
covering all of Japan (excluding colonies) and figures covering only the Ports of Yokohama
and Kobe.
Sources: Ministry of Finance (1882-1943); MITI (1952-60); Asahi Shinbunsha (1930).
During this period, Japan's export and import structures changed dramat­
ically. Raw silk and tea, which comprised 80 percent of total exports in the
early 1880s, were rapidly replaced by light manufactures, whose market share
increased from 13 percent in the 1880s to 43 percent by the early 1900s. Im­
ports of light manufactures, on the other hand, had a market share of 68
percent in the 1880s, but by the early 1900s their share had fallen to 24 per­
cent. By this time, imports of heavy manufactures and raw materials exceed­
ed imports of light manufactures with shares at 33 percent and 29 percent,
respectively.
The regional composition of Japan's foreign trade changed markedly from
1880 to 1990; Europe's share of exports and imports declined from 38 to 21
percent and 63 to 37 percent, respectively, while East Asia and North Amer­
ica became Japan's major trading partners. This reflects the evolution of
Japan's foreign trade, which began with Japan as a producer of primary goods
immediately after the opening-up of trade to Japan becoming an industrial
exporter.

112 Industry Growth and International Trade
From 1900 to 1920, the changes in Japan's foreign trade structure were
strengthened. Japan's export share of light manufactures continued to ex­
pand and reached 49 percent by 1920, and the share of light manufactures
in total imports contracted to 15 percent. Exports of a few heavy manufac­
tures also began at this time, but the share of imports of raw materials and
heavy manufactures in Japan's total imports reached 31 and 38 percent, respec­
tively. Whereas East Asia and North America continued to be major trading
partners, Europe's share of exports to Japan declined to 8 percent and im­
ports from Japan decreased to 14 percent due to the diversification of both
export destinations and import sources.
The expansion of direct trade—that is, the replacement of foreign merchants
with Japanese merchants in Japan's foreign trade—was closely associated with
the rapid change in both the commodity and regional trade structures. The
values of individual commodities that were traded by foreign and Japanese
merchants as recorded in the customs statistics for the years 1884-1900 are
the source for the data used to support this assertion below.
A sample of 110 export and 231 import commodities was selected; the to­
tal value and percentage share of Japanese merchants are calculated for each
commodity in three sets of three-year averages: 1884-86, 1891-93, and
1898-1900. The sample covers between 87 and 89 percent of total exports,
and between 88 and 94 percent of total imports throughout the three peri­
ods. Exports are classified as follows:
• indigenous (traditional Japanese manufactures) versus modern goods
(goods produced by introduced technology)
• old versus new exports (exports that took place before or after 1868)
• exports primarily to Europe and North America (Europe-North America)
East Asia, and the rest of the world.
Imports are classified as follows:
• by end use
• old versus new imports
• imports primarily from Europe-North America, East Asia, and the rest
of the world. 5
Changes in the shares of major export and import groups are shown in
Tables 6.1 and 6.2. As Table 6.1 shows, distinct shifts in exports occurred from
old to new exports, from indigenous goods to modern goods, and from
Europe-North America to East Asia. To highlight the importance of the East
Asian market, Europe and North America are grouped together. It should
be noted, however, that exports of new indigenous products contributed to
export expansion in later years. The export and import shares for the group
Indigenous, Old, and Europe-North America declined markedly and contrast

Table 6.1
Shares in Total Exports, Direct Export Ratios, and Rates of Increase in Exports by Major Commodity Groups.
Shares in Rates of increase
total exports 3 Direct export ratios 3 in exports 3
Commodity group lb 3 1 2 3 2/l c 3/2 3/1
All sample commodities 100:0 100.0 19.2 19.4 32.0 13.72 12.39 .64.42
Old exports (Old) 93.4 72.3 15.7 16.6 29.6 4.08 2.12 11.82
New exports (New) 6.6 27.8 22.6 21.9 33.7 22.85 21.95 114.22
Indigenous goods (Indigenous) 99.1 80.7 16.0 18.2 29.9 11.95 3.77 40.32
Modern goods (Modern) 0.9 19.3 38.3 25.9 43.4 24.05 59.58 204.58
Europe-North America 72.2 57.0 15.3 14.0 26.7 22.91 3.92 72.21
East Asia 21.6 38.3 19.9 20.7 32.3 9.19 19.49 67.76
Indigenous and New 5.9 8.5 16.1 20.7 30.5 21.84 5.86 76.21
Indigenous and Old 93.2 72.2 15.8 16.1 29.4 4.15 2.13 12.01
Modern and New 0.7 19.2 40.1 25.0 41.7 25.61 63.21 218.14
Indigenous, Old, and. Europe-North America 68.4 50.0 12.9 10.3 22.9 6.67 1.52 22.69
Modern, New, and East Asia 0.7 19.1 43.9 27.4 41.9 28.43 71.90 248.87
Notes:
a. Arithmetic means for major commodity groups.
b. Year 1, 1884-86; year 2, 1891-93; year 3, 1898-1900.
c. The ratio of export value in year 2 to that in year 1. See note 5 for commodity classifications.
Source: Ministry of Finance, Dainihon Gaikoku Boeki Nempyo (Foreign Trade Annual of the Empire of Japan), 1884-86, 1891-93, 1898-1900.

Table 6.2
Shares in Total Imports, Direct Import Ratios, and Rates of Increase in Imports by Major Commodity Groups
Shares in
total imports 3 Direct import ratios 3 Rates of increase
in imports 3
Commodity group lb 3 1 2 3 2/l c 3/2 3/1
All sample commodities 100.0 100.0 13.9 19.9 32.6 4.88 4.78 30.20
Old imports (Old) 63.7 39.1 12.3 17.2 29.6 3.17 .85 22.20
New imports (New) 36.3 60.9 16.0 23.3 36.6 7.13 4.68 40.69
Europe-North America 72.3 44.9 13.1 21.2 34.9 4.48 4.63 23.79
East Asia 4.8 18.4 18.9 20.9 25.7 7.18 7.07 74.10
Croup A: Unprocessed foodstuffs, raw
materials, and investment goods 20.2 57.5 19.4 26.8 38.4 8.66 5.72 71.08
Group B: Processed foodstuffs, intermediate
goods, and consumption goods 79.8 42.6 11.5 16.8 30.0 3.21 4.36 12.07
New and Group A 6.0 45.0 24.2 34.7 44.1 13.18 4.67 86.25
Old and Group B 49.6 26.8 11.3 16.4 28.2 2.64 4.12 8.45
New and Group B 30.0 15.8 11.9 17.5 32.7 4.02 4.69 17.22
Old and consumption goods and Europe-North
America 22.4 11.1 15.4 22.1 33.3 2.46 3.50 7.74
New and investment goods and Europe-North
America 1.0 3.0 21.6 49.8 59.9 18.57 4.99 136.07
Notes:
a. Arithmetic means for major commodity groups.
b. Year 1. 1884-86; year 2, 1891-93; year 3, 1898-1900.
c. The ratio of import values in year 2 to that in year 1. See note 5 for commodity classification.
Source: Ministry of Finance, Dainihon Gaikoku Boeki Nempyo (Foreign Trade Annual of the Empire of Japan), 1884-86, 1891-93, 1898-1900-

Trading Companies and Foreign Trade Expansion 115
with the shares of the group Modern, New, and East Asia, which increased
rapidly and significantly.
As Table 6.2 shows, the share of imports of processed foodstuffs, inter­
mediate goods, and consumption goods declined, whereas the share of un­
processed foodstuffs, raw materials, and investment goods increased. Old
imports were replaced by new imports far more rapidly than occurred among
exports, indicating rapid import substitution of modern manufactures. The
source of imports shifted from Europe-North America to East Asia and to
other areas. These structural changes in imports and exports, and their as­
sociation with increasing direct trade ratios, are discussed in the next section.
THE EXPANSION ROLE
PLAYED BY JAPANESE TRADING COMPANIES
The simple association between direct trade ratios and trade expansion in
aggregate terms does not reveal much about the causal relationship between
the two. The contribution of Japanese merchants to the expansion of Japan's
foreign trade is best determined by a theoretical consideration of both the
function of trading companies and the changes in economic conditions af­
fecting them. In this section, a proposition on the difference between Japanese
and foreign merchants regarding business conduct is presented. As well, two
hypotheses are presented and tested using individual commodity data to de­
rive direct trade ratios and rates of trade expansion.
The Function of a Trading Company
The primary function of a trading company is to link sellers with buyers
across national borders and carry out transactions between the two groups.
A trading company's income is generated primarily from the commissions
the company receives upon the successful completion of transactions.
Foreign trade transactions require service inputs of various types:
• The trading company must acquire both demand and supply information
(an accumulated managerial asset) so that it can connect the buyer that
is paying the highest price with the producer showing the lowest cost.
• The trading company must acquire enough skills in order to conduct the
complicated procedures of foreign transactions (an accumulated managerial
asset), and must also have the trust of both the buyer and the seller to
act as an intermediary.
• The trading company must be able to provide the seller (or buyer) with
trade credit until the transaction is completed.

116 Industry Growth and International Trade
• The trading company must be able to arrange services such as ocean trans­
portation and marine insurance.
In terms of the production function, a trading company's output is the con­
ducting of foreign trade transactions, whereas its inputs are information, skill
in conducting foreign trade transactions, and such services as obtaining trade
credit, marine insurance, and ocean transportation. The first two inputs are
provided by the trading company. Other services, which may be called trade-
promoting services, are provided through outside firms that specialize in these
services, but the trading company must procure and combine them at the right
time and place.
In conventional economic analysis, the growth of trade is attributed to such
factors as the reduction of production costs, increases in productive capac­
ity, and demand growth. In reality, however, a trading company's services are
required for these factors to be realized in the actual growth of trade. In this
sense, these factors may be called factors of trade potential. Since every trad­
ing company that can realize this trade potential would face similar profit
opportunities, the incentives for both Japanese and foreign merchants would
be equal, provided that other incentives are not at work. It appears that in
the early stages of Japan's trade expansion Japanese merchants had addi­
tional nonprofit incentives as well. They were either driven by economic na­
tionalism or assisted by any advantage they might have had over foreign
competitors. 6
Although Japanese merchants were eager to expand their business, they
were at first greatly hampered by foreign merchants. When Japan was opened
to foreign trade in 1859, Japanese merchants were inexperienced in foreign
transactions, lacking both knowledge and skill, and they had little informa­
tion about the foreign markets for their products. No trade-promoting ser­
vices were available from a Japanese source even at the end of the 1870s.
Moreover, before the establishment of the Yokohama Specie Bank in 1880
no Japanese bank provided trade credit. Japanese merchants had no choice
but to seek funds on unfavorable terms from Yokohama branches of foreign
banks such as the Oriental Banking Corporation and the Hong Kong and
Shanghai Banking Corporation. 7 It was a matter of course that almost all
Japanese foreign trade was conducted by European and American merchants,
who were equipped with the necessary managerial assets and access to trade-
promoting services.
Gradually, information and skill in foreign transactions were accumulated
by Japanese trading companies through on-the-job training and formal edu­
cation in commerce. Japanese merchants acquired the necessary knowledge
and trading skills through their contacts with foreign merchants. Some
Japanese, after working at foreign merchant houses and acquiring the neces­
sary skills, started their own businesses. Takashi Masuda, the first general

Trading Companies and Foreign TYade Expansion J J 7
manager of Mitsui and Company, is an example. The first school of com­
merce, Sh5ho Koshujo (which later became Hitotsubashi University), was es­
tablished in 1876 and began supplying graduates equipped with an education
in commerce and with knowledge of foreign languages to Mitsui and other
Japanese trading companies. The companies, in turn, sent their new employees
abroad to begin overseas branches. The new corps of specially trained per­
sonnel, the accumulation of trade skills, and an improved supply of Japanese
trade-promoting services tended to lessen the disadvantages experienced by
Japanese merchants.
In several respects, however, Japanese merchants had an advantage over
their European and American competitors. First, European and American
merchants charged a much higher commission than did Japanese merchants,
partly because of the high cost of living and partly because of the commis­
sions they in turn had to pay to Chinese or Japanese employees. Second,
Japanese merchants had an advantage both in exploiting new demand within
Japan for imported products and in organizing new Japanese supply sources
of Japanese products for export. Without these advantages, the increase in
the direct export ratio would have been much slower. It is true, however, that
these advantages tended to be offset by the disadvantage of Japan's lack of
knowledge of and access to foreign markets. Nevertheless, the advantages of
Japanese merchants at home tended to more than offset their disadvantages
abroad, which is discussed in more detail in the following section.' Because
Japan, as a small, new entrant to the world trade arena, could not offer either
new products or a large, new market, new opportunities for trade between
Japan and the rest of the world were found mainly in Japan.
The Treaty of Amity and Commerce in 1858 restricted foreign merchants
in their business activities outside foreign settlements. In exchange for ex­
traterritoriality in foreign settlements, foreign merchants could not travel freely
beyond a 40-kilometer radius outside their settlements. This regulation re­
stricted direct contact of foreign merchants with local customers or producers
and tended to strengthen the Japanese advantages mentioned above. 9 However,
since foreign merchants relied heavily on their Japanese employees in trans­
actions with the Japanese, it seems likely that the travel regulation did not
give Japanese merchants enough of an advantage to offset any disadvantage.
Comparative Advantage of Japanese Trading Companies
The direct trade ratio did not increase uniformly across all commodities
but increased early on for some commodities and for trade with some regions.
Two testable hypotheses regarding this situation are advanced below to prove
the proposition that Japanese merchants had greater incentive to expand their
businesses than did their foreign competitors.

118 Industry Growth and International Trade
Hypothesis 1 concerns the differences in the increase in direct trade in var­
ious commodities and with different regions. Compared with their foreign
competitors, Japanese merchants were at first at a disadvantage in all com­
modities, but the degree of disadvantage differed from one commodity to
the next. It seems plausible that Japanese merchants tended to increase their
shares in those commodities in which they had comparative advantage.
Hypothesis 1 may, therefore, be stated as follows: direct trade ratios tended
to increase each year (1) in new exports (or imports) rather than in old ones;
(2) in exports of modern goods rather than in exports of indigenous ones,
and in imports of such commodity categories as unprocessed foodstuffs, raw
materials, and investment goods rather than in imports of other categories;
and (3) in export and import trading with East Asia rather than with Europe
and North America.
There are several reasons behind hypothesis 1 which suggest that Japanese
merchants had comparative advantage in trade of various commodities and
with specific regions. First, trade in old exports such as raw silk and tea and
old imports such as cotton cloth and sundry Western goods began immedi­
ately after the opening of trade. European and American merchants had long-
exploited supply sources and established marketing channels for those com­
modities. When Japanese merchants entered the foreign trade market, they
found it easier to begin to export and import new commodities or to enter
areas where markets were not fully established.
Second, modern goods such as matches and cotton yarn were initially im­
ported from Europe. But they were soon produced domestically, and these
domestically produced goods eventually replaced imports (import substitu­
tion). These goods were then exported, primarily to Asian countries. Japanese
merchants often worked closely with new industries and were eager to sell
their products abroad, whereas foreign merchants, who had initially brought
the product to Japan, tended to underestimate the quality of Japanese
products and did not want to risk trading them.
Moreover, the advantage of having close ties to the market helped Japanese
merchants enter the import trade in such commodities as unprocessed food­
stuffs, raw materials, and investment goods. The demand for imported rice
and agricultural products was greatly affected by each year's domestic yields,
and because of their long experience in the domestic trade of these products,
Japanese merchants held an absolute advantage. Cotton and other raw materi­
als, as well as industrial machinery, were in demand by the new industries,
such as those producing yarn and matches, to which Mitsui and other Japanese
trading companies were closely tied from the start.
Third, European and American merchants exploited the trade between
Japan and their own countries and maintained a strong advantage over the
Japanese. However, when trade between Japan and East Asia later expanded,

Trading Companies and Foreign Trade Expansion 119
Japanese merchants were at less of a disadvantage in the region vis-a-vis their
European and American competitors. In their business with China, Euro­
pean and American merchants had typically relied on Chinese compradors,
to whom they paid a 1 percent commission. Mitsui, however, avoided this
custom by introducing instead a program of Chinese language instruction
for its Japanese employees when it attempted export expansion into the
Chinese market in the late 1890s. 10 Here, however, Japanese merchants had
another foreign competitor: Chinese merchants who had long dominated trade
within this region.
Hypothesis 2 is also based on the proposition that the Japanese merchants'
incentive to expand foreign trade was stronger than that of their foreign com­
petitors: For a given commodity, the higher the direct trade ratio or the more
rapid its increase, the higher the rate of trade expansion will be in that com­
modity.
Tables 6.1 and 6.2 show average direct export and import ratios for three
periods, 1884-86 (period 1), 1891-93 (period 2) and 1898-1900 (period 3) and
average rates of export and import expansion between these periods (i.e., 2/1
refers to the ratio of export [import] value in period 2 over the export [im­
port] value in period 1) for major commodity categories. A simple compari­
son of averages between the ratios for the different categories supports
hypothesis 1—that is, direct trade ratios of new exports and imports (group
A, Table 6.2) exceed those of old exports (group B, Table 6.2) each year. Simi­
lar differences are found between the export of modern goods and indigenous
goods, and between import groups A and B. The differences by export desti­
nation are also consistent with hypothesis 1 (part 3), but those by import source
turned out to be obscure. All differences between pairs of combined categories
such as Indigenous, Old, and Europe-North America and Modern, New, and
East Asia are consistent with the hypothesis."
Comparing the rate of trade expansion between the categories yields differ­
ences that parallel the differences between direct trade ratios, which implies
a positive association between direct trade ratios and trade expansion.
However, this relationship between averages of individual categories is insuffi­
cient to support hypotheses 1 and 2: a rigorous test of the two hypotheses
by regression analysis based on individual commodity data is needed.
Regression Analysis
of Direct Trade Ratios and Trade Expansion
The following set of regression equations is given for exports (a set of regres­
sion equations for imports can be derived by substituting DIR and GI for
DER and GE, respectively):

120 Industry Growth and International Trade
In DER, a0 + OiN + a zM + a 3Rt + a tRi + a sNR,
+ aJvfR, + a,NM (1)
In GE, •l/s bQ + b,N + bzM + biR t + b<R 2 + faNRt
+ bJWR, + b,NM + b t\n DER S
+ b,\n GDER I/S (2)
where DER,(DIR,) is the direct export (import) ratio at period t (t = 1, 2,
3); GE l/s(GI l/s) is the ratio of the export (import) value in period t to the
export (import) value in period s; GDER l/s(GDIR l/5) is the ratio of the direct
export (import) value in period t to the direct export (import) value in period
s; N and M are dummy variables for new exports (imports) and exports of
modern goods (imports of unprocessed foodstuffs, raw materials, and invest­
ment goods), respectively; /?, and R 2 are regional dummies; and /V/?„ MR lt
and NM are interaction terms between N, A/, and /?,. 12
The first equation is formulated to test hypothesis I (parts 1, 2, and 3) by
individual commodity data: that is, the equation examines how the DER(DIR)
of individual commodities is affected by commodity categories such as Old
or New. The second equation is formulated to test hypothesis 2 by estimating
the effect on the GE(GI) during a particular period of both the DER(DIR)
at the beginning of the period and the GDER(GDIR) during the same pe­
riod as well as the effect of commodity categories mentioned above.
The log-linear formulation of the two equations implies a multiplicative
relationship between variables. 13 Interaction terms are included in order to
explain high DERs or DIRs in the combined categories such as the category
New, Modern, and East Asia in Table 6.1 and the category New and Group
A in Table 6.2. The two equations are estimated using data for 109 export
and 231 import commodities for the three periods (1884-86, 1891-93, and
1898-1900). Alternative specification sets of interaction terms are estimated,
and the results from one of the regressions with the best fit are listed in Ta­
bles 6.3 and 6.4. Coefficients of determination are generally low. This is be­
cause these equations are estimated from cross-commodity data of large
sample size and because the specification does not include such powerful ex­
planatory variables as income and relative prices, which are used in conven­
tional export and import analysis. However, significant Fand /-values support
the relevance of the specification in spite of low R 2s.
This detailed examination of the individual equations allows the following
interpretation. As shown in regression equation IE in Table 6.3, the introduc­
tion of NR } produced significant and positive coefficients but resulted in the
coefficients for N and M becoming nonsignificant and of an indeterminate
sign. 14 The coefficient for R, is significant and is negative (this was true in
an alternative specification of the interaction terms). These results imply that

Table 6.3
Regression of Direct Export Ratio and Export Expansion
CONSTANT N M R, /?, AT?, In DER S In DGER l/s R 1 F
(IE) In DER t
(a) / = 1 -2.7025 -0.2535 0.8848 – 1.3026 0.7177 1.7673 0.0650 2.50 b
(0.44) (0.74) (2.24) (1-00) (2.26)
(b) t = 2 – 2.4902 -0.1741 0.6738 – 1.2765 0.8516 1.5469 0.0848 3.00 b
(0.33) (0.63) (2.44) c (1.32) a (2.20) b
(c) / = 3 – 1.5699 -0.0766 0.8735 -0.6131 0.5146 0.5369 0.0193 1.43
(0.19) (1.06) (1.52) (1-03) (0.99)
IE) In GE Us
(a) t/s = 2/1 1.2316 0.4599 -0.0674 0.1531 0.2438 0.1858 -0.1053 0.1241 3.55 c
(1.80) b (0.09) (0.58) (0.52) (2.49) c (1-18)
(b) t/s = 3/2 0.8655 0.7053 -0.5510 0.0864 0.33368 0.2464 0.2346 0.2511 7.04 c
(4.36) (1.13) (0.51) (113) (4.19) c (2.86) c
(c) t/s = 3/1 2.2212 1.1199 -0.5737 0.2524 0.5304 0.4368 0.1446 0.2643 7.47 c
(3.73) (0-63) (0.80) (0.96) (3.90) c (1.16)
Notes: Estimated by ordinary least squares method. Numbers in parentheses show /-values. See text for the explanation of symbols.
a. Significant at the 90 percent level.
b. Significant at the 95 percent level.
c. Significant at the 99 percent level.

Table 6.4
Regression of Direct Import Ratio and Import Expansion
CONSTANT N M /?, R, MR> In DIR S In GDIR t/s R 1 F
(II) In D/R t
(a) / = 1 3.2339 -0.2682 0.3277 -0.5130 – 1.0842 0.6102 0-0161 1.75
(0.85) (0.79) (I.30) a (2.44) c (1.08)
(b) / = 2 2.5526 -0.0636 0.1557 -1.1814 -1.5900 1.0650 0.0874 5.41 c
(0.21) (0.40) (3.18) c (3.80) c (1.98) b
(c) / = 3 1.6858 0.3452 0.5167 -1.0432 -0.4838 -0.2597 0.0968 5.93 c
(1.89) b (2.12) b (4.52) c (I.86) b (0.01)
!1) In GI t/s
(a) t/s = 2/1 0.6566 0.5618 0.2538 -0.0797 0.0096 0.4586 0.0681 0.0249 0.1571 7.I3 C
(3.52) c (1.21) (0.39) (0.04) (1.60) a (1.77)b (0.55)
(b) t/s = 3/2 1.3491 0.1242 0.3779 0.0775 -0.5090 -0.2643 0.1757 0.1558 0.0797 3.85 c
(0.88) (2.05) b (0.43) (2.53) c (1.05) (3.51) c (2.59) c
(c) t/s = 3/1 2.0123 0.6812 0.6173 -0.0166 -0.5320 0.2268 0.2482 0.1806 0.1852 8.47 c
(3.27) c (2.27) b (0.06) (1.83) b (0.62) (3.35) (2.13) b
Notes: Estimated by ordinary least squares method. Numbers in parentheses show /-values. See text for the explanation of symbols.
a. Significant at the 90 percent level.
b. Significant at the 95 percent level.
c. Significant at the 99 percent level.

Trading Companies and Foreign Trade Expansion 123
the DER is raised significantly when commodities N and M are sent to region
The effect of N x M x R x is illustrated in equation IE by comparing the
constant term (that is, the DER) when N = M=R l=R 2 = 0 with the
DER when N = M = R, = 1. Converted into percentages, the DERs for
the category Modern, New, and East Asia in the periods 1, 2, and 3 turn out
to be 20.6 percent, 17.9 percent, and 42.8 percent, which are two or three times
as high as the DERs for the category Indigenous, Old, and Europe-North
America (6.7%, 8.3%, and 20.8%, in the respective periods). This result is
roughly consistent with the data in Table 6.1 and supports hypothesis 1. 15 The
coefficients of R 2 are stable and positive, which may be attributed to special
circumstances, such as the fact that the R 2 category includes only a small
number of foodstuffs (rice, peas, rice wine, soy sauce, Japanese noodles) in
the category Old and Indigenous regardless of destination.
Regression equation 11 in Table 6.4 was also estimated including the vari­
able NM, though this is not shown in the table. The interaction NM has the
best fit, but it affects the coefficients of N and M. Positive and significant
NM produces nonsignificant Nand M with a positive or negative sign in peri­
ods 1 and 2, while in period 3 positive and significant Mand Nare accompa­
nied by nonsignificant NM. In the absence of an interaction term, both M
and N turn out to be positive and M is significant in the three periods. In
contrast, the negative and significant coefficients of both tf, and R 2 are not
affected by the interaction between M and N.
To conclude, both N and M, or their interaction NM, tend to raise the DIR,
whereas R t and R 2 tend to lower the DIR. This is not inconsistent with
hypothesis 1, parts 1 and 2, but it does refute part 3 of hypothesis 1. The
negative effect of R } on both the DER and DIR suggests that it was never
easier for Japanese merchants to enter the export or import trade with neigh­
boring countries than it was to enter trade with Europe and North America.
Both the DER and DIR tended to be lower instead of higher in trade with
neighboring countries, which is contrary to hypothesis 1, part 3. This may
be partly attributed to competition from Chinese and Indian merchants and
to the fact that many European and American merchants had a firm foothold
in East Asia.
In Table 6.3, regression equation 2E suggests that N and the DER tend
to significantly accelerate GE. Both R, and R 2 are positively correlated with
DER, but neither is significant. NR, is omitted from equation 2E since it
neither produces a significant coefficient with a stable sign nor significantly
effects the coefficients of other variables. An ambiguous effect of M is at­
tributed to the interdependence of the M and N categories. The DER turns
out to be positively related with export expansion, which supports hypothe­
sis 2. However, the acceleration of the GE by the GDER is significant only

124 Industry Growth and International Trade
for the period 3/2 and ambiguous for the period 2/1. For imports, both M
and N (and NM in period 2) have positive and significant correlation with
GI, but the effects of R t and /?, are ambiguous. Both the DIR and the GDIR
tend to accelerate GI signiFicantly, which is consistent with hypothesis 2.
Negative and nonsignificant coefficients for GDER and a nonsignificant
coefficient for GDIR for the period 2/1 (Tables 6.3 and 6.4) are inconsistent
with hypothesis 2. A more careful examination of GE and GDER for major
individual commodities may help to resolve this contradiction (Table 6.5). Ex­
port data for silk fabric, umbrellas, cotton flannel, and cotton yarn in period
1 illustrate that high direct export ratios (period 2 in the case of cotton yarn)
are associated with high rates of export expansion during the period 2/1 (pe­
riod 3/2 in the case of cotton yarn). All of the goods were new exports, and
Japanese merchants exploited their trade potential by expanding exports.
However, Japanese shares declined and foreign competitors' shares increased
in all cases in the following period. This seems to imply that foreign mer­
chants were not passive; rather than stay on the sidelines, they would enter
into trading those new goods that showed good prospects for expansion. This
competition between Japanese and foreign merchants was vigorous, and the
replacement of foreign merchants by Japanese merchants did not take place
smoothly or amicably.
Table 6.5
Direct Trade Ratios and Rates of Increase of Selected Commodities
Direct Trade Ratios Rates of Increase
/=1 / = 2 / = 3 t/s = 2/\ r/s = 3/2
Exports
Raw silk 15.6 9.0 29.2 2.3 1.6
Habutae (silk fabric) 61.9 16.0 19.9 73.5 4.8
Umbrellas (Western) 48.9 4.3 12.0 64.1 2.2
Cotton flannel 84.4 30.7 21.4 122.6 4.2
Cotton yarn — 80.9 59.6 — 926.0
n ports
Raw cotton 7.8 44.8 57.0 18.7 5.0
Rice 12.7 46.2 27.3 13.1 6.9
Indigo 92.4 10.1 13.7 10.9 8.9
Spinning machinery 78.8 65.1 74.3 13.7 1.7
Locomotives 6.3 52.8 80.6 4.7 6.4
Note: — = Periods when ihere were no exports.
Source: Ministry of Finance, Dainihon Gaikoku Boeki Nempyo (Foreign Trade Annual of the
Empire of Japan), 1884-86, 1891-93, 1898-1900.

Trading Companies and Foreign Trade Expansion 125
Evidence similar to that in the export analysis is found for imports of in­
digo and spinning machinery. In this case, however, the high rate of import
expansion was accompanied by an increasing Japanese share in such major
commodities as raw cotton, rice, and locomotives. This may explain the am­
biguous effect of the GDIR for the period 2/1 in Table 6.5.
The preceding analysis supports the proposition presented above. Japanese
merchants had comparative advantage in new exports and imports, exports
of modern goods, and imports closely related to modern industries, as reflected
by the higher DERs and DIRs for the commodities in those categories. The
higher ratios indicate that Japanese merchants were active in exploiting the
trade potential of those commodities. The expansion of exports and imports
also tended to be greater in commodities with higher and increasing direct
trade ratios. Thus Japanese merchants contributed to the expansion of Japan's
foreign trade.
The rapid development of Japanese trading companies did not, however,
occur independently, but was in fact strongly based on Japan's industrializa­
tion. This analysis does not refute the role of domestic industrialization in
the expansion of both exports and imports of N and M commodities, which
increased the share of Japanese merchants. This argument of a causal rela­
tionship is consistent with the previous finding that Japanese merchants were
relatively more involved in production and trade of commodities N and M
than their foreign competitors and took advantage of business opportunities
in these commodities to further increase their shares. It may be concluded
that there existed a strong interaction between the Japanese merchants' in­
crease in shares and the M- and TV-biased expansion of Japan's foreign trade.
INTENSIFICATION OF TRADING COMPANY ACTIVITY
BEFORE THE SECOND WORLD WAR
The rapid expansion of direct trade from 1900 to 1920 is attributed to the
increase in activity of Japanese trading companies, which evolved from the
earlier general trading companies. By 1910 Mitsui and Company had already
succeeded in diversifying its business in commodities and expanding to other
regions, and had become the premier foreign trading company in Japan. Mit­
subishi and Suzuki followed Mitsui's lead between 1910 and 1920, and by 1920
the three companies had established worldwide branch networks and were
trading in a wide variety of commodities.
Smaller trading companies generally specialized in exports such as textiles
and sundry goods and imports such as iron and machinery. These smaller
trading companies all contributed to domestic industrial development. Two
companies were established jointly by domestic cotton traders and spinning
firms to import raw cotton from India and the United States. Beginning in

126 Industry Growth and International Trade
the 1890s, their business expanded rapidly to meet the increasing demands
of the developing cotton textile industry. Foreign merchants, meanwhile,
watched their shares in Japan's foreign trade diminish steadily. They were
forced either to withdraw from Japan or take advantage of the few remain­
ing opportunities such as handling such commodities as fuels and chemicals.
During the interwar period, Japanese trading companies developed their
capacity far beyond the mere handling of transactions. They expanded their
operations by:
• diversifying commodities to include promising new products
• exploring new, growing markets
• and acting as brokers and organizing producers and consumers.
EMERGENCE OF GENERAL TRADING COMPANIES
Mitsui and Company led all of the general trading companies (GTCs) in
every area of activity described above. Between 1900 and 1910, Mitsui expanded
its share of coal exports from 45.8 to 82.5 percent, of machinery imports from
12.1 to 42.6 percent, and of raw-silk exports from 14.0 to 23.7 percent. Its
share for all commodities increased by 21 percent from 11.5 to 24 percent dur­
ing the same period and remained at that level thereafter. Its transactions were
diversified to include a wide variety of commodities, especially those associated
with both the export and import needs of growing industries. For example,
Mitsui imported raw cotton and spinning machinery and exported cotton yarn
and cloth—goods that were needed for the development of the cotton textile
industry. In 1910, its shares of raw-cotton imports and cotton yarn and cloth
exports reached 25.2, 33.0, and 51.0 percent, respectively.
Mitsui aggressively pursued new products that were projected to have high
domestic marketability and maintained its share of the new product market
through sole agent contracts. Exports of coal from the Miike mine in Japan
to other parts of Asia and imports of spinning machinery from Piatt and
Company illustrate this practice. In 1908, Mitsui had 80 sole agent contracts,
and by the late 1920s, the number had grown to 152, consisting mainly of
textile and electrical machinery contracts. The sole agent practice sometimes
led Mitsui to establish domestic firms for import-substituting production, as
was done for electrical machinery (Nihon Denki) in 1899 and rayon (Toyo
Rayon)in 1926.
Table 6.6 details Mitsui's regional expansion. The number of branches and
staff, as well as the value of sales, increased greatly from 1893 to 1910 and
expanded further by 1919. The percentage shares of both staff and sales
declined in Europe, Taiwan and Korea, and Manchuria, but increased in such
new areas as Southeast Asia, North America, and Oceania. In China, Mit-

Table 6.6
Features of Mitsui Branches by Region
Number of employees Sales in thousands of yen
Region Number of branches (% of total) (% of total sales)
Region 1893 1910 1919 1893 1910 1919 1893 1910 1919
Taiwan and Korea 0 5 8 0 69 186 0 15,511 67,869
(0.0) (19.9) (13.2) (10.4) (6.2)
Manchuria and Kwantung Province 0 8 8 0 67 210 0 13,906 76,100
(0.0) (19.3) (14.9) (9.3) (6.9)
China 3 9 18 26 106 521 4,790 42,871 217,835
(74.3) (30.5) (37.5) (48.2) (28.7) (19.8)
Southeast Asia 2 6 12 4 42 238 603 8,173 229,113
(11-4) (12.1) (16.9) (6.1) (5.5) (20.8)
Oceania 0 1 2 0 2 25. 0 108 14,756
(0.0) (0.6) (1.8) (0.9) (1.3)
Europe 1 3 3 5. 28 61 4,554 31,911 103,151
(14.3) (8.1) (4-3) (45.7) (21.3) (9.4)
North America 0 3 6 0 33 168 0 37,045 390,130
(0.0) (9.5) (11.9) (24.8) (35.5)
Total 6 35 57 35 347 1,409 9,947 149,525 1,098,954
(100) (100) (100) (100) (100) (100)
Note: Percentage shares are shown in parentheses.
Source: Yamaguchi et al., eds. (1974).

128 Industry Growth and International Trade
sui's share of staff increased, but its share of sales fell. In general, the changes
in shares show general expansion.
A similar trend of diversification was observed in the activities of Mitsubishi
and Suzuki. Initially, Mitsubishi was only a trading department within a larger
firm, Mitsubishi and Company, and specialized in the export of Japanese coal
to East and Southeast Asia. Eventually, Mitsubishi became a separate cor­
porate entity and established many branches in Asia, Europe, and North
America during the First World War. The firm also expanded the variety of
commodities traded. Suzuki, a trading company that was started in the late
1890s, exported specialty commodities from Japan and Taiwan such as pep­
permint, camphor, and sugar. After establishing branches in Europe, it rapidly
expanded and by the early 1920 had established a branch network.
Although not reflected in the increase in direct trade ratios in Figure 6.1,
the expansion of offshore trade between countries other than Japan with
Japanese traders as intermediaries merits brief mention. The first large-scale
trade of this type was conducted by Mitsui and Company in 1908, when it
sold soybeans from Manchuria to Europe. Mitsui conducted similar transac­
tions involving products such as raw cotton, sugar, and hemp bags. Mitsubishi
and Suzuki followed Mitsui's lead with similar moves. This implies that
Japanese trading companies had by that time progressed beyond simple com­
petition with Japan-based foreign merchants and had started competing for
shares in the world market.
The exporters' unions, organizations of domestic producers promoting
cotton-cloth exports, also deserve mention. Sanei and Nippon-Menpu, both
established in 1906, are well known. The former was established jointly by
three cotton weavers to promote exports to Korea, and the latter by five cot­
ton weavers for exports to Manchuria. However, they were actually led by
Mitsui and Company who conducted the export business for both unions.
Mitsui was the principal player in the efforts to expand these exports, and
Mitsui received little or no commission and provided the unions with trade
credits on very favorable terms.
Mitsui's success in expanding exports was remarkable in both the Korean
and Manchurian markets. In 1901, British cotton cloth had more than a 70
percent share of the Korean market and American cotton cloth almost 99
percent of the Manchurian market. But Mitsui succeeded in greatly reducing
prices (to 20-25% lower than American cotton cloth in Manchuria) and
quickly replaced the British and American products. By 1910, Mitsui had cap­
tured more than a 90 percent share of the market. 16
Trade-Promoting Services
The improved implementation of trade-promoting services should not be
overlooked. During the early 1880s, the government started to promote agents

Trading Companies and Foreign Trade Expansion 129
specializing in trade-promoting activities to supplement its direct trade pol­
icy, but this only came to fruition in the late 1890s and its effect in support­
ing the expansion of direct trade could only be felt after 1900. Marine
transportation, faced with severe competition from foreign shipping compa­
nies, had been heavily protected since the early Meiji period. In 1893, Japanese
ships entered the ocean transportation business with the introduction of ships
on the Indian route. This was followed by the introduction of ships to the
European, North American, and Australian routes in 1896. Japan's share of
the world's total freight tonnage rose from 22 percent in 1890 to 35 percent
in 1900, further increasing to 46 percent in 1910 and 64 percent in I920. ,T
The first Japanese marine insurance company, Tokyo Marine Insurance
Company, was established in 1879 with government support, but Japanese
traders had to continue to depend on foreign insurance companies. The for­
eign companies initially discriminated against Japanese traders by refusing
to insure ships that were steered by Japanese captains. In 1896, the second
marine insurance company, Nippon Kaijo, was established. It was followed
by a few others but these had not fully developed before the First World War."
By around 1890 the Yokohama Specie Bank had expanded its foreign ex­
change business, with substantial government support, to handle the trans­
actions of a large proportion of Japanese traders. But foreign banks continued
to dominate the Japanese market and colluded to set the market rate of the
yen around its gold par value. It was only after the First World War that
Japanese trading companies were free of the disadvantages resulting from
the dominance of these foreign banks.
POSTWAR EVOLUTION OF JAPANESE GTCs
After the Second World War, Japan experienced rapid economic growth
and dramatic change in its industrial and trade structure. In the 1960s, this
growth centered on heavy industries. The share of the metals, machinery, and
chemicals industries increased from 49 percent of manufacturing in 1950 to
61 percent in 1960 and reached 72 percent by 1970. The export structure also
changed. The shares of heavy industrial goods in total manufacturing exports
in 1951, 1960, and 1970 were 36, 44, and 78 percent, respectively.
Following the Second World War, ten major GTCs were established." Of
these ten, Mitsubishi and Mitsui are of zaibatsu origin and had already diver­
sified in the prewar period. Although these two companies were divided into
more than a hundred companies by the Occupation authorities in 1947, they
were later reunited—Mitsubishi in 1954 and Mitsui in 1959. Export substitu­
tion of heavy industrial goods was common for individual GTCs and was
most evident in the case of the Kansai Gomen, the five big trading compa­
nies in the Kansai (Osaka) area that specialized in textiles before the war. They
took over some domestic steel distributors to promote steel exports. At the

Table 6.7
Export Structure of Japanese GTCs: 1960 and 1970
Chemical
Company Year Foodstuffs Textiles products Metals Machinery Others Total
Mitsui 1960 20.2 22.5 13.4° 28.6 a I8.2 b 14.4 I17.4 a
1970 30.2 62.9 78.6 246.3 255.2 49.7 722.9
Mitsubishi 1960 24.6 20.2 0.7 b 20. l a 29.9* 16.5 111.9 s
1970 39.4 51.4 55.8 197.8 259.5 43,2 657.2
Marubeni 1960 8.1 46.6 2.5 b 11.4 b I8.5 b 10.2 97.3 a
1970 13.3 75.9 26.1 121.0 249.7 36.1 522.2
C. Itoh 1960 5.0 56.5 1.0 b 8.3 b 5.7 b 7.6 84.1
1970 14.5 101.6 21.3 94.3 132.9 21.1 385.7
Sumitomo 1960 0.7 a 0.6 b 3.2 b I6.1 a 4.1 b 2.6 27.4 b
1970 6.0 7.6 34.5 150.8 115.2 11.6 325.6
Nissho-Iwai 1960 c 1.4 b 11.9 — 14.5 a 2.8 b 4.2 a 36.7 a
1970 8.6 40.6 — 110.7 100.5 31.9 292.3
Toyo Menka 1960 2.5 b 31.7 0.8 b 5.1 a 3.8 b 7.7 51.6
1970 25.4 60.6 11.8 32.0 65.3 2.4 197.4
Nichimen 1960 1.6 a 40.6 1.6 b 8.0 14.4 6.2 72.3
1970 10.1 38.7 18.6 39.9 54.5 8.5 170.4

Chemical
Company Year Foodstuffs Textiles products Metals Machinery Others Total
Kanemaisu-Gosho 1960 d 3.4 25.6 — 9.3 11.1 12.1 61.6
1970 10.7 34.8 — 28.6 28.4 20.4 122.9
Ataka 1960 0.2 b 4.6 1.2 b 8.4 a 3.0 a 4.7 22.1
1970 3.9 23.4 12.5 44.3 19.6; 8.2 111.8
GTCs total e 1960 67.7 260.8 24.4 b 129.8 a 111.5 b 86.2 582.4
(70.3) (59.6) "(27.0) (58.9) (30.3) (33.8) (46.8).
1970 162.1 497.5 259.2 1,065.7 1,280.8 243.1 3,508.4
(69.5) (62.6) (58.3) (77.8) (39.8) (36.9) (50.4)
Japan's total f 1960 96.3 430.5 90.4 a 220.3 a 367.4 a 254.7 1,459.6
Notes: All figures, except those in parentheses, are in billions of yen. Fiscal years except as noted. Export price indices in 1970 are (1960= 100): food­
stuffs, 129; textiles, 103; chemical products, 77; metals, 103; machinery, 94; all commodities, 106.
a. The ratio of 1970 to 1960 exports for this company is above the 5.14 ratio of total 1970 GTC exports to total I960 GTC exports.
b. The ratio of 1970 to 1960 exports for this company is over 10.
c. The total of Nissho and Iwai, which were separate companies in 1961.
d. The total of Kanematsu and Gosho, which were separate companies in 1961.
e. Figures in parentheses are GTCs' percentage share of the total for Japan.
f. Calendar year.
Sources: Financial statements of the companies and Yamazawa and Yamamoto (1979a).

132 Industry Growth and International Trade
same time, Nissho, Iwai, and Ataka, which were originally steel and machinery
trading companies, became more diversified.
Diversification of Transactions
Diversification of product and geographic markets and types of transac­
tion characterized the strategy of the postwar GTCs and mirrored the general
move towards diversification of the economy. From 39 percent of the total
exports of Japan's GTCs in 1960, exports of heavy industrial goods such as
metals, machinery, and chemicals expanded to make up 74 percent of the to­
tal exports of GTCs in 1970. Although this pace of change in export struc­
ture was common to every GTC, it was especially remarkable for companies
that had begun as textile trading companies. For example, the share of heavy
industrial goods exported by the Kansai Gomen increased from a range of
18 to 33 percent in 1960 to a range of 54 to 76 percent in 1970; textiles fell
from 50 to 66 percent to 15 to 31 percent in the same period. Such a sharp
decline in the textile share of total exports is comparable to what occurred
at Mitsui in the prewar period.
The rate of increase for total exports of Mitsui, Mitsubishi, Marubeni,
Sumitomo, and Nissho-Iwai is close to or above the average rate of increase
of the ten GTCs (see Table 6.7). This indicates that the greater the scale of
the company's initial export level, the greater the rate of increase for exports
in the 1960s. The rate of increase for heavy industrial goods is much higher
than the rate of increase for all commodities. As Table 6.7 shows, the share
of chemical products, metals, and machinery are 10.6, 8.2, and 11.5 times
higher, respectively, in 1970 than in 1960. Although the ten GTCs exported
over three-fourths of Japan's total metal exports in 1970, their share of
machinery is much lower (about two-fifths) because of the prevalance of
manufacturers to directly export electrical machinery and motor vehicles. As
the firms sought to expand into new markets, the number of overseas branches
increased 150 percent between 1961 and 1970, primarily in Europe, Africa,
and Australia. Mitsui and Sumitomo were the most aggressive in opening new
offices (see Table 6.8).
There was also diversification in the type of transaction. Offshore trade
(involving non-Japanese buyers and sellers) became more important, although
it comprised only 3 to 8 percent of the activity of individual GTCs in the
1960s. Exports from Japan provided 14 to 19 percent of total offshore trade
volume, while imports to Japan provided 16 to 26 percent.
Less than half of all the GTCs' volume involved foreign trade. Domestic
transactions accounted for 50 to 62 percent. The GTCs were thus in a good
position to hedge between domestic and foreign markets. Although exact
figures are unavailable, it is widely agreed that domestic transactions have

Trading Companies and Foreign Trade Expansion 133
the highest profit margin, followed by imports and exports. Offshore trading
is the least profitable because of high operating costs and greater risks.
Organizing Capability of GTCs
Expansion of heavy industries in the 1960s provided GTCs with ample op­
portunity to demonstrate their organizing capability:
• Increasing demand for energy and raw materials led GTCs to explore im­
port sources in such unfamiliar areas as Australia, Latin America, and Af­
rica, where they risked investment in new mines and related infrastructure.
• GTCs utilized their information networks to investigate American and Eu­
ropean suppliers of technology and equipment, and joined domestic
manufacturers in new import-substitution production.
• GTCs promoted joint ventures in import-substitution production in Asian
and Latin American developing countries, and the shift in Japanese ex­
port structure from final goods to intermediate goods and machinery. This
contributed significantly to the resumption and further expansion of trade
by Japanese GTCs in Southeast Asia. Equity investments by the GTCs were
aimed not so much at earning direct returns on the investments as obtain­
ing commitments from the new ventures to give the GTCs exclusive rights
to handle their trade.
• GTCs also participated as prime contractors or consultants in the export
of plant and equipment, and in aid programs.
Trading companies specializing in promoting products frequently merged
with like companies to expand business activity. In the late 1950s Japanese
steel firms actively invested in new manufacturing plants and equipment to
expand capacity. At the same time, trading companies took over the steel ex­
port business from domestic steel distributors through a series of mergers.
Although knowledgeable in the steel business and long affiliated with the large
steel mills, the domestic steel distributors were inexperienced in conducting
foreign transactions. Thus, to expand exports they sought partners with over­
seas experience and pursued the diversification strategy of the large textile
trading companies to meet their needs. In 1955 Marubeni absorbed Taka-
shimaya-Iida, which specialized in machines and steel, and in February I960,
Marubeni absorbed Daiichi Kozai. C. Itoh took over Morioka Company in
October 1961, and Toyo Menka merged with Nankai Company in October
1963.
The success in diversifying the activities of the GTCs was largely due to
the organizing capability of the GTCs' top and middle management. To achieve
this level of skilled personnel, the GTCs attracted competent university gradu-

Table 6.8
Overseas Branches of Japanese GTCs: 1961 and 1970
Company Year Asia North
America Latin
America Europe Middle
East Africa Oceania Total
Mitsui 1961 23 a 14 13 a 6b 3a 5a 3b 67 a
1970 39 18 26 25 8 14 10 140
Mitsubishi 1961 18 lO 3 10 5 5 4a 3a 55
1970 23 17 11 6 7 11 7 82
Marubeni 1961 21 8a 9 5a 4 5a 4 56 a
1970 27 15 10 14 6 9 6 87
C. ltoh 1961 15 9 12 5a 4 5a 3a 53
1970 17 13 12 14 6 8 8 78
Sumitomo 1961 16 7 4a 4a lb lb lb 34 a
1970 20 10 11 II 6 6 6 70
Nissho-Iwai 1961 c 25 9 9 10 4 5a 6 68
1970 21 12 5 14 6 9 8 75
Toyo Menka 1961 12 7 8 2b 3 lb 3 36 a
1970 17 10 9 8 3 4 4 55

Company Year Asia North
America Latin
America Europe Middle
East Africa Oceania Total
Nichimen 1961 23 12 6 4a 3a 6 2 56
1970 20 10 8 9 5 7 3 62
Kanematsu-Gosho 1961 d 20 9 3 4a 3 2 2b 43
1970 15 11 4 7 1 3 6 47
Ataka 1961 12 5a 2a 4 la 0 3a 27 a
1970 14 10 5 6 2 2 7 46
Total 1961 185 90 76 49a 31 a 34 a 30 3 495
1970 213 126 101 114 50 73 65 742
Noies:
a. The total of Nissho and Iwai, which were separate companies in 1961.
b. The total of Kanematsu and Gosho, which were separate companies in 1961.
c. The ratio of the number of overseas branches in 1970 to the number in 1961 is greater than 1.5, which is the ratio of total GTC branches in 1970
to the total in 1961.
d. The ratio of 1970 branches to 1961 branches is over 3.
Sources: Company financial statements.

136 Industry Growth and International Trade
ates and trained them on the job to be business experts. In addition, GTCs
frequently sent their staff, a major supply source for entrepreneurship in
Japan's private companies, to affiliated companies to perform functions that
otherwise could not be handled by the respective companies.
Direct Export by Manufacturers
Not all manufacturers in Japan left overseas marketing to the GTCs;
producers of electrical machinery and motor vehicles exported most of their
products directly or via their own subsidiary trading companies. By the late
1960s, the value of exports by Matsushita, Toyota, and Nissan was almost
at the same level as the major GTCs.
Some common reasons for manufacturers relying on their own sales ef­
forts rather than use the established channels of the GTCs were the technical
sophistication of the products and the timing of export expansion. For au­
tomobiles and electrical goods, for example, significant technical skills and
knowledge were required and after-sale service had to be provided. Thus, it
was not considered profitable for GTC branches to market these goods, es­
pecially since the volume was small at the start. Moreover, when exports be­
gan to expand in the early 1970s, the big manufacturers were prepared to do
their own marketing overseas. However, the technical skill requirement does
not completely explain the lack of involvement by GTCs in these products.
It is more likely that the GTCs simply failed to see the rapid growth potential
of these exports and allowed the manufacturers to rely on their own sales
efforts.
CONCLUSION
The existence of a number of profit-making opportunities and a shortage
of entrepreneurial and managerial skills (a typical characteristic of backward
economies) in Japan before the Second World War induced the evolution of
GTCs. But the GTC was not invented by the Japanese. Diversification of com­
modities and markets was a natural extension of trading company activities,
as was the case for Jardine, Matheson and Company and East Asiatic Com­
pany in East and Southeast Asia during the late nineteenth century. The
Japanese developed this institutional device so fully that Japan's industriali­
zation and foreign trade expansion were promoted considerably by the GTCs
despite the various handicaps they had to overcome when they started. In
this sense, the GTC in Japan exemplifies Gerschenkron's (1966) institutional
devices of late-starting countries to overcome barriers in their industrializa­
tion process.

Trading Companies and Foreign Trade Expansion 137
Such business historians as Nakagawa (1967) and Morikawa (1976) dispute
whether the GTC served as the core of zaibatsu, financially backed by a bank
affiliated with the company and the sending of personnel to new industrial
undertakings, as well as serving as a powerful marketing arm for the affili­
ated firms. Despite Nakagawa's and Morikawa's reservations, the argument
does, however, apply quite well to prewar Mitsui and Mitsubishi, the two major
firms of the period. But after the Second World War, control by zaibatsu fam­
ilies was eliminated. Today, both financial and personal relationships are a
less significant element in transactions between the GTCs and their affiliated
banks and manufacturers.

Part III
The Role of Government
in Industrial Growth

Seven
Japan's Industrialization
and Protection Policy
Before the Second World War
Although many free-trade advocates attribute Japan's industrial success to
its free trade policy at the beginning of its industrialization, Japan's heavy
industrialization was actually achieved under tariff and nontariff protection.
Japan's industrialization began under the low tariff protections agreed upon
with the Western industrialized nations in 1866. But Japanese tariffs gradu­
ally rose until the early 1930s when Japan promoted heavy industrialization
through import substitution. Because tariffs were a major policy instrument
of governments around the world before the Second World War, this chapter
analyzes Japan's tariff structure in the context of its trade policy during its
prewar industrialization.
HISTORY OF TARIFF PROTECTION
The tariff treaty with foreign powers in 1866 (Kaizei Yakusho) established
the initial conditions for the development of Japan's industry and trade. Un­
der the terms of the treaty, Japanese import and export duties were limited
to unilateral rates as low as 5 percent (ad valorem equivalent) on most com­
modities. Except for foodstuffs, grains, and coal, these duties were levied on
most raw materials used in industrial production. Due to inflation, the equiva­
lent ad valorem rate of these duties declined to between 2.0 and 2.5 percent
in the first two decades of the Meiji period. These conventional tariffs were
levied on 40 percent of all dutiable imports and were maintained until com­
plete tariff autonomy was achieved in 1911.
After continued efforts to revise the tariff treaty of 1866, Japan partially
restored tariff autonomy in 1899, and the Tariff Law that was enacted in March

142 Role of Government in Industrial Growth
1897 took effect in January 1899. Subsequently, general revisions of the Tariff
Law were made in 1906, 1911, 1926, and 1932, and partial revisions were made
almost every year. Meanwhile, the tariff treaty with the United Kingdom and
other nations in 1866 continued to restrict tariffs on the products from these
countries to levels as low as 5 to 15 percent.
In addition to the Tariff Law and conventional tariffs, a third category—
special tariffs—was created for all imports including rice and other essential
foodstuffs. Under the First and Second Emergency Special Tariffs laws that
were enacted in 1904 and 1905 to finance the Russo-Japanese War, surcharges
from 5 to 20 percent were levied on most imports except those under conven­
tional tariffs. 1 Although intended to be temporary at the outset, these spe­
cial tariffs were maintained for some time and were eventually incorporated
into the tariff schedule. For example, the 100 percent duty on luxury mer­
chandise, consisting of 147 items in 1924, was initially levied primarily for
balance-of-payment purposes; but after meeting its intended objective the duty
was incorporated into the tariff schedule in 1926 and remained in effect until
the Second World War.
Industrial Protection
Raising revenue was the main purpose of tariff imposition in Japan before
the First World War. During the early Meiji period (1868-1912), Japan im­
posed a duty of 5 percent on most exports. The revenue from these export
duties was on average two-thirds the revenues obtained from import duties.
Because the senselessness of export duties was well known, a movement for
their elimination spread nationwide in the early 1890s. Duties on such major
exports as raw silk and tea were, however, maintained until 1899, when, un­
der partial tariff autonomy, import duties were raised to yield a greater por­
tion of tariff revenue. The main source of government revenue in the Meiji
period, however, was land taxes, and tariff revenue constituted at most be­
tween 5 and 10 percent of total government revenue.
As industry grew, however, industrial protection became the main reason
for tariff imposition. Beginning in the late 1890s, tariffs were used selectively
to protect domestic industries. Figure 7.1 shows changes in the average tariff
rates, which are defined as the ratio of total tariff revenues to either the value
of total imports or the value of dutiable imports.
Average tariffs on manufactures followed a pattern similar to that of dutia­
ble imports (see Figure 7.1); these are analyzed in greater detail in the follow­
ing sections. Raw material imports were initially subject to the same tariffs
as manufactures but were exempted from duty by the late 1890s. For exam­
ple, ginned cotton was exempted from import tariffs in 1896 and iron ore was
exempted in 1901. Furthermore, the number of duty-free raw material items

Industrialization and Protection Policy 143
Figure 7.1
Average Tariffs of Japan: 1868-1980

1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980
Year
Notes: Solid line = average tariff rates calculated for ,total imports; dotted line = average
tariff rates calculated for dutiable imports.
a. Tariff act, recovery of tariff autonomy (1899)
b. Tariff revision (1905)
c. General tariff revision (1911)
d. General tariff revision (1926),
e. Surcharge of 35 percent on specific duty items (1932)
f. . General revision (1951)
g. General revision and shift to BTN (1961)
h. Kennedy Round reduction started (1968)
i. Unilateral reduction by 20 percent (1972)
j. Tokyo Round reduction started (1980)
Sources: Yamazawa and Yamamoto (1979a: table 22) for the years 1868-1970; Ministry of
Finance (1982) for the years 1970-80.
increased from 49 to 89 in the general revision of 1906. This, in effect, was
tariff differentiation, which implies a deliberate policy of industrial protec­
tion. If policy makers had been seeking complete self-sufficiency in the
Japanese economy, they would have levied high tariffs uniformly on all im­
ports. But protection of a given industry can only be undertaken at a cost
to all other manufacturing and nonmanufacturing industries.
Japan's import dependence on raw materials began with the elimination
of import duties on raw cotton. Initially, as part of its import substitution
policy, the Meiji government encouraged domestic cultivation of cotton, but

144 Role of Government in Industrial Growth
Japanese raw cotton was unsuitable for spinning fine quality yarn. In the early
1890s, while domestic production of cotton textiles was growing, and because
of the lack of domestic raw cotton of fine quality to produce cotton yarn,
imports of cotton yarn were increasing. Thus, to achieve import substitution
of cotton yarn, imports of fine quality raw cotton were necessary and import
duties on raw cotton were eliminated. The elimination of the import duties
resulted in an increase in domestic spinning of imported, higher-quality raw
cotton. At the same time, export duties on cotton yarn were also eliminated,
although Japan's export market at the time was small. In a few years, however,
exports of the higher-quality cotton yarn exceeded imports. Domestic culti­
vation of raw cotton for both the export market and domestic consumption
decreased rapidly after 1896.
Start of Agricultural Protection
Tariffs on agricultural products were originally intended to supplement
emergency surcharges on the land tax during the Russo-Japanese War. Tariffs
on cereals set the precedent for agricultural protection in Japan. Rice, which
had been exempted from duty, became subject to a duty of 15 percent in 1905.
In the same year, duties on wheat and barley were raised from 5 to 15 per­
cent. These agricultural tariffs remained in effect even after the emergency
surcharges were eliminated, and were incorporated into the tariff schedule
in 1906.
In 1911, the duty on wheat was raised further to 20 percent. The duty on
rice, on the other hand, provoked a public controversy between landowners
and manufacturers over agricultural protection. The controversy was finally
resolved in the Japanese Diet in 1913 when compromise tariffs were agreed
upon. Under the agreement, a specific duty of one yen per 60 kilograms (the
ad valorem equivalent was 23% of average import prices in 1910-12) was levied
on imports of rice from outside the Japanese empire, but imports of rice from
Taiwan and Korea were exempted. Unlike the effect of the British Corn Laws
in the United Kingdom, which led to the expansion of grain imports, the tariff
policy for rice in Japan resulted in self-sufficiency in rice throughout the
Japanese empire. Tariffs on refined sugar continued to be as low as 10 per­
cent under the tariff treaty of 1866 until 1911, when they were raised to an
ad valorem equivalent of between 50 and 60 percent. Since these higher tariffs
were imposed on sugar imports from outside the empire, they tended to en­
courage colonial agriculture in Taiwan and Korea, which expanded produc­
tion of sugar and rice for export to Japan (see Table 7.1).
It should be emphasized that Japan's tariff policy did not remain unaffected
by the alternating worldwide waves of free trade and protectionism (Haberler
1964). Initially, free trade was forced on Japan during the free-trade move-

Table-7.1
Rice and Sugar Imports of Japan Proper: 1898-1933 (Vo)
Product and origin 1898 1903 1908 1913 1918 1924 1928 1933
Rice and paddy
from outside the empire 97;7 91.4 69.1 63.3. -52.4 25.1 16.9 5.0
from Korea 3 14.9 33.1 57.8 56.4 66.7
from Taiwan 2.3 8.6 30.9 21.8 14.5 17.1 26.7 28.3
Sugar b
from outside the empire 94.7 87.2 67.5. 70.4 38.0 34.1 34.7 9.7
from Taiwan 5.3 12.8 32.5 29.6 62.0 .65.9 65.3 90.3
Notes:
a. Korea was not part of the Japanese empire during the first three periods and is therefore treated separately.
b. Imports of sugar from Korea were negligible.
Source: Toyo Keizai Shinposha (1935).

146 Role of Government in Industrial Growth
ment that was initiated by the 1860 Cobden-Chevalier Treaty between Great
Britain and France. But Japan's initial period of free trade was followed by
a period of protectionism that extended from the late 1870s until the First
World War. During this protectionist period, Japan resumed tariff autonomy
and began to raise tariff barriers. After the First World War, protectionism
was further strengthened, escalating to autarky within each trade bloc in the
worldwide tariff war of the 1930s. By this time Japan had the highest level
of tariff protection in the world.
ANALYSIS OF INSTITUTIONAL
INDUSTRIAL TARIFF PROTECTION
In the preceding section it was suggested that Japan's imposition of tariffs
became selective, and that the main objective was to protect industry. To fur­
ther support this assertion, more evidence is presented in the following sec­
tions. The structure of tariff protection for manufactures during the period
1890-1940 is analyzed in detail, and special attention is given to changes in
the tariff structure in response to the progress of industrialization. Two com­
plementary approaches are used in the analysis. The first examines the prin­
ciples that justify tariff imposition and the structure of the nominal tariff
schedule. The second compiles tariff data on individual principal commodi­
ties and quantitatively analyzes the changes in tariff protection for manufac­
tures over time.
Principles of Tariff Policy
Evidence of industrial protection by selective tariffs can be found in tariff
reports that were made at each general tariff revision (Ministry of Finance
1938). The tariff report for the general tariff revision in 1911 proposed that
tariffs be levied principally to raise revenue and that industrial protection be
given secondary consideration. But this proposition merely indicates that the
government did not want to appear protectionist. In fact, industrial protec­
tion dominated the structure of the tariff schedule. Even the two special
tariffs—one that allowed emergency surcharges in 1904-05 and one that im­
posed luxury duties in 1924—eventually ended up protecting domestic indus­
tries even though they were initially motivated by the necessity to increase
revenue. The emergency surcharges began the move toward agricultural pro­
tection, and the luxury duties were incorporated into the general tariff schedule
as prohibitive tariffs on imported consumer goods.
After the First World War and during the early 1920s, special tariffs were
occasionally levied to protect particular industries. Manufacturers of iron and
steel products, and synthetic dyestuffs and other chemical products had been

Industrialization and Protection Policy 147
established or had developed rapidly during the First World War. After the
war, enthusiasm to industrialize was still strong, and the strong desire for au­
tarky was rooted in the belief that all existing manufacturing industries should
be maintained at any cost. Thus, when these industries were confronted with
reorganized foreign competition after the war, the domestic producers de­
manded and received tariff protection.
While explaining the general tariff revision of 1926 in a speech to the
Japanese Diet, Prime Minister Osachi Hamaguchi explicitly proposed that
Japan's tariff policy be based on the concept of industrial protection. His
general criteria for tariff imposition are as follows: 2
• Raw materials for industrial production that are either not produced domes­
tically or only produced in small quantities should be exempt from duties
or have low duties.
• Important industries not yet fully developed but with prospects for fur­
ther development should be given enough protection to enable them to
compete with foreign industries.
• Tariffs for important industries that are already fully developed and that
compete with foreign products should be lowered or kept at the present level.
• Daily necessities should either be exempted from tariffs or have the lowest
possible tariffs.
• Luxury consumer goods should have the highest tariffs in order to dis­
courage their consumption domestically.
The criteria were reasonable and could easily have been expanded to be even
more protectionist. Further protection was in fact realized with the adoption
of a series of tariff revisions in the 1920s and early 1930s. Newly developed
heavy manufactures were given higher tariff protection under the general tariff
revision of 1926. This was followed by tariffs on sugar and starch in 1927,
lumber in 1929, artificial silk in 1931, and tariff increases on pig iron and
other heavy manufactures in 1932. Another general tariff revision that was
planned in 1936 to add protection to new industries resulted in increased pro­
tection for several manufacturing sectors including automobile production
and petroleum refining in 1937.
Strong objections to this increasing tendency toward protectionism were
raised by free-trade advocates. They argued against fully embracing protec­
tionism and contended that protection should be confined to a few promis­
ing industries whose progress could be regularly reviewed. They also demanded
the elimination of tariff protection for fully developed industries. In most
cases their advice went unheeded. One example of this is the movement to
eliminate the duty on cotton yarn. Cotton yarn was subject to a duty of 5
percent (although it was not being imported at the time). In 1925 cotton

148 Role of Government in Industrial Growth
weavers and hosiers, which were typically small and medium-sized firms,
pleaded publicly and in the Diet for the elimination of the duty on cotton
yarn to increase their competitiveness with large firms. The 5 percent duty
on Chinese cotton yarn, which because of the depreciation of silver had be­
come more competitive, was an added cost to small and medium-sized firms
trying to compete with large firms. Large firms could offset the higher cost
resulting from the duty by dealing in large volumes. But several big cotton-
spinning firms were able to exert their influence in maintaining the tariff un­
til 1930, when it was finally lowered to 3.3 percent.
Disaggregation in the General Tariff Schedule
The degree of detail in the commodity classification of the general tariff
schedule provides more evidence for tariff differentiation. The tariff sched­
ule that was in effect in 1899 included only 532 classes. However, with the
general tariff revision of 1906, the tariff schedule was broken down into a
standard commodity classification consisting of 19 categories, 538 classes,
and 819 items. In 1911, the tariff schedule was disaggregated further into 1,599
items and in 1926 to 1,699 items. This system was used until the Brussels Tariff
Nomenclature (BTN) system was adopted in 1961. The disaggregation of the
commodity classification reflected not only the emergence of new industries
and commodities in foreign trade but also the need for deliberate protection
of domestic industries.
Escalated Tariff Structure
The first tariff schedule of 1899 set up nine classes of tariff rates between
0 and 40 percent at intervals of 5 percent. Rates were assigned to the various
stages of processing or type of good. Individual commodities were then as­
signed one of these rates. For example, a rate between 0 and 5 percent was
levied on raw materials, a rate of 10 percent was levied on semimanufactures,
a rate between 15 to 20 percent was levied on finished manufactures, and a
rate of more than 25 percent was levied on luxury goods.
The differential between tariffs increased with the degree of processing
(tariff escalation) and the passage of time. In the tariff schedule of 1906, raw
materials were assigned tariffs from 0 to 5 percent, finished manufactures
had tariffs from 30 to 40 percent, and luxury goods were subject to tariffs
of 50 to 60 percent. In the tariff schedules of 1911 and 1926, commodities
were differentiated within the same escalated structure by such factors as the
need for protection for import-competing production, the future prospects
for domestic production, and the effects on export competitiveness (in the
case of materials used in export industries). The effective rates of protection

Industrialization and Protection Policy 149
produced by this escalated structure differed from the nominal tariffs; in fact,
nominal tariffs gave higher effective protection to manufacturing than im­
plied by their levels because of this escalated structure of protection.
. Ad Valorem Versus Specific Duties
The Tariff Law of 1899 mandated the adoption of ad valorem duties but
permitted, for administrative convenience, specific duty rates (that is, a set
amount per piece or per ton) using average import prices for the preceding
six months. The Tariff Law of 1906 increased the role of specific rates by
commodity, and the number of imports subject to specific duties was ex­
panded. The ad valorem equivalent rates of the specific duties declined prior
to 1920 when import prices were rising, but they were later revised upward
several times. For example, during the First World War, rapidly rising import
prices had lowered ad valorem equivalents to levels as low as 1-2 percent. In
1921, however, tariffs were raised back to their former levels by changing many
specific duties to ad valorem rates (but most of these were changed back to
specific duties in the general revision of 1926).
In the 1920s, the ad valorem equivalent of specific duties increased as im­
port prices decreased, but specific duties were seldom revised downward. In
June 1932, specific duties were raised uniformly by 35 percent to adjust to
the devaluation of the yen. The adjustment was needed to maintain the ad
valorem equivalent of specific duties vis-a-vis the higher prices of imports
resulting from the decrease in the value of the yen relative to the dollar.
However, since prices for many imports were raised by less than 35 percent
and since currency depreciation by itself produces the same effect as a uni­
form schedule of import tariffs and export subsidies, this uniform adjust­
ment of specific duties effectively doubled protection for domestic producers. 3
Other Tariffs
Although tariffs had increased industrial protection by 1906, supplemen­
tary tariffs to protect domestic producers against "unfair foreign competi­
tion" were also established that year under new tariff laws. Specifically, three
special-purpose tariffs—the countervailing, retaliatory, and antidumping
duties—were created, although they were seldom enforced before the 1930s.
In keeping with the worldwide trend toward industrial protection in the 1930s,
the special-purpose tariffs were strengthened and enacted into the Trade Pro­
tection Law of 1934. Under the new law, a 50 percent surcharge was imposed
on all imports from Canada and Australia in retaliation for the discrimina­
tory tariffs those nations imposed on Japanese goods. However, after six
months, new trade agreements between Japan and the two nations eliminated

150 Role of Government in Industrial Growth
these tariffs but established exemption and payback rules for materials used
in export industries.
Another set of tariff laws was adopted in 1900 under the Temporary Storage
Yard Law. The laws exempted imported materials from duties provided the
imported materials were simply processed for export in special areas known
as temporary storage yards, which, in effect, were the first free port (or export-
processing) zones. In order to further promote exports, domestically produced
materials were permitted to be similarly processed in 1912. By the early 1920s,
the concept of free ports was openly discussed in business circles and the de­
bate moved to the Diet in 1925. In 1927, the government responded by enact­
ing the Tariff Factory Law, which changed the temporary storage yards into
"tariff factories." Under the new law, improvements such as the simplifica­
tion of procedures and extension of the storage period from six months to
one year were made. Owing to these changes, by the late 1920s, the number
of tariff factories had increased.
INDUSTRIAL PROTECTION BY TARIFF:
A QUANTITATIVE ANALYSIS
Thus far, the history of tariffs in Japan has been discussed in terms of the
average tariffs on total imports or on large commodity groups such as
manufactures, foodstuffs, and raw materials. But in order to establish a pat­
tern of industrial protection through the imposition of tariffs, analyzing the
average level of tariffs is not as important as analyzing the upward or down­
ward deviations from the average—that is, analyzing the structure of individual
tariffs. To this end, data on individual tariffs have been compiled for 61 prin­
cipal commodities at five-year intervals from 1893 to 1938. The 61 commodi­
ties under consideration do not correspond to the original commodity
classification in Ministry of Finance statistics (1882-1943). The 61 commodi­
ties make up 60-70 percent of the value of total imports. Based on this data,
a quantitative analysis of the tariff structure is presented in the following
sections.
Tariffs on Individual Commodities
The value of every commodity imported into each of Japan's main ports
from outside the Japanese empire was recorded along with the tariff revenues
collected on the commodity. The tariff rate was estimated by dividing tariff
revenues by the value of the imported commodity for each of the 61 individual
commodities. 4 The commodities were then divided into six to eight groups
by industry and by economic use (or stages of processing), and the average

Industrialization and Protection Policy 151
tariff rates of individual groups are compared to identify the structure of tariff
protection.
The tariff rates collected were used in the analysis rather than the nominal
tariff rates listed in the tariff schedule. While this ratio tends to underesti­
mate the restrictive effect of tariffs on imports, 5 it is used here nevertheless
because of the following difficulties pertaining to nominal tariffs:
• In addition to general tariff revisions, nominal tariffs were revised almost
every year (including special tariffs) and were subject to conventional tariffs
or complete or partial exemptions. Thus, the actual nominal tariff rate
on each of the 61 commodities throughout the period 1893-1936 would
be difficult to determine.
• Many commodities were subject to specific duties. Aggregating these com­
modities into larger categories requires calculations of the ad valorem
equivalents of the specific tariffs. Determining reliable levels of import
prices to derive ad valorem equivalents is beyond the scope of this analysis.
The tariff rates that were calculated for the groups that are classified by
industry and by economic use are summarized in Tables 7.2 and 7.3. The tariff
rates are calculated approximately every five years: 1893, 1898, 1903, 1908,
1913, 1918, 1924, 1928, 1933, and 1938. Rates for a few of the selected years
were affected by several events. For example, because of the Great Earthquake
of 1923, original data on goods imported through the Port of Yokohama in
1922 and 1923 are not complete; the year 1924 was therefore selected instead
of 1923. In addition, the rates for 1924 were lower since daily necessities were
exempt from import duty in March of that year. In 1937, the publication of
statistics on important imported commodities was discontinued for security
purposes, and in 1938, data were available for only 30 percent of the com­
modities covered in the sample.
Table 7.2 and Figure 7.1 summarize changes in average tariffs (simple and
weighted arithmetic means, respectively) for the 61 commodities. The simple
mean (last row, Table 7.2) increased after 1898 and, except for the decreases
in 1918 and 1924, continued to increase until the Second World War. The arith­
metic mean weighted by import values of individual commodities (solid line,
Figure 7.1), however, increased until 1913 and tended to decline in the 1920s
and 1930s. This occurred in part because the share of duty-free or low-duty
items (such as raw materials) increased, but it also reflects underestimation
of the restricting effect of higher tariffs. The standard deviation of tariffs
of the 61 commodities increased from 1900 until the First World War, and,
despite the decreases in 1918 and 1924, continued to increase until the Sec­
ond World War. This evidence supports the discussion above on the selective
imposition of tariffs.

Table 7.2
Simple Average Tariffs on Individual Commodity Groups by Industry (%)
Classification by industry 1893 1898 1903 1908 1913 1918 1924 1928 1933 1938
A Agricultural products 2.52 2.49 8.57 20.61 19.99. 14.39 8.63 14.37 23.08 24.24
C Raw materials 3.96 2.79 5.66 8.73. 6.42 3.34 1.77 4.01 7.63 4.16
S Primary products (A + C) 3.43 2.68 6.63 12.69 10.94 7.02 4.06 7.46 12.78 12.19
B Manufactured foodstuffs 3.22 3.36 12.61 35.89 42.67 24.44 19.02 47.37 50.55 58.31
D Textile manufactures 3.20 2.84 12.43 14.86 20.68 9.55 11.85 26.31 25.54 39.17
E Other light manufactures 4.71 4.62 11.86 16.21 21.11 11.71 10.33 20.97 19.15 18.00
Q Light manufactures (B+D + E) 3.78 3.69 12.24 20.87 26.13 13.99 13.07 29.35 29.13 34.96
F Chemical products 4.63 4.75 6.29 6.51 13.50 4.13 17.15 32.63 28.43 47.15
G Metals and metal products 4.17 3.83 9.83 12.06 15.31 6.30 3.74 17.19 17.99 21.45
H Machinery 4.20 4.24 12.59 24.33 25.45 18.25 17.34 22.38 26.94 19.46
R Heavy manufactures (F + G + H) 4.32 4.24 9.32 13.33 17.66 9.06 12.34 23.47 23.89 31.75
Total 3.91 3.71 9.88 16.19 19.81 10.68 10.93 22.60 23.76 29.16
Note: Percentages represent simple arithmetic means of individual tariffs belonging to each group. Commodities not imported are excluded from calculation-
Source.- Yamazawa and Yamamoto (1979a).

Table 7.3
Simple Average Tariffs on Individual Commodity Groups by Economic Use (°7o)
Classification by economic use 1893 1898 1903 1908 1913 1918 1924 1928 1933 1938
A + B Foodstuffs 2.91 2.97 10.81 29.78 33.60 20.42 14.86 34.17 39.56 44.69
C Raw materials 3.96 2.79 5.66 8.73 6.42 3.34 1.77 4.01 7.63 4.16
1 Intermediate goods 1-L a 4.04 3.45 6.60 4.07 12.25 5.74 6.80 13.60 16.48 9.54
J Intermediate goods 1-H a 4.84 4.44 3.72 '3.84 6.41 2.84 13.92 14.68 15.02 13.99
O Intermediate goods 1 (1 + J) 4.49 3.95 5.16 3.96 9.60 4.42 10.04 14.09 15.82 10.65
K Intermediate goods 2-L a 4.24 4.17 10.59 8.10 18.11 7.67 6.14 15.27 11.05 6.75
L Intermediate goods 2-H a 4.04 4.03 8.88 9.42 16.21 3.44 7.08 19.88 17.29 15.97
P Intermediate goods 2 (K + L) 4.14 4.09 9.56 8.89 16.97 5.13 6.67 17.90 14.61 10.85
M Capital goods 3.67 3.86 8.75 13.75 17.33 6.98 10.60 14.27 16.89 7.27
N Consumer goods 4.04 4.12 17.31 31.21 30.12 20.55 20.43 41.23 41.59 63.10
Total 3.91 3.71 9.88 16.19 19.81 10.68 10.93 22.60 23.76 29.16
Notes: Percentages represent simple arithmetic means of individual tariffs belonging to each group. Commodities not imported are excluded from
calculation.
a. Intermediate goods are divided into four groups, cross-classified by stages (1 and 2) and factor intensity (light (L) and heavy (H) manufactures).
Source: Yamazawa.and Yamamoto (1979a).

154 Role of Government in Industrial Growth
Escalated Tariff Structure
The tariff structure continued to be differentiated by degree of processing
throughout the period. The sample commodities are classified into groups
by industry and by economic use, thus facilitating comparisons of changes
in average tariffs (see Tables 7.2 and 7.3). The average tariffs for four broad
groups—agricultural products, raw materials, light manufactures, and heavy
manufactures—remained at about the same level until 1898, but tended to
diverge after 1903. The infant industry argument leads to the hypothesis that
tariffs on heavy manufactures exceeded those of light manufactures at the
early stage of heavy industrialization; but this was not the case. The tariff
rates for both light and heavy manufactures followed a pattern similar to that
of the total average tariff (see Table 7.2), but the rate for light manufactures
exceeded that for heavy manufactures from 1903 onwards. The rate for agricul­
tural products increased and kept pace with that of light manufactures until
1913 when its rate of increase slowed, and eventually fell below that of heavy
manufactures after the 1920s. The rate for raw materials fluctuated between
2 and 8 percent from 1893 through 1938 (the fluctuation of average rates of
raw materials seems to be the result of changes in tariffs on fuels).
The disaggregation of light and heavy manufactures shows that the tariff
rates on manufactured foodstuffs increased earlier and remained at a level
higher than those of any other group. The tariff rate for textiles kept pace
with that of other light manufactures until 1924 but exceeded the tariff rate
on other light manufactures after 1928. The tariff rates for heavy metals and
chemicals lagged behind that of the other groups, but after 1924 they rose
rapidly and exceeded those for textiles. Average tariffs on machinery were
raised earlier and remained between 20 and 25 percent, but they were sur­
passed by the rates for metals in the 1930s. From the data, no clear principle
of tariff differentiation is discernible.
When classified by economic use, the average tariffs of the groups show
an escalated structure of tariffs and a pattern of tariff differentiation that
is clearer than when the groups are classified by industry (see Table 7.3 and
Figure 7.2). Average tariffs for both consumer goods and foodstuffs increased
earlier than any other group and remained at a level higher than any other
group throughout the period. Intermediate goods 1 and 2 and capital goods
had average tariffs that showed no rapid increases and remained at about 15
percent. Raw materials had the lowest average tariffs—between 2 and 8 percent.
As Table 7.2 shows, there is no relationship between tariff rates for light
and heavy manufactures. The rapid increase in average tariffs on chemicals
and textiles after the late 1920s was largely the result of the increase in tariffs
on individual commodities in the consumer goods category within the two
groups. This leads to the conclusion that classification by economic use (or

[
Figure 7.2
Average Tariffs on Commodity Groups Classified by Economic Use
60
50
20
10 1 I I T I
— -— Consumer goods "' T | 1
Foodstuffs
Intermediate 1
L
— • — Capital goods 1
1
— — Raw materials 1
1
1
i
-1
1 _
/
/
1
1
1 '/
1 /
/ /
' S
t /
1 /
1 /
1 f
1 I
if ^ \ 1 I
1 I
1 I 1 j
'/ \ 1 j
I I
' / A 1 I
/ / \\ 1 I
/ / \ 1 I / / V
/ / > / /
/ /
•7 – A / / / / \
/ LSJ , \ •''/J ~ N
1 * ^ * ^» •. \
egg^ ' N;
i i i i i i t i
1893 1898 1903 1908 1913 1918 1924 1928 1933 1938
Year
Source: Calculated from Yamazawa and Yamamoto (1979a: table 21).

156 Role of Government in Industrial Growth
stages of processing) of individual commodities was the main criterion relied
upon to determine tariff levels.
OTHER NONTARIFF
INDUSTRIAL PROTECTION POLICIES
While the Japanese government used tariffs as a principal policy instru­
ment for industrial protection before the Second World War, especially after
tariff autonomy had been fully restored, it also resorted to the use of other
policy instruments such as quantitative import restrictions, production sub­
sidies, and other direct and indirect government assistance to affect the course
of industrial development. Although it is difficult to measure the impact these
instruments had on industrial protection, they are discussed briefly in the sec­
tions that follow.
Quantitative Restrictions on Imports
The import-approval system, which was first introduced for rice with the
enactment of the Rice Law in 1921, empowered the government to intervene
in the market to adjust the supply of rice. The system was extended in 1924
to include synthetic dyestuffs to confront German dumping, and in 1931 to
ammonium sulfate after the breakdown of the international ammonium sul­
fate cartel. But it was not until 1937 that the Japanese government resorted
to general quantitative restrictions on imports. 6
In January 1937, Japan's quantitative restrictions began with the estab­
lishment of an exchange control system, which was introduced with the aim
of stabilizing yen exchange rates. To this end, trading licenses were issued to
importers based on their past performance. But in September 1937 the Ad­
justment Law introduced a stricter import quota system that was devised to
improve the persistent deficit in the balance of payments and to secure sup­
plies of commodities for military purposes. Except for a few industries, such
as automobiles and machine tools, quantitative import restrictions had little
effect on industrial development before the Second World War.
Production Subsidies
Production subsidies were used by the Japanese government to promote
the modernization of the Japanese economy, especially during the 1880s, the
1890s, and the interwar period. This is reflected in the ratio of industry sub­
sidies to government expenditure, which, for the period 1880-90, was between
4 and 5 percent. This ratio declined to 1 to 2 percent between 1900 and the
First World War and increased again to 8 percent during the interwar period.

Industrialization and Protection Policy 157
The distribution of subsidies changed over time. In the 1880s, construc­
tion firms were the main recipients of subsidies. Between 1890 and 1910, sub­
sidies went mainly to transportation and communications firms, and after
the late 1920s subsidies were used mainly to support agriculture. Production
subsidies for mining and manufacturing were relatively small (ranging from
3.6 to 5.4% during the period 1925-38). The ratio of subsidies to manufac­
turing income was less than 1 percent from 1880 to 1938. However, it is not
so much the quantitative magnitude but the qualitative impact that should
be considered, since subsidies were directed at building infrastructure and at
fostering strategic industries such as shipbuilding (under the Shipbuilding En­
couragement Act of 1896), steel, and automobile production.
The exemption from corporate and other taxes had the same effect as a
production subsidy. For example, although the impact was not incorporated
in the calculation of subsidies mentioned above, steel-producing firms with
more than a predetermined production capacity were exempted from all
domestic taxes under the Steel-Manufacturing Encouragement Act of 1917.
Direct Government Aid
Direct government aid to initiate and develop new industries can also be
viewed as a form of protection. One example is government financing, which
lessened the risks entailed in the start-up of new industries. In the initial stages
of the industry's development, the government established experimental plants
for the silk, cotton, and woolen industries. These plants were then turned
over to private firms after it was apparent that they could be operated as in­
dependent, viable entities. The early development of the chemicals, metals,
and machinery industries also benefited greatly from the activities of
government-owned foundries, research institutes, and national universities.
These governmental institutions played an important role in introducing new
techniques, and in training and supplying engineers for employment in pri­
vate firms.
For two decades, private firms had failed to establish a viable steel produc­
tion industry. It was only after the government exercised strong leadership
in developing steel production did the integrated production of iron and steel
succeed for the first time in Japan. The government's efforts were directed
toward the creation and development of the Yawata Iron Works. During the
start-up period, the losses of Yawata and the subsequent expansion of invest­
ment in the firm were financed by the government. After nine years of oper­
ation, Yawata began to show a profit. Yawata played a central role in the
expansion of steel production in Japan throughout the period before the Sec­
ond World War (see Chapter 5).

158 Role of Government in Industrial Growth
Government Purchases
The pace of an industry's development depends largely on the scale and
growth of its potential domestic market. In the process of import substitu­
tion, which is achieved by lowering the costs and improving the quality of
domestically produced goods, domestic producers have to compete with for­
eign producers for a share of the domestic market. If, however, part of domes­
tic demand is reserved for domestically produced goods regardless of the cost
and quality differences between domestically produced and imported goods,
the pace of import substitution will be considerably accelerated.
Recognizing this, the Kokusanshoreikai (the Association to Encourage
Domestic Manufacturing) initiated a semiofficial "buy-Japanese" movement
at the end of 1914. The aim of this movement was not only to work toward
the substitution of domestic for imported goods but also to increase the use
of domestic goods. Led by Baron Eiichi Shibusawa, one of the most eminent
business leaders in the Meiji and Taisho periods (1868-1926), the association
"pledged itself to conduct the following program: survey home industries;
hold fairs and exhibitions to display domestic products; give public lectures;
answer inquiries on home manufactures; collect and display samples and cata­
logues; publish a review, etc." (Kobayashi 1930:238-39). The government ap­
propriated 3,000 yen for the association's maintenance in 1914 and increased
the appropriation to 5,000 yen a year from 1915 onwards.
The imported goods that were targetted in the buy-Japanese movement in­
cluded drugs and chemicals, iron and steel, machinery, glass, paper, and
woolen fabrics. These goods had been mostly imported in the past. With the
outbreak of the First World War in 1914, the volume of these imports decreased
and their prices rapidly rose. Domestic products in these commodity groups
were, however, widely perceived to be inferior to imports, and foreign products
such as heavy electrical machinery were considered a necessity in government
factory work (Takeuchi 1972).
In the 1920s, the buy-Japanese movement was given further impetus by
domestic industries seeking further protection against renewed competition
from foreign competitors and by the need for improvement in the balance
of payments. The British consulate in Tokyo reported on the buy-Japanese
movement as follows:
Though it is not strictly within the category of legislation, mention should
be made of an extremely strong official campaign to encourage the use
of domestic products. This has ramifications in all quarters, such as, for
example, the issue of instructions by the Railway department laying down
what domestic goods must be used and what imported goods may be used.
It is understood that certain government departments when calling for
tenders now specify in many cases articles of Japanese manufactures ir­
respective of quality. 7

Industrialization and Protection Policy 159
It is difficult to make a quantitative appraisal of the effect of the buy-Japanese
movement on import substitution. Before the First World War, a large part
of domestic demand for such manufactures as steel, ships, and automobiles
came from the Japanese government, which needed these items in areas such
as railroad construction and the military. For example, in 1903 two-thirds of
Yawata's sales of steel products were to the government. In 1921 only one-
half of Yawata's sales were to the government, and by the late 1920s the share
of Yawata's sales to the government had declined to less than one-fourth. But
such data are misleading since domestic demand would be much higher if
the indirect demand by private producers of machinery and metal manufac­
tures that was induced by government demand is included. There is no doubt
that the government pursued a policy of preferential purchasing of domestic
goods, weighing the differences in price and quality between domestically
produced and imported goods.
At the same time, domestic producers tried to match imports both in qual­
ity and cost. In the mid-1920s, in industries in which a wide technological
gap existed between domestic and foreign goods (such as in the electrical
machinery industry), major Japanese manufacturers rushed to introduce ad­
vanced technology into their manufacturing processes under license contracts
from European and American manufacturers.
Export Promotion
As European competition revived in the world market after the First World
War, Japan's exports fell and the government was prompted to make export
promotion a more important part of trade policy. Various export promotion
measures were pursued and enacted in the 1920s. These measures, which in­
cluded the establishment of a quality control system for traditional Japanese
exports such as silk, cotton textiles, and celluloid, were strengthened in the
1930s. In addition, new markets for these exports in Latin America, the Mid­
dle East, and Australia were encouraged, sometimes through the use of govern­
ment guarantees that supported bank acceptance of export bills.
The government's policies toward its colonial possessions allowed for the
development of markets for the export of emerging manufactures such as met­
als, chemicals, and machinery Exports of these commodities expanded stead­
ily to Manchuria and Kwantung Province in the 1930s, but they were tied to
the growth of Japanese investment in these areas.
FOREIGN INVESTMENT POLICY
The discussion of foreign investment policy in this section is limited primar­
ily to commodity trade policy. However, since government policy toward direct

160 Role of Government in Industrial Growth
foreign investment (DFI) by foreign firms and toward imports played an im­
portant role in the introduction of a part of the heavy industrial sector from
abroad, this policy is also discussed.
Japan adopted three methods to establish new industries. The first was to
produce by imitation, that is, importing the necessary machinery and em­
ploying foreign technicians in the process. This method was principally
employed in light industrialization, a good example of which is the cotton-
spinning industry. The second method was to purchase patents and expertise
from firms in developed countries, which is similar to the first method in
that it also involved machinery and technicians being introduced from abroad.
This second method was used in a number of heavy industries, including the
iron and steel and ammonium sulfate industries. The third method was to
allow stock ownership of patented supplies, skills, and machinery. This method
allowed joint ventures to be established.
Direct Foreign Investment
From the 1900s onwards, especially before the First World War and in the
1920s, Japan was a host country for DFI. 1 As was shown in Chapter 2 (see
Table 2.2), the value of DFI was low, at 70-100 million yen from 1904 to 1913
and only 145 million yen from 1920 to 1931. Examination of its composition,
however, shows that its role in heavy industrialization in Japan was significant.
Two characteristics of DFI in Japan should be noted (see Table 7.4). First,
DFI in Japan was concentrated in heavy industry. Because of the industriali­
zation taking place during the period 1900-1940, domestic demand for heavy
industrial goods and imports grew. The gradual rise in the tariff rates for in­
dustrial goods that occurred following the recovery of partial tariff autono­
my in 1899 further induced growth of domestic heavy industries. In some
cases, DFI was used to jump the tariff barrier in these industries.
Second, DFI in Japan generally took the form of joint ventures (with the
exception of 100 percent foreign ownership in the automobile assembly and
rubber manufacturing industries), with management in the hands of the
Japanese partner. The foreign partner was limited to capital participation and
supplying technology and machinery. In this way, the main objective of tech­
nology borrowing—the establishment of import-substituting firms—was most
easily achieved.
Although certain firms (Teijin for rayon and Hitachi for electrical
machinery) developed technology independently, many firms relied on tech­
nology borrowing through DFI to shorten the time needed for technological
development.* The end result was not, however, the mere copying of technol­
ogy; rather, in adapting such technologies to domestic conditions, firms gained
the capacity for independent technological development. This was spurred

Industrialization and Protection Policy 161
by the expansion of domestic demand and other needs, but the government's
strong policy of encouraging domestic production—as in its preferential pur­
chasing of domestically produced goods—also played an important role.
Moreover, incentives in the form of tariff exemptions and subsidies only ap­
plied to firms in which foreign stock ownership was below a certain percent­
age. And, even more directly, a number of control policies were adopted to
lower foreign capital participation in domestic firms. Thus, by the end of the
1930s, capital, management, and production in most venture firms had been
transferred to Japanese nationals.
Localization Policy
The heavy electrical machinery industry exemplifies industrial development
via joint-venture companies. The rapid electrification that took place in Japan
in the 1900s lured foreign portfolio investment and at the same time raised
the domestic demand for heavy electrical equipment such as motors and dy­
namos. To meet this demand, major electrical machinery firms were set up
as joint ventures.
Once this happened, import substitution rapidly occurred in the industry.
From the 1920s to the first half of the 1930s, import dependence on various
kinds of heavy electrical equipment rapidly declined (Table 7.5). Even during
the rush to build the large-scale thermal- and hydraulic-power dynamos after
1933, large-sized dynamos could be entirely supplied by domestic production.
This was the result of the rationalization and cost-reducing programs initi­
ated during the recession in the latter half of the 1920s. Moreover, in 1930,
under a Ministry of Communications ordinance, the use of domestically
produced equipment was subsidized. The ordinance specified, however, that
only firms that had less than 50 percent of their stock owned by foreign in­
terests were to be considered domestic firms and were thus eligible for product
subsidies. As a result, each joint-venture firm tried to develop technology on
its own to reduce its foreign-equity participation.
The automobile industry exemplifies another industry established by Japan's
powerful localization policy. In 1929, domestic demand for automobiles was
35,000 units, of which 84 percent was imported as disassembled parts and
components to be re-assembled in Japan and 14 percent was imported as
finished automobiles. Domestic production could meet a mere 2 percent of
total demand. Consequently, the Japanese government adopted a policy of
subsidization to nurture domestic firms and through diverse measures was
able to control the activity of foreign firms. In 1932, for example, the yen
was depreciated by 43 percent, tariffs on automobile components and parts
were raised, and increases in the number of units assembled in Japan were
controlled. As a result, with the enactment of the Remittance Control Law

Table 7.4
DFI in Japanese Manufacturing Industries: 1899-1940
Year
Industry Name of firm established
Electrical machinery Nihon Denki 1899
Tokyo Denki 1905
Shibaura-seisakujo 1908
Fuji Denki 1923
Mitsubishi Denki 1924
Toyo-Otis Elevator 1932
Sumitomo Denki 1932
Tokyo-Western 1929
Rubber products Yokohama Gomu 1917
Chuo-gomu 1917
Automobiles Japan Ford 1925
Japan GM 1927
Sheet glass Nihon Itagarasu 1918
Artificial fibers Asahi Kenshoku 1924
Nihon Bemberg 1929 Foreign
parent firm Foreign equity
participation
Western Electric 54; 32 a
General Electric 55 b
General Electric 24 b
Siemens 30; 30
Westinghouse 40; 4
Otis Elevator 66; 70
I.S.E. 13; 7
ERP 100; 0
Goodrich 50; 9
Dunlop 100; 99
Ford 100 c
General Motors 100 c
Libby Owens 34; 17
Glanzsoff 25
J.P. Bemberg 25; 5

Year Foreign Foreign equity
Industry Name of firm established parent firm participation
Phonograph records Nihon Chikuonki 1927 Columbia 59; 2
Nihon Victor 1929 R.C.A. 68; 25
Other machinery Nihon Seiko 1907 VickerSrArmstrongs 50; 6
Toyo Kikan 1928 Babcock & Wilcox 66; 66
Kyozo Seisaku 1928 United Steel and Signal 20; 13
Toyo Carrier 1930 Carrier Corporation 50; 46
National Kinsen-Torokuki 1935 National Cash Register 70; 70
Notes:
a. First item is the percentage share at the time of establishment; the second is the percentage share in 1941.
b. Merged to become Tokyo Shibaura Denki in 1939. International Electric (33%; 16%).
c. Stopped production in 1939.
Source: Kogyo Ginko (Industrial Bank of Japan) (1950), "Gaikoku Gaisha no Honpo Toshi (Japanese Investment in.Foreign Companies)," in Minis­
try of Finance (1950(5):64-67).

164 Role of Government in Industrial Growth
Table 7.5
Decreased Import Dependence on Heavy Electrical Machinery (°7o)
1922-27 1928-32 1933-37
Water turbines and wheels 66.0 48.0 21.0
Hydraulic power dynamos 59.0 42.0 14.0
Steam turbines 97.0 82.0 23.0
Thermal-power dynamos 88.0 80.0 21.0
Transformers 46.0 25.0 0.5
Note: Import dependence is determined by the ratio of the number of units of newly built domes­
tic equipment to the number of units of imported equipment.
Source: Takeuchi (1972).
of 1939, General Motors and Ford ceased re-assembly of automobiles in Japan.
In 1939, 165,000 automobiles were produced domestically, representing 65 per­
cent of total domestic demand. Then, in 1941, because of the war between
the United States and Japan, all imports were halted.
The localization policy was implemented so that Japan could take the best
advantage of DFI, that is, the transfer of new technology to this newly in­
dustrializing country.

Eight
Trade and Industrial Policies
During the High-Growth Period
This chapter analyzes Japan's trade policy during the so-called high-growth
period of the Japanese economy from the end of postwar reconstruction to
the first oil crisis (1955-73). The objective is to identify the characteristics
that distinguish Japan's trade policy during the high-growth period from previ­
ous periods. The high growth that characterized the period 1955-73 is attrib­
uted to the successful policy of nurturing growth industries. There is a vital
connection between this policy and the growth of trade. However, Japan's
trade policy should not be viewed as a consistent policy that was adopted
methodically over the high-growth years, but rather as one that gradually
evolved from an initially protective policy to one of trade and capital liberali­
zation.
Japan was able to reap benefits from the postwar liberalization and the
gradual economic integration of the Western developed nations. As a late en­
trant into an international system characterized by division of labor, Japan
had to agree to work toward liberalization as a precondition for entry into
the system. After the high-growth period, Japan's continuing development
and export expansion led to trade friction with Japan's developed trade part­
ners, which provided further incentive to hasten the adoption of liberaliza­
tion measures.
Japanese government policy characteristically allowed industries a period
of adjustment before exposing them to international competition, thus delay­
ing the advent of liberalization. Once trade and capital liberalization began,
the government's ability to regulate the activity of private firms weakened.
Ultimately, this led to a transformation of Japan's industrial and trade poli­
cies during this period. As a result of trade friction, even the export policy
in the high-growth period gradually weakened.

166 Role of Government in Industrial Growth
PARTICIPATION IN
WORLDWIDE ECONOMIC INTEGRATION
For Japan, the recovery of the international trade system in the post-Second
World War period meant a second opening. As was the case a century be­
fore, worldwide economic integration was in progress. According to Haber-
ler (1964), it was, broadly speaking, a period of economic integration involving
phased trade liberalization, as well as the easing of controls on international
investment. Table 8.1 summarizes Japan's participation in international trade
relative to the progress of worldwide economic integration.
Although the Havana Convention did not achieve its objective of estab­
lishing the International Trade Organization (ITO), the 1947 negotiations in
Geneva opened the doors for the establishment of the General Agreement
on Tariffs and Trade (GATT) Secretariat, which performed the functions of
the proposed ITO Secretariat. The beginning of the postwar period was charac­
terized by continuous negotiations on tariff reductions for specific commodi­
ties and by the easing of trade and foreign-exchange control measures that
had been adopted in the 1930s.
At the same time, in Europe the European Atomic Energy Community
(EURATOM) and the European Coal and Steel Community (ECSC), the
predecessors of the European Economic Community (EEC) of the same six
member countries, began their activities; the remaining seven countries, en­
couraged by the EEC, banded together to form the European Free Trade As­
sociation (EFTA). In 1961, for the first time, the major Western European
nations adopted the position of the IMF's Article 8 countries, so that for­
eign exchange controls could no longer be used for balance-of-payments rea­
sons, and convertibility of the major currencies was revived. The EEC
countries dismantled their tariff barriers against member countries during
the 1960s, thus promoting the economic integration of Western Europe. At
the same time, they established common tariff barriers vis-a-vis nonmember
countries and also adopted a common agricultural policy. In 1973, the United
Kingdom joined the EEC, furthering the trend toward European economic
integration.
In response to the trend toward economic integration in Western Europe,
US. President Kennedy, desiring to promote the economic integration of the
free world's economies, initiated the package of tariff-slashing negotiations
known as the Kennedy Round. Moreover, the Organisation of European Eco­
nomic Co-operation (OEEC), which had been established to funnel US. aid
to Western Europe, was reorganized into the present-day Organisation for Eco­
nomic Co-operation and Development (OECD) to coordinate economic policy
among developed countries.
As a result of a number of negotiations, Japan was allowed to join GATT

Trade and industrial Policies: High-Growth Period 167
in 1955 and was able to share equally the existing GATT concessions, includ­
ing the tariff concessions, of other member countries under the most favored
nation (MFN) clause, particularly of the United States, that had been agreed
to in previous negotiations. However, Japan was initially allowed to temporar­
ily postpone trade and foreign exchange liberalization.' The two GATT prin­
ciples regarding tariffs are (1) reciprocal reduction of tariffs among member
countries, and (2) nondiscriminatory treatment of member-country tariff
reductions (MFN clause). That is, in the case of general tariff negotiations,
bilateral negotiations on a number of commodities must occur simultane­
ously among the member countries, and each country's tariffs that were agreed
upon during such negotiations must be applicable to all member countries,
including its new member Japan. The MFN clause allowed Japan to take ad­
vantage of three general negotiations prior to its entry into the GATT (1947,
1949, and 1950-51) as well as of the fourth and fifth general negotiations
of 1956 and 1960-61 in which it did not actively participate. Although dis­
criminatory treatment of Japan by a number of European countries, such
as France and Belgium, continued with the suspension of MFN treatment
under GATT's Article 24, it cannot be denied that Japan's trade expansion
resulted partly from a "free ride" in progressive tariff reductions, especially
those undertaken by the United States. Japan finally joined the OECD in
1964 but was able to completely postpone capital liberalization until 1967.
Table 8.2 shows the growth of industrial production and trade for devel­
oped and developing countries during the high-growth period. The data, which
includes part of the postwar recovery growth period, shows that growth of
industrial production for developed countries averaged between 4 and 7 per­
cent due to the high growth of the major developed countries, the United
States and the countries in Western Europe. Moreover, growth of the EEC
member countries—the core of Western Europe—during the first two peri­
ods was also high (7.6 and 6.7%, respectively). A comparison of the growth
performance of the two periods 1953-60 and 1960-70 with that of the 1970s
in Table 8.2 shows a drastic change in the international economic environ­
ment. Growth of trade in industrial goods largely exceeded that of industrial
production in the 1960s, reflecting a steady trend of international trade liber­
alization in industrial goods trade. Thus when Japan's industrial production
rose annually by 14.5 percent in the 1950s and by 13.1 percent in the 1960s,
growth in industrial exports was even higher, at 15.9 percent and 15.8 percent
for the 1950s and 1960s, respectively.
FOREIGN EXCHANGE MANAGEMENT
AND INDUSTRIAL POLICY
What was the role of government in the Japanese economy during the high-
growth period? Scholars are divided in their opinions (see Patrick and

Table 8.1
Japan's Participation in Worldwide Economic Integration after the Second World War
Month/year World Month/year Japan
Oct. 1947 IMF and GATT established
July 1948 OEEC established
July 1952 ECSC established
Mar. 1957 Rome treaty concluded
Jan. 1958 EEC and EURATOM established
Mar. I960 EFTA established
Jan. 1961 Intra-EEC tariffs on industrial goods
reduced by 10 percent
Sept. 1961 OEEC reorganized as OECD
July 1962 Common agricultrual policy begun for
EEC Dec. 1949
Apr. 1952
Apr. 1954
Sept. 1955
June 1960
Feb. 1961
Apr. 1962 Basic exchange rate set (360 yen to the U.S. dollar)
Peace treaty effected
General revision of tariffs
Joined GATT
Master Plan for trade and foreign exchange liberaliza­
tion announced
Convertibility of major European currencies revived
"Negative list" formula for import control
implemented

Month/year World Month/year Japan
Oct. 1962 U.S. Trade Expansion Act effected Feb. 1963
Apr. 1964 Reclassified under GATT Article 11
Joined OECD; reclassified under IMF Article 8
May 1967 Kennedy Round tariff negotiations
concluded July 1967
Apr. .1968-
Apr. 1973 First liberalization of DFI in Japan
Kennedy Round tariff
reduction implemented
July 1968 Intra-EC tariffs abolished and common
extra-EC tariffs effected
Aug. 1971
Nov. 1972 Generalized Scheme of Preferences (GSP) imple­
mented
Unilateral tariff reduction of 20 percent
Jan. 1973 EC enlarged (U.K. and three other
countries join)
Source: MITI (1982):

170 Role of Government in Industrial Growth
Rosovsky 1976; Shinohara 1982; Komiya et al., eds. 1988). Few in Japan claim
that the government played the role of prime mover, with private Firms merely
following orders in a "Japan, Inc." type of relationship (Patrick and Rosovsky
1976: chap. 1). In Japan, governmental guidance did not resemble that in some
developing or socialist countries. Rather, the government-business relation­
ship found in Japan has been unique—one that is unknown in the United
Table 8.2
World Industrial Production and Trade in Industrial Products:
Market Economy Countries
Industrial production Industrial product exports 3
(growth rate, %) (growth rate, °/o)
Year Developed
countries Developing
countries Developed
countries Developing
countries
1948-53
1953-60
1960-70
1970-79
1970-80 6.6
4.2
5.8
3.0 C 4.0
7.2
6.2
5.9 C 6.8 b
9.3
6.(f 3.5 b
9.8
11.3 C
Industrial product exports 3
(US$ billions) Industrial product exports 3
(growth rate, %)
Year To
developed
countries To
developing
countries To
developed
countries To
developing
countries
From developed countries
1960 37.7
1970 128.0
1979 572.3 16.4
33.5
191.8 } 11.5
} 5.6 5.8
7.4
From developing countries
1960 2.5
1970 8.9
1979 57.5 1.2
3.6
31.6 } 9.1
} 10.0 11.0
13.5
Notes:
a. Industrial products include Standard International Trade Classification (SITC) 5-8.
b. Includes growth rate of all commodity exports. The growth rate of industrial products from
developed and developing countries combined was 7.3 percent.
c. The growth rates for the columns under Industrial Production were available for the period
1970-1980, while those for Industrial Product Exports were available only for the period
1970-1979.
Sources: United Nations, Monthly Bulletin of Statistics, various issues.

Trade and Industrial Policies' High-Growth Period 171
States or Europe. The government has had a protective attitude toward pri­
vate firms, which have basically cooperated with the government. Further­
more, there have been many opportunities for communication and information
exchange between them, and many scholars agree that industrial policy has
been adopted on the basis of consensus. Opinions diverge, however, as to the
degree of government leadership and its effectiveness. Even during the high-
growth period, there was no single industrial policy. The changes that took
place at that time must therefore be examined carefully.
Governmental guidance was relatively strong during the postwar reconstruc­
tion period under the priority production policy (1947-50). In order to revive
the coal, iron and steel, and electric power industries, the government assumed
authority and heavily subsidized these industries. But the policy generated
huge fiscal burdens and inefficiency in assuring the supply of goods (see the
discussion on steel in Chapter 5). During the industrial rationalization pe­
riod (1950-61), greater stress was placed on achieving industrial efficiency and
strengthening international competitiveness. The government adopted a pro­
motion policy for growth industries such as iron and steel, machinery (machine
tools and other types of machinery), electronics, and petrochemicals. This
approach was a model for the one adopted by the Industrial Structure Coun­
cil (1961- ) in the 1960s. These industries pulled the economy onto the path
of high growth, and the success of this industrial policy, which the OECD
labeled a "picking-the-winners" policy, has attracted attention both in Japan
and abroad (see Chapter 9).
In selecting industries to promote, two criteria were employed: the indus­
tries had to have high income elasticity and show high productivity increases.
These industries included the priority growth sectors, which were promoted
by a concentration of public and private resources that induced investment
by private firms expecting high growth under the Income Doubling Plan of
1961-70. However, the policy was broadened into a promotion policy for in­
dustry in general and included industries such as nonferrous metals and
petrochemicals in which comparative advantage was difficult to achieve be­
cause of Japan's lack of resources.
The government's intervention in the market can be broadly categorized
by the priority policies adopted for specific industries and the enactment of
various types of regulatory measures. Priority policies include the many in­
centives granted to industries, such as direct lending by government financial
institutions (including the Bank of Japan's low-interest financing of equip­
ment investment and working capital through private banks), special tax mea­
sures such as tax deductions and grace periods for repayment of loans, and
direct subsidies. 1 But the peculiarity of Japanese industrial policy was that,
despite the abundance of incentives, many of these measures had little quan­
titative importance. Although the incentives were initially attractive, they lost
much of their appeal as firms began to improve their capital financing ability.

172 Role of Government in Industrial Growth
Some of the government's regulatory measures had no direct legal basis,
as in the case of administrative guidance measures. These measures tried to
limit competition from outside an established industry by regulating the en­
try of firms into specific industries. They were intended to end "excessive
competition" in industries as well as to maintain a stable market and pro­
mote structural reorganization. 1 But administrative guidance, which relies
largely on persuasion, has its limitations and entails costs. Altering market
behavior through administrative guidance was not very effective. Regulations
having a direct legal basis proved more effective, particularly the foreign ex­
change and trade controls. These controls were adopted at the beginning of
the 1950s to maintain balance-of-payments equilibrium. However, because
the IMF allowed these controls to continue throughout the 1950s, they also
served as an instrument of industrial protection (that is, infant industry pro­
tection during the period of industry rationalization) and gave the govern­
ment leverage to control private firms. Thus foreign exchange management
became an important foundation of industrial policy in Japan.
Foreign exchange was managed in the following manner. First, under the
Foreign Trade and Foreign Exchange Control Law/ foreign exchange owned
by the private sector was held at foreign exchange banks under the control
of the Ministry of Finance, 1 Licenses were needed to buy the foreign currency
that was used to pay for commodity imports. These licenses were classified
into three categories, depending on the type of imports involved. The first
was the Import Quota category (IQ), in which the value of the import quota
was predetermined according to the type of goods. Under this category, for­
eign exchange quotas were allocated among users. The second category was
the Automatic Import Quota (AIQ), in which the import quota was variable
and the foreign exchange allocation was approved so long as there was no
interference with foreign trade and domestic economic development. The third
category was the Automatic Approval (AA), in which requested values of for­
eign exchange were approved automatically and on a nondiscriminatory basis. 6
In the 1950s, many commodities fell under the IQ classification. Imports
of coal, scrap iron, raw cotton, and other raw materials that were indispensa­
ble to the operation of the iron and steel industry, the cotton-spinning indus­
try, and other industries were put on the IQ list. The limited amount of
imported raw materials were then rationed among firms according to such
criteria as equipment capacity or operation levels. For instance, during the
recessionary period of 1957 when there was a glut in the steel market, the
Iron and Steel Bureau of MITI set up a rationing cartel for imported scrap
iron. The cartel designated the production volume for specific commodities,
and each firm had to reduce production by between 10 and 40 percent in or­
der to prevent further price drops. Similarly, in 1957 MITI's textile bureau
controlled the volume of cotton-spinning production by rationing raw cot-

Trade and Industrial Policies- High-Growth Period 173
ton to only 8.38 million of the 92 million spindles registered. Although ra­
tioning of raw material imports was the strongest and most effective regula­
tory measure used to back administrative guidance under MITI's industrial
policy, import allocations on raw materials were discontinued in 1961, and
the manner of enforcing industrial policy changed. 7
PHASED IMPLEMENTATION OF LIBERALIZATION
By 1960, Japan was ready for trade and foreign exchange liberalization.
In June, the government, through its Economic Planning Agency, announced
its "Master Plan for Trade and Foreign Exchange Liberalization (Boeki Ka-
wase Jiyuka Keikaku Taiko)V l The motives behind this liberalization move­
ment included international pressure stemming from the strong trend in
international economic liberalization, the improvement in Japan's balance-
of-payments position in the previous few years, and increases in foreign ex­
change reserves—as well as the expectations of higher productivity resulting
from the rationalization efforts of firms. However, out of prudent concern
for industry protection, liberalization was implemented in stages.
According to the Master Plan, traded goods were divided into four
categories:
(A) Goods to be liberalized at an early date;
(B) Goods to be liberalized in the near future (that is, within three years);
(C) Goods requiring a longer period of time before liberalization; and
(D) Goods for which a suitable time period for liberalization was difficult
to determine.
The liberalization of imports such as raw cotton and wool (category A
goods) was expected to contribute to cost reduction in industries utilizing them
as inputs. Thus, goods such as textiles and iron and steel, whose domestic
industries were already internationally competitive, and those goods not di­
rectly competing with domestically produced goods were put under category
A or B. Toward the end of the 1950s, because of the strides made in domestic
production, synthetic textiles were also put under category B. Included un­
der category C were those goods for which technology development was un­
der way or those goods belonging to the core sectors of the machinery industry
where nurturing was necessary. These commodities included machine tool
equipment or implements, automobiles, heavy electrical equipment, and chem­
ical machinery parts. Category D included such agricultural commodities as
wheat, rice and other starches, and dairy and edible meat products.
Table 8.3 shows the products whose trade was liberalized after 1960. The
progress of liberalization under the Economic Planning Agency's Master Plan

174 Role of Government in Industrial Growth
can be clearly seen in the trends revealed in the table. It is also important
to note how the stages of development in domestic production are reflected
in the timing of the liberalization of various commodities such as automo-
Table 8.3
Phased Implementation of Trade Liberalization
Rate of
trade liberal­
ization 3 Number of
restricted
products b,c
Year (%) AIQ IQ Major products liberalized
1960 44 Coffee beans, movie film
1961 70 Raw cotton, radio receivers,
instant coffee, and watches
1962 88 254 Socks, sheet glass, fountain
pens, crude petroleum
1963 92 192 155 Bananas, crude sugar
1964 93 162 123 Lead, zinc, electric generators
1965 93 161 122 Automobiles
1966 93 167 124 Cocoa powder, streptomycin
1968 93 164 121 Perfume, cream
1969 93 161 118 Brandy
1970 94 133 90 Wine, margalin working
machinery, electric generators
(more than 400 kWh), chassis
fitted with engine, woolen fabrics
1971 95 87 40 Pork, candy, black tea
1972 95 86 33 Ham and bacon, refined sugar,
heavy and light petroleum
1973 31 Electrical calculators, some
marine products
1974 29 Malt, integrated circuits
1975 27 Electronic computers and parts
1977 27 Tobacco
Notes:
a. Percenl share of AA category using 1959 import composition.
b. Numbers in terms of the four-digit Brussels Tariff Nomenclature (BTN).
c. By August 1988, the number of restricted products (both AIQ and IQ) was reduced to 22,
which included 21 agricultural products and coal.
Source: Ministry of Finance, Nihon Gaikoku Boeki Nempyo (Foreign Trade Annual) (1978).

Trade and Industrial Policies: High-Growth Period 175
biles, integrated circuits, and electronic computers. For instance, the liberal­
ization of automobiles took place in 1965 when the import/demand ratio was
2 percent and the export/production ratio was 16.8 percent. In 1971, when
half of the intended capital liberalization measures had been achieved, these
ratios were 0.8 percent and 38.4 percent, respectively, clearly demonstrating
a carefully designed liberalization plan. The exception was the liberalization
of the computer industry, which was achieved hastily and put into effect in
1975 when the import/demand ratio stood at 20 percent and the ex­
port/production ratio was still 4.5 percent.
The extent of liberalization can be measured by estimating the proportion
of goods in the AA and AIQ categories to all goods, using the 1959 import
value composition. In 1963 liberalization was about 92 percent complete.
However, this estimate should be viewed with reservation because of a large
change in import composition that occurred after 1959. In particular, the IQ
category includes commodities under import control, although these were
mostly agricultural and included only 27 products in 1983, that in fact be­
long in category D. 9
Tariff Policy Reform
Tariff policy reform went hand in hand with trade liberalization (see the
discussion on tariff and nontariff restrictions in Chapter 7). In place of the
direct controls mandated by the discontinued quota system, the use of tariffs
as a tool of industrial policy increased. Because existing tariffs were unsuit­
able for the level of development of the country's industries in 1961, tariff
law reform was undertaken in June of that year (Ministry of Finance 1963).
Tariff rates were thus altered to account for the competitiveness of specific
industries, and the Brussels Tariff Nomenclature (BTN) classification used
in the United States and Europe was adopted. As a result, tariff rates on
agricultural commodities, petrochemicals, and some types of machinery were
raised. 10 The tariff rates on both total imports and dutiable imports rose in
1965-67 (see Figure 7.1).
What type of tariff structure emerged as a result of the reform? Table 8.4
shows estimates of the nominal as well as effective protective rates on selected
commodities in major industries based on the tariff rates prevailing on April
1, 1962 (Japan Tariff Association 1962). The nominal tariff rates are expressed
as a percentage of the commodity unit price, whereas the effective protective
rate is calculated in terms of the percentage of value added of labor and cap­
ital inputs in the production of the given commodity. This estimation proce-

Table 8.4
Tariff Structure at the Start of Trade Liberalization: April 1962
Effective
Nominal Rate Protective Rate
Sector 3 (%) f»
Metals and products
Iron ore, scrap iron b 0 —
Coal products 5.0 —
Crude petroleum 10.0 —
Pig iron 10.0 24.4
Steel ingots 12.5 47.0
Hot-rolled steel 15.0 35.1
Steel pipe 15.0 18.9
Nonferrous metal ore 0
Nonferrous metal 10.0 25.2
Copper products 22.2 63.7
Aluminum 19.3 40.7
Machinery
Metal-working 19.7 20.9
General parts 20.0 22.9
Optical 24.1 28.7
Watches and clocks 30.5 43.2
Heavy electrical 17.5 17.5
Automobiles 36.0 66.5
Textile products of natural fiber
Wool, raw cotton 0
Cotton yarn 5.0 9.9
Wool yarn 10.0 41.0
Cotton fabric 16.0 36.2
Woolen fabric 20.0 40.0
Knitwear 23.0 49.2
Apparel and accessories 27.8 48.8
Textile products of synthetic fiber
Synthetic textile material 25.0 31.0
Synthetic fiber 25.0 33.8
Synthetic fabric 25.0 26.3
Knitwear 23.2 24.7
Apparel and accessories 27.8 51.5
Notes: Nominal rates as of 1 April 1962. Effective protective rates are calculated by the fol­
lowing formula using input-output coefficients of 1961:
r,=«,-!, V;>'0-,?,«</>
where Tj and tj denote effective and nominal rates, respectively.
a. Commodities in each sector are ranked by amount of processing required.
b. Effective protective rates for raw materials were not estimated and are indicated by dashes.
Source: Calculated by author.

Trade and Industrial Policies: High-Growth Period 177
dure is used because it is more accurate in estimating the extent to which tariff
protection raises the value added of inputs to production (Yamazawa 1969).
The following findings can be ascertained from Table 8.4:
• The escalating structure of the nominal tariff rates (that is, rates are lowest
at the raw material stage and rise with further processing) is clearly evi­
dent for the sectors directly related to raw materials, such as metals and
natural fibers;
• The structure of effective rates is obscure. For instance, even if the nomi­
nal tariff rates are high, the effective protective rates decline when nomi­
nal tariffs on inputs are also high, thus raising the value added (compare,
for example, cotton and synthetic fabrics). In commodity groups such as
metals and natural fabrics, there is no significant rank correlation between
nominal tariff rates and effective protective rates. This is also seen in Balas-
sa's (1965) international comparison of developed countries' tariff
structures.
• High effective protective rates are observed in the primary processing stages
close to the raw material stage. For crude steel and cotton fabrics, for ex­
ample, nominal tariff rates are not high but effective protective rates are.
Because of the zero tariff or low tariff rates on imported industrial raw
materials by developed industrialized countries, even low nominal tariff
rates on processed primary goods yield high real protective rates.
• In the machinery and textile groups, luxury consumer goods carry high
nominal tariff rates and thus high effective protective rates as well.
• For goods such as machine tools and heavy electrical equipment, synthetic
fiber materials, and filaments, for which the liberalization Master Plan
intended rates to be low for both nominal and effective protection, tariffs
did not replace the quota system.
Figure 7.1 clearly shows that tariff rates were highest during the period
1960-65. Tariffs were later reduced during the Kennedy Round negotiations
(1967-71), the implementation of the Generalized Scheme of Preferences (GSP)
for developing countries (1971-), the two unilateral tariff reductions that were
undertaken primarily to prevent further yen appreciation (1972-, 1976), and
the implementation of the Tokyo Round agreements on tariff reductions
(1980-87), under which Japan's tariffs on mining and industrial products
declined to a level comparable to that of the European Community (EC).
Figure 8.1 compares the distribution of tariffs on mining and manufactur­
ing products of Japan, the United States, and the EC. Of particular interest
is the downward shift in the distribution after the Tokyo Round. The distri­
bution of tariffs for Japan, the EC, and the United States is concentrated
in the 5 to 10 percent range. However, changes in the distribution for the United

Figure 8.1
Changes in the Frequency of Tariffs on Mining and Manufacturing Products:
Comparison of Japan, the EQ and the United States
# 50 Japan
1 Basic rates
i Temporary rates (1978)
Offered rates (Multilateral Trade
Negotiations, Tokyo Round)
:.-.–! ._.
1 1 1
?50 EC
J 1 I L
2 50 United States
0 0-5 5-10 10-15 15-20 20-25 25-30 30-35 35-40 40-45 45-50 50 +
Tariff rate (%)
Note: Individual commodities with different tariff rates are grouped according to range of tariff
rate within which it falls.

Trade and Industrial Policies: High-Growth Period 179
States contrasts with that for the EC. Prior to the Tokyo Round, the distri­
bution of tariffs on mining and manufacturing products for the EC ranges
between 0 and 20 percent and is particularly concentrated in the ranges be­
tween 5 and 15 percent. However, with the tariff reductions under the Tokyo
Round, the distribution shifts downward to between 0 and 15 percent, con­
centrating especially in the ranges between 0 and 15 percent and between 0
and 10 percent. On the other hand, the distribution for the United States ranges
between 0 and 45 percent; with the tariff reductions, the distribution shifts
downward, spreading over the ranges between 0 and 30 percent. Japan's dis­
tribution is between those of the EC and the United States. Japan had two
rates before the Tokyo Round: basic rates that appeared in the official tariff
table, which were binding by GATT rules and could not be raised without
negotiation, and temporary rates, which were rates that were lowered tem­
porarily on many imported industrial products (the two unilateral tariff reduc­
tions). Although the Tokyo Round negotiations to reduce the basic tariff rates
were completed, the tariff distribution actually imposed in 1978 was that of
the temporary rates. Before the Tokyo Round, Japan's unilateral tariff reduc­
tions had caused a shift toward the EC type of distribution. After the Tokyo
Round, distribution of tariffs for the EC and Japan became even more similar.
Liberalization of Direct Foreign Investment in Japan
Even after becoming a member of the OECD, Japan was allowed to post­
pone the date of complete capital liberalization. In July 1967, the first liber­
alization measure was adopted, and liberalization gradually proceeded until
the fifth liberalization measure of May 1973. Liberalization was accomplished
by shifting from the individual (case-by-case) investigation system that was
set up under the foreign exchange laws to the automatic approval system. Thus,
foreign capitalization of a new firm's equity acquisition could be approved
up to 50 percent for category 1 industries or up to 100 percent for category
2 industries. 11 Just as liberalization gradually moved industry categories from
the individual investigation to the liberalized system, they were also moved
from category 1 to category 2 of liberalized industries.
The government was extremely careful in choosing the industries to be liber­
alized. During the first stage of liberalization, category 1 listed 33 industries,
whereas category 2 included industries such as iron and steel, cement, weav­
ing, and shipbuilding, in which domestic firms were competitive internation­
ally. Thus there was hardly any possibility of foreign capital flows into the
17 industries included on the category 2 list. During the second liberalization
stage that began in March 1969, 160 industries were included under category
1 and 44 industries under category 2. This increase in the second stage
represented a fourfold increase in the total number of liberalized industries

180 Role of Government in Industrial Growth
as compared to the number of liberalized industries in the first stage. During
the third stage of liberalization in September 1970, there was another large
increase in the number of liberalized industries, with 447 industries included
under category 1 and 77 industries under category 2. The total value of ship­
ments from the industries in both categories made up 75 percent of the
manufacturing sector's total shipments.
By April 1971, the automobile industry was fully liberalized. In August 1971,
under the fourth stage of liberalization, a shift to the "negative list system,"
in which only nonliberalized industries were announced, was undertaken. Un­
der the fifth stage of liberalization of May 1973, with the exception of 22
industries (7 industries required individual investigation and 15 industries were
from category 1), all industries were liberalized. However, foreign stock owner­
ship in existing firms continues to be closely monitored, and regulations are
stricter on takeover bids and stock held by foreigners.
The remaining regulations and the cautious liberalization process described
above show the exclusive nature of Japan's domestic investment policy and
its protection of domestic firms. Indeed, the cautiously planned postpone­
ment of trade and capital liberalization gave private firms the time needed
to strengthen their international competitiveness and technological develop­
ment. The success of this strategy is evident, because upon liberalization,
Japanese firms were already competitive (Shinohara 1982, Pt. 1). However,
it is partly a result of this success that Japan and its trading partners are now
experiencing trade friction. For instance, if American and European firms
had been able to enter Japan's domestic automobile and electrical machinery
industries and had taken part in the export expansion to the United States
and Europe along with Japanese firms, the trade problems of today might
have taken a different course, and international industrial cooperation would
probably have proceeded more smoothly.
EXPORT-PROMOTION POLICIES
In the 1950s, with the reopening of private firms to normal export trade,
Japan was plagued by a scarcity of foreign exchange and its related problems.
In particular, after the end of the export boom during the Korean War
(1950-53) the number of exportable goods was limited. Only light industrial
goods such as silk cloth and cotton fabrics were highly competitive interna­
tionally, whereas pre-Second World War exports such as iron and steel, tex­
tile machinery, and chemical fertilizers, which had been destined only for the
East Asian markets such as Manchuria, Korea, and China, were no longer
competitive. The dim prospects of the postwar export markets were worri­
some. Given this background, policymakers became conscious of the need
to nurture export industries and adopted a number of measures to promote

Trade and Industrial Policies: High-Growth Period 181
exports (Table 8.5). These took the form of preferential export financing,
export-promoting fiscal policy, and the creation of export-promoting institu­
tions, each of which is discussed in the following sections.
Preferential Export Financing
The procedure for export financing is as follows: When exporters ship their
merchandise, it is accompanied by a bill of lading against which a usance
export bill can be issued, allowing the exporter to recover export funds. Even
before lading, advance export financing can be obtained on the basis of the
export contract; thus, the needed cash is available for output processing and
Table 8.5
Export-Promotion Policy
Preferential Export Financing
A Export advance-bill system 11/1949-6/1960
A' Export trade-bill system 7/1960-6/1972
B Foreign currency-secured loan system 2/1953-8/1961
B' Foreign exchange loan system 9/1961-3/1972
C Foreign exchange bill rediscounting system 12/1965-6/1972
D Yen usance bill rediscounting system 5/1970-6/1972
E Japan Export-Import Bank's long-term trade credit 1954-1968
Special Tax Measures
F Export-income exemption 1953-63
F' Tax exemption on export-income increase 1957-61
G Export-increase depreciation system 1964-71
H Tax exemption on technology export income 1959-
I Depreciation allowance on opening of overseas branches 1958-63
J Overseas market development reserve fund 1964-72
K Overseas investment loss reserve fund 1964-74
L Special depreciation allowance for export-contributing firm 1968-70
Export-Promoting Institution
M Export insurance system 1950-
N Japan Export Bank (Japan Export-Import Bank) 1951-
O Japan Export Trade Organization (Japan External Trade
Organization) 1954-
P Supreme Export Council 1954-
Q Acknowledgment of export contribution 1963-72
Source: MITI (1982).

182 Role of Government in Industrial Growth
freight charges. Although the major portion of the export financing is denomi­
nated in foreign currency, it could be converted into yen by having the promis­
sory note discounted at foreign exchange banks.
Preferential export financing consisted of the Bank of Japan charging an
official rediscount rate on export bills or collateral loans from foreign ex­
change banks that was lower than the official discount rate for other com­
mercial bills. Foreign exchange banks then set their own rate on export bills
at the official rediscount rate plus a commission fee. Thus, export bills en­
joyed lower rates than other commerical bills.
The export advance-bill system was adopted in 1949 but was changed to
the export trade-bill system in 1960 (see A and A', Table 8.5). However, in
the case of undelivered letters of credit, the Bank of Japan used the discounted
bills of the foreign exchange banks as guarantees and lent yen funds at a higher
rate than the rediscounting rate. In the case of usance export bills after lad­
ing, short-term loans could be obtained from the foreign exchange banks un­
der the acceptance system. The system of foreign currency-secured loans and
funds (see B and B', Table 8.5) provided more profitable financing for ex­
porters. The usance export bills purchased by foreign exchange banks from
exporters were exchanged for yen-denominated promissory notes, which were
then used as collateral to draw the appropriate value of yen loans from the
Bank of Japan at low international rates of interest. Under this setup, for­
eign exchange banks became buyers of foreign currency-denominated bills
and sellers in the forward exchange market, causing a trend toward lower quo­
tations. To control this trend, the foreign exchange-bill transaction system (see
C, Table 8.5) was adopted in the second half of the 1960s so that foreign
currency-denominated bills could be bought. 12
Column 2 in Table 8.6 shows the Bank of Japan's preferential rates. The
rate for export advance-bills of the foreign exchange banks (that is, eligible
for the Bank of Japan's rediscounting [column 5, Table 8.6]) changed in
response to changes in the Bank of Japan rate and was kept higher than that
for ordinary commercial bills (column 4, Table 8.6). The Bank of Japan's
preferential export margin (column 1 minus column 2, Table 8.6) ranged from
1.4 percent to 2.1 percent. Characteristically, during periods of financial re­
straint (such as June 1957 and September 1961), export loan rates, in contrast
with the increase in official lending rates, were either not changed or were
actually lowered, so that the preferential margin increased. Thus with the over­
heating of the economy during the high-growth period and the balance-of-
payments deficits, financial restraints were applied simultaneously with ex­
port promotion.
Although the lending rate for foreign exchange loans for postlading finance
(column 3, Table 8.6) was low (2.56%) when the system first began, the lend­
ing rate rose in the second half of the 1960s in response to the rise in US.

Trade and Industrial Policies: High-Growth Period 183
interest rates. Nevertheless, preferential export financing was generally
strengthened in the 1960s. The facilities for the foreign exchange funding sys­
tem discussed previously were enlarged and export bill preferential margins
were widened. The utilization of preferential export financing, denoted by
the ratio of Bank of Japan export-related loans to the exporurelated loans
of all banks, rose from about 30 percent in the early 1960s to about 50 per­
cent in the period 1967-71. u
In the 1970s, however, the ballooning of current account surpluses, together
with the policy adopted to prevent yen appreciation, forced a drastic reduc­
tion of the preferential export margins, and in August 1971, the margins were
Table 8.6
Preferential Export Financing: 1955-72
Bank of Japan Foreign Exchange Banks
Discount Loans Discount
Discount rate of against Discount rate of
rate of export foreign rate of export
commercial advance exchange commercial trade
bills bills assets" bills bills
Month/year (1) (2) (3) (4) (5)
Aug. 1955 7.30 5.84 — 7.67 6.94
June 1957 8.40 5.84 8.40 6.57
Sept. 1958 7.30 5.48 — 7.30 6.21
Feb. (Mar.) a 1959 6.94 5.11 — 6.94 5.84
Dec. 1959 7.30 5.48 — 7.30 6.21
Aug. 1960 6.94 5.22 — 6.94 5.84
Jan. 1961 6.57 4.75 — 6.57 5.48
Sept. (Oct.) 3 1961 7.30 4.38 2.56 7.30 5.11
Mar. 1963 6.21 4.02 2.56 6.21 4.75
Mar. 1964 6.57 4.02 3.29 6.57 4.75
June 1965 5.48 4.02 3.65 5.48 4:75
Sept. 1969 6.25 4.25 4.00 6.25 5.00
May 1970 6.25 5.25 5.00 6.25 6.00
Aug. 1971 5.25 5.25 5.25 5.50 6.00
Dec. 1971
(Jan. 1972) a 4.75 4.75 4.75 5.00 5.50
June 1972 4.25 4.25 — 4.50 5.00
Notes:
a. The change in the Bank of Japan's rate was followed by rale changes at foreign exchange
banks in the month shown in parentheses.
b. This measure started in 1961 and ended in 1971.
Source: Bank of Japan, Keizai Tokei Nempo (Economic Statistics Annual), various issues.

184 Role of Government in Industrial Growth
cancelled altogether (see Table 8.6). The preferential export financing system
was finally discontinued in June 1972. 14
Whereas export financing of small and medium-sized firms came from small
and medium-sized finance corporations, long-term loans for shipping and
plant exports were extended by the Japan Export-Import Bank (Ex-Im Bank).
The latter provided cooperative financing together with commercial banks,
in which the Ex-Im Bank usually financed 70 percent of the loan and ex­
tended loans at a lower rate (4Vo) than the rate charged by commercial banks
mi
Fiscal Policy for Export Promotion
In order to promote exports in the 1950s, a special tax measure, the export-
income exemption (see F, Table 8.5), was adopted. The exemption allowed
exporting firms to claim income tax exemptions equal to a fixed percentage
of their export values (1-5%). It was strengthened in the late 1950s into the
system of accelerated tax exemptions on export income (F' ( Table 8.5) and
income for technology exports (H, Table 8.5) and subsidies for the opening
of overseas branches (1, Table 8.5).
Due to the prohibition on export subsidies under GATT Article 16, this
straightforward export subsidization system was discontinued in 1963. Indirect
tax-exemption measures were, however, adopted to take its place. For exam­
ple, the accelerated depreciation allowance for export income (G, Table 8.5)
increased the accelerated depreciation rate of exporting firms, and the over­
seas market development reserve fund (J, Table 8.5) allowed a fixed propor­
tion of sales to be set aside for losses. As a proportion of actual exports,
the indirect export subsidies did not differ much from those of the tax deduc­
tions. However, existing measures were strengthened and new measures were
adopted, such as the promotional measures for technology exports (for
proprietary rights, copyright, consulting roles) and overseas investment (H
and K, Table 8.5).
In 1966, premium rates under the accelerated depreciation allowance sys­
tem were reduced and the reserve fund for overseas market development was
increased. In 1968, the premium accelerated depreciation for firms contributing
to the country's export growth was strengthened. However, as a result of the
accelerating trend in exports, between 1970 and 1972 certain measures (G,
J, and L, Table 8.5) were discontinued, although technology export and over­
seas investment-related measures were retained.
Table 8.7 presents the values of subsidies under the major fiscal policies
for export promotion (value of presumed deductions). Since these incentives
apply to all export commodities, the ratio of the total value of subsidies (line
5) to the value of total exports (line 6) can be calculated (line 7). For the pe-

Trade and Industrial Policies: High-Growth Period 185
riod 1956-70, the ratio is about 1 percent for all industries. Since both lines
5 and 6 are for a five-year period, the ratio is the yearly average over the pe­
riod. Similar estimates are given for the iron and steel industry, including the
ratio of total subsidies to iron and steel exports. From 1950 to 1960, the ra­
tios are similar to those of all industries but increase during the period 1966-70
for iron and steel. Since iron and steel were growth export commodities dur-
Table 8.7
Reduction of Tax Revenues under Export-Promotion Measures
(¥100 millions, Vo)
1950-55 1956-60 1961-65 1966-70 197]-75
A. All Industries
1. Export-income exemp­
tion, tax exemption on
export-income increase 122 582 792 1,483 515
2. Overseas market de­
velopment reserve fund 234 475 439
3. Technology export-
income depreciation 18 198 362
4. Overseas investment loss
reserve fund 63 60 889
5. Total (1-4) 122 582 1,107 2,216 2,205
6. Total exports 30,143 56,686 106,823 246,570 590,81
7. Implicit subsidy rate
.W(5-6) 0.40 1.03 1.04 0.90 0.37
1953 -55 1956-60 1961-65 1966-70
B. Iron and Steel Industry
8. Export-income exemption 804 4,567 4,882 —
9. Tax exemption on export-:
income increase 58,025
10. Overseas market develop­
ment reserve fund 11,329
11. Technology export-income
depreciation 2,303
12. Total (8-11) 804 4,567 4,882 71,656
13. Steel exports 203,858 476,537 580,739 3,343,164
14. Implicit subsidy rate (%)
(12^13) 0.39 0.96 0.84 2.14
Sources:
A. Tsuruta (1982:60).
B. Tecson (1985).

186 Role of Government in Industrial Growth
ing this period, the strengthening of fiscal policy to promote exports was
strongly reflected in the exports of these goods.
The tariff rebate system for export goods was a system of rebates on tariffs
paid on imported materials used in export production. Adopted first in 1966,
this new export tariff rebate system employed a fixed rate of tax rebates for
61 commodities in order to avoid the complexities of separate computations.
Essentially, if the rebate was within the limits of the value of the tariff rate,
it was neutral with respect to resource allocation so that even GATT allowed
it. In the case of fixed average rates, however, the portion that exceeded the
tariff rate was considered an export subsidy. Under the new system, the value
of such tariff rebates was 16.6 million yen as opposed to the 1.56 billion yen
for total commodity exports (or about 60% of total exports), so that the tariff
rebate rate was 0.11 percent. The tariff rebate rates for specific commodity
groups were 0.13 percent for textile goods (average of 12 commodities) and
0.09 percent for metals (average of 14 commodities). This is because value-
added rates were higher in Japan and there were many imported raw materi­
als with few or no tariffs. In consideration of GATT criticism, the average
tariff rebate was reduced, leading to the low average rates (Nihon Kanzei
Kyokai 1965).
Export-promoting Institutions
In the early 1950s, with the reopening of direct private export trade, a num­
ber of export-promoting institutions were established by the government. These
institutions provided a useful infrastructure for the export growth of the late
1950s, although it is difficult to quantify their promotional effects. Many of
them continue to function, although in the 1970s their objective shifted from
simple export promotion to other important areas of overseas investment,
resource development, and import promotion.
One such institution is the export insurance system, which was set up in
1950. Under the export insurance system (M, Table 8.5), the government made
up part of the losses incurred by exporters resulting from unforeseen events
overseas. The system's scope of responsibility was widened in 1956 to include
foreign investment. Another institution, the Japan Export-Import Bank (N,
Table 8.5) (originally the Japan Export Bank), was established to extend long-
term financing to ship and plant exports. Gradually, resource development
imports and the financing of foreign investments increased. A third institu­
tion, the Japan External Trade Organization (O, Table 8.5) (originally the
Japan Export Trade Organization), concentrated its efforts on surveying over­
seas markets to promote exports. Today its activities include resource develop­
ment, economic cooperation, import promotion, and the promotion of the
internationalization of domestic firms.

Trade and Industrial Policies: High-Growth Period 187
The Supreme Export Council (P, Table 8.5), a semiprivate agency, was origi­
nally set up in the mid-1950s to set annual targets for exports and to discuss
export-promotion policy. The export contribution of various firms was evalu­
ated using a point system that estimated the level of the firms' export perfor­
mance (Q, Table 8.5). Beginning in 1968, firms contributing to exports were
allowed premium accelerated depreciation allowances and increases in for­
eign currency reserve funds, which were, in effect, export subsidies. In 1969,
however, the Supreme Export Council was reformed into a trade council; the
Council's activities were discontinued, and by 1972, all of the measures that
had been in effect were eliminated. In 1982, the Supreme Export Council was
revived but this time its goal was to concentrate on generating policies for
import expansion.
A number of scholars have stressed the resource misallocation effects of
the government's export-promotion measures. Indeed, the menu of incentives
was abundant and was retained until the early 1970s, despite changes in form,
because of pressure from special interest groups. However, to what extent these
promotion measures affected Japan's export expansion is another question.
Estimates of the impacts of the subsidies involved in the preferential export
financing and fiscal measures per unit value of exports need to be analyzed.
In the case of preferential export financing, the rate of reduction in in­
terest payments per unit value of exports can be estimated. Assuming that
export bills are normally three-month bills, the interest rate paid is one-fourth
of the annual rate, and the rate of reduction in interest payments is one-fourth
of the preferential margin. In the case of advance export financing, it is at
most 0.5 percent; for postlading financing, 1.2 percent, or a total of less than
2.0 percent. Although the Japan Export-Import Bank's financing was long
term, with a large preferential margin, it is disregarded here because this type
of financing was limited to shipping and plant exports. Moreover, as shown
earlier, the rate from special fiscal measures was at most 2 percent so that
the combined financial and fiscal rate was about 4 percent. It is difficult,
however, to measure the degree of the export-promotion effect of a subsidy
rate of this magnitude.
One approach to measuring the export-promotion effect of a subsidy is
to compare it with the export-promotion effect of other variable changes such
as yen depreciation. In the second half of the 1960s, the exchange rate was
360 yen to the dollar (the yen was considered undervalued in many respects).
Due to the yen's appreciation under the Smithsonian Agreement of Decem­
ber 1971, the value of the yen rose by 14.4 percent (that is, to 308 yen to the
dollar), and, between March and October 1973 (before the first oil crisis),
under the fluctuating exchange-rate system, its value increased by 26.4 per­
cent (to 265 yen to the dollar). In any case, the rates of appreciation greatly
exceeded the export subsidy rate estimate.

188 Role of Government in Industrial Growth
Another approach is to compare the export-promotion policy with those
of contemporary developing countries. Many of today's developing countries
are already employing preferential export financing, export-promoting fiscal
measures, and a system of tariff rebates for export production. In South Korea,
at the commercial bank level, the export bill preferential margin averaged 18.4
percent during the period 1966-70 and 8.5 percent during the period 1975-78.
Assuming that these were three-month export bills as in Japan, the interest
rate reduction in preferential financing was one-fourth of the preferential mar­
gin, or 4.6 percent for the period 1966-70 and 2.1 percent for the period
1975-78. Moreover, the value of domestic tax exemptions related to exports
was on average 9.8 percent during the period 1966-70 and 9.2 percent during
the period 1975-78. 11 The ratios of tariff rebates on exports to the value of
exports were 12.6 percent and 8.8 percent for the respective time periods in
South Korea. This is because value added was low and import tariff rates
were high.
Compared with the policies adopted by South Korea, Japan's export-
promotion measures were modest. The promotion effects were probably
greater than is implied by actual export figures in the 1950s, when quantita­
tive controls on financing were severe. However, after the 1960s the effect might
not have been that great.
In the 1960s, when such export measures were being adopted, voluntary
export restraint measures were applied on export commodities that encoun­
tered import restrictions and demands for export control in U.S. and Western
European markets. This is discussed in detail in Chapter 9.

Nine
Trade Conflicts
and Structural Adjustment
This chapter analyzes the dramatic changes in the Japanese economy in the
1970s and 1980s. In these two decades, the world experienced serious fluctu­
ations in prices and supplies of petroleum and other primary commodities.
The major industrial countries have been affected by sudden constraints in
raw materials and energy inputs, which they have tried to overcome by im­
plementing new policies and adjusting their industrial structures. Some coun­
tries have fared better than others, but huge trade imbalances have occurred.
Japan, with its high import dependence, was seriously affected by the fluctu­
ations in commodity prices, but it made appropriate adjustments and recov­
ered quickly from the setback in growth.
Along with the improvement in Japan's economic growth performance, the
country also did well in export growth. Japan's share of world exports in­
creased from 6.6 percent in 1970 to 9.8 percent in 1987, while that of the United
States decreased from 14.9 percent to 10.6 percent.
Japan Ts persistent trade surplus has been the source of several conflicts with
its major trading partners. Although Japan has been investing its current ac­
count surplus abroad, it needs to restructure its economy so that its external
balance is not significantly out of line with its partners. The convention of
adjusting policies at the border alone is not enough. Because Japan is now
well recognized as a major economic power, Japan's strategy and approach
to its domestic and external activities should be completely reoriented. The
country has now entered the stage where it shares a leading role with the United
States and Western Europe in research and development (R&D). Japanese
firms are moving from the traditional goal of expanding shares in existing
markets to exploring new markets.
Japan has begun to make adjustments and restructure its economy, and

190 Role of Government in Industrial Growth
this trend will continue in the 1990s. The restructuring and adjustments of
the Japanese economy in the next decade may be undertaken in the multilateral
context of the Uruguay Round and Pacific cooperation.
GLOBAL ADJUSTMENT
AND DIVERGENT PERFORMANCE
In the 1970s and 1980s, there were large fluctuations in global supply and
demand conditions for major inputs. In the late 1960s and early 1970s, prices
of petroleum and other primary products increased dramatically. The sud­
den rise in commodity prices contributed to upward pressures on the price
levels of many countries, and runaway inflation and the threat of unemploy­
ment were serious concerns in the latter half of the 1970s. In response to these
conditions, deflationary policies were instituted by the industrialized coun­
tries in the early 1980s. By 1987, commodity prices had declined to pre-1980
levels. Figure 9.1 shows the cyclical changes in commodity prices and the vola­
tile growth performance of the industrialized countries, although their ag­
gregate, annual average figures show a much smoother pattern of change.
The patterns of change are described in greater detail below.
Fluctuation and Policy Response
A quadrupling of petroleum prices between October 1973 and early 1974
ended the period of steady and high economic growth and stable prices that
had continued in the world economy since the mid-1950s. Even before this,
however, other commodity prices had begun to increase in the late 1960s,
although less drastically, and wages were rising in most industrial countries
throughout the 1960s. The rise in prices of many commodities and labor largely
resulted from the excess demand for productive resources toward the end of
the rapid-growth period, which led to accelerated inflation. International
monetary disturbances in the late 1960s, which in 1971 eventually led Presi­
dent Nixon to suspend the official convertibility of the U.S. dollar, added
to excess international liquidity and aggravated the inflationary trend.
The oil shock of 1973 not only caused rapid inflation, but it also contributed
to large trade imbalances, heavy deflationary pressures, and negative growth
in the major developed countries. To restore the level of output and employ­
ment, governments undertook active fiscal expansion. Although medium eco­
nomic growth was restored by the latter part of the 1970s, inflation continued
to be a problem and in fact was further accelerated by the second oil shock
of 1979-80.
Governments responded to the second oil shock with deflationary mea­
sures. The U.S. Federal Reserve, for example, tightened the U.S. money sup-

Figure 9.1
Fluctuations in the Global Economy
125 I 1 1 1 r
°/o

250 1 1 1 1 1
Exchanae rate (¥/$)
– ""*\ Japan's terms of trade
• \ (1980 = 100)
_ \ \ –
• \
l i i i
1960 1965 1970 1975 1980 1985 1990
Year
Source: IMF (1988).

192 Role of Government in Industrial Growth
ply, but this led to much higher interest rates than had been planned. The
higher interest rates attracted substantial inflows of money to the United States,
which caused the U.S. dollar to appreciate. Japan and the EC countries fol­
lowed the United States in tightening their monetary policies, partly to curb
their own inflation and partly to discourage the outflow of money and stop
the depreciation of their currencies. But the overall tightening of money sup­
ply by the United States, Japan, and the EC countries resulted in very high
global interest rates and the U.S. dollar remained overvalued.
Although all of the countries responded with deflationary measures, the
fiscal responses to these conditions varied among the industrialized coun­
tries. Japan adopted an austere fiscal policy and succeeded in reducing its
fiscal deficits. Moreover, Japanese production, faced with stagnant domestic
demand, became even more oriented toward the export market, which resulted
in a huge trade surplus. Despite the trade surplus, the undervaluation of the
yen continued because of capital flows to the dollar market.
In contrast, the United States continued with fiscal expansion and accumu­
lated deficits in both its current accounts and fiscal budgets. Among the EC
countries, examples of both types of responses were found. West Germany's
performance was similar to that of Japan, whereas that of the United King­
dom and France was closer to that of the United States.
The agreement in September 1985 by the Group of Five (Japan, the United
States, the United Kingdom, West Germany, and France), however, changed
this. High interest rates and the overvalued U.S. dollar were quickly adjusted.
The downward adjustment of interest rates occurred first as inflation slowed
and tight monetary policies were eased in the major industrialized countries.
Although the fluctuations in the world economy have returned to their pre-
oil shock levels, the large imbalances between the developed economies re­
main unresolved.
Disparity in Economic Performance
The economic growth and trade performance of the developed countries
prior to 1973 stand in marked contrast to their performance after the first
oil crisis. The average GDP growth rate of all developed countries was 4.8
percent in the period 1960-70, but fell to 2.3 percent over the period 1974-83.
Similarly, average annual growth of total exports was 8.2 percent in the former
period, but dropped to 6.3 percent in the latter period.
Table 9.1 compares the macroeconomic performances of five developed
(Japan, the United States, the United Kingdom, West Germany, and France)
and two Asian developing (South Korea and Thailand) countries over four
five-year periods. The four five-year periods are the high-growth period
(1966-70), the recovery from the first oil shock (1975-79), the period of dis-

Table 9.1
Growth and Trade Performance Variation During Global Adjustment
Japan United
States United
Kingdom West
Germany France South
Korea Thailand
Growth rates of real GDP
(period average, °7o)
1966-70 11.0 3.0 2.5 1.7 6.4 12.7 8.6
1976-79 5.2 4.2 2.7 3.9 3.5 10.6 8.0
1980-84 3.9 1.9 0.7 3.6 1.5 5.9 5.5
1985-87 3.9 3.3 3.5 2.2 2.0 9.4 4.3
Balance of trade
(annual, period average, USS billions)
1966-70 84 955 -2,692 3,811 – 1,185 -901 -446
1976-79 4,160 -27,302 -9,753 14,792 -4,517 -2,312 -1,190
1980-84 11,742 – 62,209 -5,558 14,664 – 14,029 -3,039 -2,829
1985-87 70,044 – 163,978 – 16,657 47,829 -7,042 2,846 -1,247
Source: IMF (1988).

194 Role of Government in Industrial Growth
inflationary policies (1980-84), and the period of readjustment (1985-87). The
last period is cut short because data are unavailable for later years. During
the global adjustment following the first oil shock, economic growth of the
five industrialized countries did not differ greatly, reflecting the stagnant econ­
omies of Japan and West Germany and the more dynamic economies of the
others. But a significant difference can be found in the trade balances be­
tween the two groups.
South Korea and Thailand, which experienced high growth beginning in
the 1960s, continued to experience rapid growth throughout this period (see
Table 9.1). In general, the Asian NIEs experienced a growth rate of 8-10 per­
cent, and the ASEAN countries 6-8 percent on average. In the early 1980s,
however, the slowdown in the world economy seriously affected the Asian
developing countries. While their exports to the developed countries' mar­
kets stagnated and their export earnings decreased drastically, their demand
for imports grew as their development needs expanded. As a result, many
of the Asian developing countries incurred large trade deficits. To avoid fur­
ther accumulation of foreign debt, many countries were forced to cut back
on their development programs. Both the Asian NIEs and ASEAN countries
experienced setbacks in their growth between 1982 and 1985. However, since
the end of 1986, these economies have been recovering and their trade balances
have improved steadily.' South Korea's deficit turned to a surplus in 1986,
and Thailand's deficit improved in 1987-88. Japan's recovery has also bene­
fited from the continued rapid growth of its neighbors.
TRADE CONFLICTS AND COUNTERMEASURES
Sectoral and Overall Conflicts
In recent years Japan's better trade performance has caused conflicts with
its trading partners, but trade conflict is not a new phenomenon. In the 1920s
and 1930s, Japanese cotton and silk fabric exporters encountered prohibitive
restrictions in India, Canada, and Australia. In the early 1950s, restrictions
were imposed on selected goods, including canned tuna, sewing machines,
pottery and cotton fabrics, by the United States and unbleached cotton fabrics
by the United Kingdom. Trade conflicts spread to iron and steel in 1966, to
televisions in 1968, and to synthetic fabrics in 1969. Complaints were made
about Japan's cheap export prices and the rapid increases in their exports,
and antidumping suits were filed. In the 1970s and early 1980s, complaints
were also heard with regard to industrial machinery (1978), automobiles (1981),
and VCRs (1981).
As Japan caught up with the industrialized countries in terms of industri­
alization and development, trade conflicts increased, especially as new ex-

Trade Conflicts and Structural Adjustment 195
ports were developed. Every export was shipped to the industrialized coun­
tries' markets to compete with domestic products. American and European
producers demanded protection, and the conflict was often settled with volun­
tary restrictions placed on Japanese exports, monitoring of export prices, and
more tightly controlled trade arrangements. Textiles are a prime example. From
1969 to 1971, the United States was in serious conflict with its trading part­
ners over the unrestricted import of textiles into the United States. The con­
flict culminated in a voluntary export restraint (VER) arrangement and later
became a focal point of negotiations between the United States and the Asian
NIEs. These negotiations eventually resulted in a managed trade network
under the 1974 Multi-Fibre Arrangement (Destler et al. 1979; Yamazawa
1988a). Many export commodities destined for the United States or Europe
are still governed by VERs, which indicates continuation of the conflicts be­
tween Japanese exporters and American or European producers in particu­
lar sectors.
In the 1980s, however, a conflict emerged in the overall trade relationship
between Japan and its major trading partners. The export strategy of Japanese
firms, as well as the trade and industrial policy of the Japanese government,
became targets of criticism at summit and OECD ministerial meetings. The
Japanese government received complaints from the EC (in September 1981),
the United Kingdom (in October 1981), and the United States (in December
1981) about their persistent trade deficits with Japan. Complaints were also
heard about the complicated distribution system of the Japanese market and
the commercial practices that prevent foreign access. The trading partners
demanded market liberalization and import expansion.
Japan responded with a series of trade liberalization programs including
unilateral tariff reductions, expansion of import quotas, improved import
financing, and the dispatch of import missions. In July 1985, the Japanese
government announced an Action Program—the seventh in a series of trade
liberalization programs since 1972—to further reduce barriers and promote
imports. The Action Program included the reduction or elimination of tariffs
on over 1,800 items as of January 1986, and a package of liberalization and
simplification measures in standards, certification, and government procure­
ment procedures. This program goes beyond previous trade liberalization pro­
grams in its principles and scale and is expected to increase imports and reduce
surpluses in the long run.
Factors Causing Trade Conflicts
The factors causing trade conflicts are complicated and interrelated. First,
individual sectoral conflicts result mainly from the catching-up industrial­
ization process of developing countries and the resulting adjustment dif­
ficulties faced by the importing industrialized countries. International

196 Role of Government in Industrial Growth
competitiveness comes not only from cheap labor costs, but also from dy­
namic gains such as the new technology and economies of scale acquired in
the catching-up process. The prolonged process involved in reallocating em­
ployment between sectors in mature economies adds to the difficulty of ad­
justing to changing comparative advantage. Trade conflicts have now spread
to some high-tech industries where the importing countries wish to protect
v domestic production for security or other strategic reasons.
Second, large trade imbalances have been generated in the process of ad­
justment to the changing conditions and comparative advantage positions.
This is most evident in the conflicts between Japan and the United States.
Japan's trade surplus with the United States was around USS12-S13 billion
in 1981-82, but increased rapidly to reach US$51.4 billion in 1986. The United
States made stronger demands for further liberalization of the Japanese mar­
ket and a reduction of the bilateral imbalance to a tolerable level. The in­
creasing bilateral trade imbalance between Japan and the United States simply
reflects the increasing imbalances of these two countries with the world, which,
as Bergsten and Cline (1987) correctly analyzed, can be attributed to the mac-
roeconomic policies of the two countries. Japan's trade surplus with the world
expanded from around US$20 billion in 1981-82 to US$93 billion in 1986,
whereas the U.S. trade deficit with the world expanded from US$28 billion
in 1981 to US$148 billion in 1986.
The prolonged trade imbalances between the United States and Japan have,
however, aggravated the sectoral conflicts between the two countries, and trade
negotiations and VER agreements have continued to proliferate. The con­
flict over semiconductors was concluded in August 1986 with a new com­
promise agreement on American price surveillance of Japanese manufacturers.
But, in March 1987, the U.S. government invoked Article 301 of the U.S. Trade
Act because of an alleged violation of the antidumping provision by a Japanese
producer, and prohibitive tariffs were imposed on the imports of related
Japanese products.
Market-Oriented Sector-Specific (MOSS) talks were also held to discuss
the liberalization of the Japanese market. At first, the talks dealt with four
sectors—namely telecommunications, medicine and medical equipment, elec­
tronics, and forestry products. In subsequent discussions, the talks have been
expanded to include principles such as equal access in the bidding process
for construction projects, which American construction companies have
sought in their efforts to bid on construction projects such as the Kansai In­
ternational Airport project. The American demand for Japanese market liber­
alization has been most forthright with respect to agricultural products. The
United States submitted Japan's residual import restrictions on 12 agricul­
tural products to the GATT panel, which decided that in 10 of the 12 cases
Japan had violated GATT rules. The U.S. Rice Millers Association also
denounced Japan's policy of rice self-sufficiency.

Trade Conflicts and Structural Adjustment 197
These demands for Japan to liberalize its markets are well founded, and
the Japanese government has begun to accommodate them as announced in
the Action Program and the Maekawa Report. These changes are beneficial
to the Japanese economy. Trade conflicts with the United States, together
with the rapid appreciation of the yen, have directly prompted the restruc­
turing of the Japanese economy..
Trade Conflicts with Asian Countries
Japan's trade conflicts are not confined to the United States and Western
Europe. Consistent trade surpluses with the Asian NIEs and some ASEAN
countries have raised major issues at bilateral governmental talks and the
Japan-ASEAN Forum. Japan's trade surplus basically reflects the strong im­
port demand for machinery, parts, and industrial materials needed by these
countries in their development. It also reflects the different stages of develop­
ment of Japan and these Asian countries. Thus the demands made of Japan
focus on the expansion of imports to Japan, improved market access for major
ASEAN exports such as boneless chicken, bananas, and plywood, and an
improved Generalized Scheme of Preference (GSP). Since these Asian coun­
tries are aiming to catch up in their industrialization, Japan was also asked
by the NIEs and ASEAN countries to increase imports of new manufactured
products (Indonesia), invest in new industries (Thailand), and provide better
access to high technology (South Korea and Taiwan).
The Action Program included some official measures to respond to these
demands and requests. But like producers in the United States and Western
Europe, Japan's domestic producers complain about increasing imports, es­
pecially as the exports of the Asian countries become more competitive in
world markets. The increased competition between Japan and its Asian neigh­
bors has been accelerated by the rapid appreciation of the yen, but this com­
petition has also tended to promote Japan's economic restructuring.
JAPAN'S MACROECONOMIC POLICIES
AND PERFORMANCE
In the 1970s and early 1980s, macroeconomic policies focused on overcom­
ing the two oil shocks, but lately they have been increasingly geared toward
correcting external imbalances and mitigating trade conflicts. The macroeco­
nomic policies in both periods have helped to provide a framework for the
restructuring of the Japanese economy, with individual firms changing their
strategies and behavior accordingly.
Table 9.2 displays the basic characteristics of Japan's macroeconomic per­
formance during the 1970s and 1980s. The effects of the first and second oil

198 Role of Government in Industrial Growth
shocks are distinctly revealed by the fluctuations in the growth rate of real
GDP and in the current account balance. Negative growth was recorded in
1974, and positive growth, though very low by Japanese standards, was re­
corded in 1975, with current account deficits in both years. Between the two
oil shocks, during 1976-78, the economy recovered. Medium growth was main­
tained and a current account surplus was accumulated. Growth rates declined
again in 1979-80 (though not as much as in 1974-75), and the current ac­
count deficit was US$10 billion for both years. After 1981, growth reached
a low of 3 percent, but the current account surplus attained in 1981 continued
to expand. In 1984 and 1985, the economic growth rate increased with fur­
ther increases in the current account. In 1986, although the growth rate
dropped to 2.6 percent, the current account surplus ballooned to US$86 bil­
lion, which in 1986-87 led to a drastic change in macroeconomic policies.
The relationship between two growth rates in Table 9.2—the growth rate
of national expenditure and the growth rate of GDP—invites further investi­
gation. National expenditure is the sum of consumption, investment, and
government expenditure. If the growth rate of national expenditure is multi­
plied by the ratio of national expenditure to GDP, it will give the contribu-
Table 9.2
Pattern of Japan's Economic Growth: 1974-87
Growth rate
of real GDP
m Growth rate
of national
expenditure
m Current account
(US$ billions)
1974 -1.4 -0.5 -4.7
1975 2.7 1.8 -0.7
1976 4.8 4.0 3.7
1977 5.3 4.4 10.9
1978 5.2 6.0 16.5
1979 5.3 6.5 -8.7
1980 4.3 0.8 -10.8
1981 3.7 2.1 4.8
1982 3.1 2.8 6.9
1983 3.2 1.8 20.8
1984 5.1 3.7 35.0
1985 4.9 3.8 49.2
1986 2.4 3.8 85.8
1987 4.2 5.0 87.0
Source: Calculated from national account and balance-of-payments statistics in the 1980 con­
stant price series (Bank of Japan, Keizai TokeiNempo [Economics Statistics Annual], 1988).

Trade Conflicts and Structural Adjustment 199
tion of domestic demand to overall growth of real GDP. 2 For the period un­
der study, this ratio is 0.98-1.01, which indicates that the growth of national
expenditure closely approximates the entire contribution of domestic demand
to overall growth. The difference between the two growth rates gives the con­
tribution of external demand (exports minus imports of goods and services),
which is associated with the change in the current account balance. If the
current account increases, then the contribution of external demand is posi­
tive, and if it decreases, then the contribution of external demand is negative.
Table 9.2 shows that the current account position improved and that its con­
tribution to overall growth was positive from 1975 to 1978 and from 1980
to 1985, but negative for 1979 and 1986-87. 3 This macroeconomic perfor­
mance directly resulted from changes in the macroeconomic policy mix, which
is reviewed below. 4
High-Growth Period
The period 1955-70 was characterized by a clear policy mix: The exchange
rate was pegged at 360 yen to the U.S. dollar and was allowed to fluctuate
within a 1.5 percent band. Capital did not move freely across the border. Fis­
cal policy had almost a neutral effect on the business cycle under the balanced
budget principle. Monetary policy was therefore used to deal with both the
balance of payments and the business cycle. Fortunately, the policy goals often
synchronized with each other. Economic booms coincided with balance-of-
payments deficits because of the increasing imports of raw materials for in­
dustrial production. The Bank of Japan tightened its money supply by mainly
raising official rediscount rates combined with a tighter discount window for
lending to firms. During the recession, export sales rose and an external pay­
ment surplus resulted, which led to the easing of the money supply. The
balance of payments tended to constrain continued excessive growth, and un­
der the IMF regime the Japanese government was only concerned with its
own external and internal balances. Since the early 1970s, however, the mac­
roeconomic policy mix has been quite different.
Smithsonian Agreement
In December 1971 the Smithsonian Agreement set the yen-dollar exchange
rate at 308 to 1, and in March 1973 the yen was floated with other major
currencies. Since then, changes in the value of the yen vis-a-vis other major
currencies have played a major role in adjusting external imbalances, but the
time lag in adjustment (the J-curve effect) has often been much longer than
expected. Capital movement has become more sensitive to international differ­
ences in interest rates and has often caused erratic movements in exchange

200 Role of Government in Industrial Growth
rates. This, in turn, has led central banks to maneuver their monetary policy
to achieve internal balance.
A balanced budget could no longer be maintained in Japan because ex­
penditures tended to increase as inflation increased and revenues decreased
due to the depressed economy. The financing of the accumulating budget
deficits required an increasing share of revenues, which, in turn, reduced avail­
able revenue for other expenditures. All macroeconomic policy measures be­
came severely constrained.
First Oil Shock and Recovery
The effect of the first oil shock on Japan was more severe than on any
other industrialized country. Following the quadrupling of petroleum prices
in 1973-74, wholesale and consumer prices rose by 28 and 23 percent, respec­
tively. To cool the inflationary pressure, the Bank of Japan tightened the money
supply by raising its rediscount rate to 9 percent. Government expenditures
became inflated and, combined with decreased tax revenues, resulted in huge
fiscal deficits. Business activities were curtailed both by limited oil supplies
and a tightened money supply. This slowed Japan's industrial activity by as
much as 17 percent, and GDP growth became negative.
In the period 1973-75, Japan had a huge current account deficit, which
together with the oil price hike, lowered Japan's terms of trade by 42 percent
(see Figure 9.1). Stagnant domestic demand led to the promotion of exports
to achieve a positive growth rate. The Bank of Japan managed to keep the
exchange rate at between 284.90 and 307.00 yen to the dollar during 1975-76
as compared to about 253.00 yen to the dollar in 1973-74. In 1976, growth
recovered to around 5 percent and the current account showed a surplus; thus,
monetary policy was gradually relaxed. Japan was criticized, however, for un­
dervaluing the yen. From early 1977, the Bank of Japan allowed the yen to
continue to float on the market, and by November 1978 the yen was trading
at 175.50 to the dollar. Japan was also requested at the London Economic
Summit in May 1977 to adopt an expansionary policy to stimulate domestic
demand and increase imports, so as to play a "locomotive" role, together
with the United States and West Germany, in leading the world's recovery
from the oil shock. The Bank of Japan's rediscount rate was lowered to 3.5
percent, and the fiscal deficit was continued.
Second Oil Shock and Disinflationary Response
The macroeconomic policy mix that was geared to international harmony
was terminated by the second oil shock, which occurred as incremental price
increases in 1979-80. All countries became concerned about the state of their

Trade Conflicts and Structural Adjustment 201
own economies. The United States led the disinflationary response to the sec­
ond oil shock with high interest rates. This invited capital inflows, which re­
sulted in an overvalued dollar. The yen-dollar exchange rate rose to as high
as 263.65 to 1 in February 1985. The Bank of Japan's rediscount rate was
raised to 9 percent in March 1980, and an austere fiscal policy ("zero ceil­
ing" on all government expenditures) was introduced in Japan to reduce the
public debt from 4.5 percent of GNP in 1982 to 2.1 percent in 1985.
Although Japan was able to adjust more readily to the second oil shock
and the setback in economic growth in Japan was much less than in other
industrialized countries, growth slowed to around 3 percent. The disinflation­
ary policy mix was continued and production was increasingly oriented toward
exports. This was helped by the fiscal expansion and increase in imports of
the United States, although the trade conflict between the United States and
Japan widened.
Redirection of the Macroeconomic Policy Mix
The agreement of the Group of Five in September 1985 was aimed at cor­
recting the overvalued dollar and resolving trade imbalances. Japan was un­
happy with the undervalued yen and welcomed its appreciation to 190-200
yen to the dollar, although no target rate was agreed upon at the summit.
The yen's appreciation was initiated by the joint intervention of the Bank
of Japan and the U.S. Federal Reserve but its value increased beyond expec­
tations in market trading. In March 1986, when the yen fell below 180 yen
to the dollar, the Bank of Japan intervened to halt further appreciation of
the yen by buying dollars and selling yen. However, this effort failed to stop
the yen's further appreciation.
After 1986, the yen-dollar rate fell repeatedly and rapidly, with short peri­
ods of limited stability, until the middle of 1988 when it reached the 120-yen
level. Throughout this period, monetary policy remained relaxed. The Bank
of Japan's rediscount rate was reduced from 4.5 percent in January 1986 to
a historical low of 2.5 percent in February 1987. The rediscount rate was main­
tained at this low level in case the narrowing interest rate difference between
the United States and Japan should invite Japanese capital back from the
United States.
The switch to an expansionary fiscal policy was delayed by concerns over
fiscal deficits. In September 1986, the Japanese government implemented a
public investment program of 3 trillion yen in order to boost domestic de­
mand and achieve a 4 percent growth rate. The growth rate of real GDP,
however, turned out to be only 2.6 percent in 1986 and as low as zero in April-
June 1987, although it increased after that. In May 1987, before the Venice
Economic Summit, the Japanese government pumped an additional 6 tril-

202 Role of Government in Industrial Growth
lion yen (about US$40 billion) into the economy in order to boost domestic
demand. Private housing investment, private consumption, and company ca­
pacity investment were stimulated, and as a result GDP grew at 4.2 percent
in 1987 despite the negative contribution of net exports.
The correction of Japan's current account imbalance was delayed because
of the J-curve effect. Although the volume of Japanese exports decreased
after the middle of 1986, the trade surplus did not decrease in dollar terms
until April 1987. The value of imports did not increase because of the decline
in oil prices. As a result, the Japanese current account declined continuously
after May 1987, and only a small increase was recorded for the current ac­
count surplus for 1987 in comparison with the previous year.
Although a huge trade imbalance still remained, the restructuring of the
Japanese economy proceeded in the context of the following macTo policy
mix changes. First, slower growth itself forced firms and industries to under­
go structural adjustments. Second, the rapid appreciation of the yen and the
improved terms of trade in 1973 before the first oil shock, in 1978, and after
1985 severely affected export-oriented and import-competing industries and
drastically altered their comparative advantage. Third, the trade imbalances,
trade conflicts, and resulting series of market liberalization measures have
encouraged imports and increased domestic competition. Fourth, lively domes­
tic demand encouraged more Firms to produce for the domestic market. Fifth,
the government's policies were increasingly geared toward international har­
mony and promotion of the restructuring of the economy.
THE RESTRUCTURING OF INDUSTRIES AND FIRMS
Table 9.3 summarizes the MITI report on structural changes (MITI 1988a).
Figures for the period 1975-84 are based on past performance, while those
for the period 1990-95 are based on an analysis of current trends. Although
the industry classification does not reveal the current trends precisely, Table
9.3 provides a general picture of the changes taking place.
From "Heavy and Big" to "Light and Small"
Table 9.3 shows that the shares of primary, consumer, and basic materials
industries declined steadily from 1975 to 1984, and the decline is expected
to continue until 1995. Construction and other services sectors, which also
declined until 1984, are expected to either remain steady or increase their
shares. Machinery production increased rapidly during the period 1975-84
and will continue to increase at a decelerated rate. The increase in services
will continue at a faster rate,
MITI's industrial production index presents a more detailed picture of trends

Trade Conflicts and Structural Adjustment 203
by individual sectors (MITI 1988b). Within the basic materials sector, metals
and chemical fertilizers recorded severe declines. Aluminum production
declined from 1,091 tonnes in 1980 to 41 tonnes in 1987. Crude steel produc­
tion decreased by 12 percent, urea'production by 59 percent, and ammonia
production by 29 percent; during the same period. Ethylene, representing
petrochemicals, decreased by 14 percent from 1980 to 1982, but recovered and
surpassed its 1980 level by more than 10 percent. Color film increased by 154
percent.
Within the machinery sector there have also been mixed performances. Ship­
building declined by 71 percent and steel structures by 16 percent during the
Table 9.3
Changes in the Industrial Structure
(% composition of output at current prices)
1975 1980 1984 1990 1995
Primary
(agricullure, forestry, and fishery) 3.9 3.0 2.7 2.1 1.7
Consumer
(food processing, textiles and
clothing, paper, and
miscellaneous) 14.9 13.7 13.0 10.9 10.0
Basic materials
(metals, metal products,
chemicals, coal and petroleum
products, and ceramics) 16.0 16.7 14.3 11.6 10.6
Machinery
(industrial, electrical, transpor­
tation, and precision) 12.3 13.7 16.6 16.6 17.0
Construction
(construction and mining) 10.7 10.6 8.6 9.6 9.4
Services
households and firms (advertis­
ing, information, and leasing)
and public services (education,
medical, and governmental) 14.1 15.1 16.9 21.3 23.7
Other services
(utilities, transportation,
commerce, and finance) 28.1 27.2 27.9 27.9 27.6
Note: Figures for 1990 and 1995 are forecasts based on an analysis of the trend of changes
described in the text.
Source: MITI (1988a).

204 Role of Government in Industrial Growth
period between 1980 and 1987, whereas automobile production expanded by
46 percent. Staggering increases have occurred from 1980 to 1987 for such
items as watches (133%), computers for general use (153%), industrial cir­
cuitry (267%), VCRs (525%), and industrial robots (681%). Within the con­
sumer industries, foodstuff and textile production decreased, while new
products increased as much as in other sectors.
The increasing share of services reflects the move toward "a services econ-
omyj* but changes in its content should be examined carefully. The share of
household expenditures as a percentage of total expenditures on services such
as restaurants, barbers, education, and medical care increased from 28.3 per­
cent in 1975 to 35.5 percent in 1986, which partly corresponds to the rising
expenditures on commodities (consumer industries). It should be noted that
the share of intermediary services, which include consulting, leasing, and in­
formation services that support business activities, increased rapidly as a per­
centage of total expenditure. Inputs of these support services tend to increase
in all industries as the industries expand, but service input coefficients are
twice as high in the machinery industries as compared to the consumer and
basic materials industries. The increase in the service input coefficient in the
machinery industries is expected to continue and accelerate. The fact that sup­
port services formerly provided within individual firms are now procured from
outside companies by the firms partly explains the increase in the service in­
put coefficients.
The structural changes mentioned above are often described as changes
from "heavy, thick, long, and big" to "light, thin, short, and small." This
implies a shift in production from Heavy and Big (H&B) industries such as
basic materials to Light and Small (L&S) industries such as household and
office equipment and services. It also implies a shift in consumer tastes at
the higher income level toward more sophisticated and differentiated goods
and shorter cycles of fashion changes.
Table 9.4 shows changes in the export and import structure. Although it
only covers commodities, the classification corresponds to that of Table 9.3.
The rapid decline of metals exports and the equally rapid increase of
machinery exports during the period 1980-87, and especially 1985-87, should
be noted. The four individual machinery items listed together increased from
9.0 to 32.1 percent during the period 1970-87, accounting for four-fifths of
the increase of machinery's share.
In imports, raw materials decreased steadily over the whole period, while
fuels increased until 1980 and then rapidly declined after 1985. The increase
in imports of other manufactured goods, consisting of such standardized
products as textiles and steel, was evident after 1980 and has been growing
more rapidly since 1985. The combined share of chemicals, machinery, and
other manufactured goods as a percentage of total imports—the manufac-

Trade Conflicts and Structural Adjustment 205
tured imports ratio—increased at an accelerating rate in the 1980s to 44 per­
cent. Both the increase in manufactured imports and the decrease in imports
of raw materials and fuels correspond to the decrease in domestic produc­
tion of the consumer and basic materials industries and are related to the
shift from H&B to L&S industries.
Retrenchment of "Heavy and Big" Industries
H&B were the leading industries in the high-growth period during which
they achieved economies of scale and cost reductions, thereby acquiring in­
ternational competitiveness. However, the price increases of oil and raw materi­
als decreased their competitive edge, while slower economic growth after the
Table 9.4
Changes in the Export and Import Structure
(% shares of total exports and imports)
1970 1975 1980 1985 1987
Export structure
Textiles 12.5 6.7 4.9 3.6 3.0
Chemicals 6.4 7.0 5.2 4.4 5.1
Metals 19.7 22.4 16.4 10.5 7.9
Machinery 46.3 53.8 62.8 71.8 74.6
Office equipment 1.7 1.4 1.8 4.4 6.3
Semiconductors 0.4 0.8 1.8 2.7 3.6
VCRs — — 1.5 3.8 2.6
Automobiles 6.9 11.1 17.9 19.6 19.6
Import structure
Foodstuffs 13.6 15.2 10.5 12.0 15.0
Raw Materials 35.4 20.1 16.9 13.9 14.7
Fuels 20.7 44.3 49.8 43.1 26.2
Chemicals 5.3 3.6 4.4 6.2 7.9
Machinery 12.2 7.4 7.0 9.6 12.8
Office equipment 1.7 0.9 0.7 1.2 1.5.
Manufactured goods 12.8 7.4 11.4 15.2 23.4
Iron and steel 1.5 0.3 0.6 1.1 1.7
Textiles 1.7 2.3 2.3 3.0 5.1
Manufacturing subtotal 30.3 18.4 22.8 31.0 44.1
Note: Manufacturing subtotal is the sum of chemicals, machinery, and manufactured goods.

206 Role of Government in Industrial Growth
oil shocks forced a low utilization rate of their existing capacities. Interest
payments on the underutilized capacities added financial burdens to the H&B
firms, so that many opted to reduce existing capacities.
Table 9.5 shows the reduction of iron- and steel-making capacities in the
steel industry. Since 1970, five large steel mills have introduced integrated
production, utilizing both blast furnaces and converters to produce steel, while
smaller mills operate only converters or electric furnaces of a much smaller
size. Prior to the first oil shock, Japanese steel mills had planned to increase
their capacities, and some furnaces were, in fact, completed after 1973. But
the low utilization rate of furnaces led to a reduction in iron and steel produc­
tion capacities beginning in the late 1970s and intensifying in the 1980s. The
reduction was most severe in iron production—the primary processing of ores
and scrap iron. Since Japanese iron producers relied entirely on imported
materials, their competitive edge was badly affected by the increased price
of fuel. Electric steel mills nearly ceased iron production, while integrated
steel mills eliminated 19 furnaces and reduced their capacity by 25 percent.
On the other hand, in steel production, where Japanese mills have the ad­
vantage of producing high-tech finished steel and have a large domestic mar­
ket, the number of furnaces decreased but the capacity of those remaining
fell only slightly, as old, obsolete furnaces were abandoned in favor of new,
efficient ones.
Table 9.5
Reduction of Iron and Steel Production Capacity
1970 1975 1980 1985 1987
Iron making
Blast furnaces
No. of furnaces 62 69 65 54 50
Capacity ('000 tonnes/yr.) 76,545 120,308 136,245 123,672 103,064
Electric furnaces
No. of furnaces 23 20 10 5 2
Capacity ('000 tonnes/yr.) – 227 224 94 67 29
Steel making
Converters
No. of furnaces 83 98 94 85 85
Capacity ('000 tonnes/yr.) 91,480 125,319 129,967 123,976 124,080
Electric furnaces
No. of furnaces 739 705 626 555 514
Capacity ('000 tonnes/yr.) 13,599 24,972 28,757 28,389 28,079
Source: MITI in Bank of Japan (1988). Keizai Tokei Nempo (Economic Statistics Annual 1988).

Trade Conflicts and Structural Adjustment 207
From November 1986 to March 1987, the five biggest integrated steel mills
announced in succession their plans to further reduce capacity and person­
nel in order to adjust to the crude steel production target of 90 million tonnes
per year. Seven blast furnaces were to be demolished and 52,000 employees
out of a total 191,000 were to be laid off or moved elsewhere in 1988-89.
Another cost reduction measure adopted by Japanese steel mills was the
more economic use of fuels in the various stages of operation. The energy
input per tonne of crude iron ore was reduced steadily from 5,290,000 kcal
in 1975 to 4,783,000 kcal in 1980, and to 4,258,000 kcal in 1985. The switch
from costly oil to coal was also promoted. In blast furnaces especially, oil was
almost totally replaced by blast and coking gases.
A sharp reduction of capacity and personnel was similarly carried out in
the aluminum, shipbuilding, chemical fertilizer, and petrochemicals indus­
tries. As in iron production, competitiveness in these industries had deterio­
rated due to the increased costs of raw materials and fuels, and very low
operation rates under stagnant demand. 1
Exploring New Frontiers for L&S
The shift toward L&S is most apparent in the expansion of the machinery
industry and in services. Free from the disadvantages of H&B (the large-scale
use of energy and resources under increased energy prices and slower growth),
L&S has vast technical possibilities in responding to the great variety of de­
mand and changing consumer tastes for more sophisticated equipment. In­
stead of mass producing a small variety of goods, the L&S approach envisions
small-lot production of a large variety of goods. Production must be achieved
without increasing costs and with a shorter lead time. Fortunately, the Japanese
have a lot of experience and expertise, an established machinery industry, and
a large, high-income domestic market in which to test new products.
Not all machinery meets the L&S requirements. Only the electronics, au­
tomobile, and some industrial and precision machine industries were able to
develop new frontiers and expand production rapidly. Once the technology
is standardized for mass production, the Japanese competitive edge is easily
eroded and lost to the neighboring NIEs. This has been the case for house­
hold electronics and compact automobiles. The domestic market, with its vast
spending power and sophisticated tastes, has enabled Japanese manufacturers
to explore and test new products. The macroeconomic policy mix shift in 1986
toward boosting domestic demand has accelerated this, with such products
as digital audio tape-players (DATs) and furniture components designed for
limited room space appearing in stores.
The development toward L&S requires a high precision control system and
the effective use of market information. This has, in turn, resulted in an in-

208 Role of Government in Industrial Growth
crease in demand for information and lease services, as mentioned above. 6
These services were originally provided by service departments within the pi­
oneering firms. As the demand for the services by other organizations grew,
the service departments became independent businesses specializing in these
services. In conventional statistics, since these new businesses produce "ser­
vices to firms," they are classified as such, thereby increasing the output of
the services sector.
The restructuring toward L&S is also active in other industries such as tex­
tiles and clothing. Clothing production involves the assembly of textiles and
accessories and is very sensitive to rapid market changes. In the clothing in­
dustry, manufacturers-cum-wholesalers do the planning and marketing. They
purchase fabrics, design and cut the fabrics using computers, subcontract the
sewing, and ship the finished clothes to markets. They create a basic design
and fashion, produce more than a dozen varieties using different colors and
sizes, and supply the market with those varieties most in demand. Major cloth­
ing producers have at their disposal highly automated sewing factories or sub­
contracted firms both in Japan and in neighboring countries. Clothing
producers try to coordinate production and sales so as to minimize their in­
ventory (the "just-in-time" system). Parent companies and their subsidiaries
are linked by a nationwide network of highway and express delivery systems
(taku-kyu-bin) that provide quick, reliable, and economical transportation
of materials and finished goods in small lots. This departs from the old im­
age of labor-intensive clothing production. In a textile industry policy report,
the move toward L&S was emphasized as a major factor in the industry's
revitalization.
The shift to L&S became evident in the 1980s. Although statistics with con­
ventional classifications do not show this, evidence can be found in the in­
creased use of semiconductor/industrial circuitry, a basic component of
microelectronics. The capacity of semiconductors has improved almost a
thousandfold. At the same time, prices of semiconductors have been halved
over the past decade, enabling their use in industrial and transportation
machinery as well as in electronics.
Further evidence can be found in the increased use of point of sales (POS)
among the retailers of consumer products. In the POS system, cash registers
read the Japanese Article Number (JAN) code attached to each article sold.
Sales figures can then be compiled and trends analyzed automatically. The
JAN code system was begun in 1978, and by 1982 the number of registered
producers and retailers equipped with POS registers was 217 and 78, respec­
tively. However, by 1987, these figures had risen dramatically to 26,438 and
11,711 and include all of the major producers and retailers.
Other areas of exploration (other than microelectronics) are new materials
technology (such as fine ceramics and metals with special properties) and bio-

Trade Conflicts and Structural Adjustment 209
technology (such as medicine and agriculture). Their industrial applications
have not been readily identified in conventional statistics. The MITI report
predicted, however, that by 1995 their potential market size would amount
to US$460 and US$53 billion, respectively, compared to US$1.3 trillion for
microelectronics (MITI 1988a).
Globalization of Firm Activities
During the macroeconomic changes of the 1980s, Japanese industries and
firms had to abandon their conventional strategy of domestic production for
export, and turned to exploring new frontiers, such as L&S production, at
home. Another response to the macroeconomic changes of the 1980s was the
relocation of production abroad, and selling the products in the host mar­
ket, exporting to a third market, or importing the product back to Japan.
As Figure 5.5 shows, many Japanese industries were at the export and ma­
ture stages as of 1980, but in the following years some export industries were
forced to the mature stage, and some mature industries to the reverse import
stage. CPC development was accelerated in these industries so that direct for­
eign investment (DFI) promoted the restructuring of the Japanese economy.
Further analysis is needed to determine the interrelated decisions of both
domestic and external redirection in individual industries and their represen­
tative firms, but the external redirection of Japanese firms can be classified
by means of the aggregate surveys by the Ministry of Finance and MITI.
Japanese DFI, which had earlier been subject to restrictive rules to keep
domestic savings in Japan, expanded after the investment rules and regula­
tions were relaxed during the period 1967-70. The first boom in DFI occurred
in 1973 (US$240 million), and increased again in the late 1970s after the econ­
omy recovered from the first oil shock. DFI exceeded US$10 billion in 1984,
and according to Ministry of Finance statistics, it rose again to US$32 bil­
lion in 1986-87.
In which industries and to which countries was DFI directed? Throughout
the 1960s, Japanese DFI was directed to the development and procurement
of such resources as petroleum, lumber, and minerals. During the first in­
vestment boom in 1973, Japanese DFI was directed to the relocation of labor-
intensive manufacturing to Southeast Asia because of increasing labor costs
at home. In that year, the Asian share of total Japanese DFI was 28.6 per­
cent; of total Japanese DFI in manufacturing, 32.5 percent; and of total
Japanese DFI in textiles, 58.6 percent.
In the 1980s, however, both the destination and industry composition of
Japanese DFI changed. In recent years, the direction of Japanese DFI has
shifted toward the developed countries. The shares of Japanese DFI to the
United States and Europe made up 40.8 percent and 15.5 percent of total

210 Role of Government in Industrial Growth
Japanese DFI, respectively (in terms of values of investment in fiscal year
1986, ending March 1987). For Latin America and Asia, respectively, the shares
were 21.2 percent and 10.4 percent. In terms of industry composition, 17.1
percent of Japanese DFI was in manufacturing, of which 25.9 percent was
in electric and electronic goods, 21.8 percent in automobiles, and only 1.7
percent in textiles. Greater shares were occupied by the nonmanufacturing
sectors, such as finance and insurance (32.4% of total DFI), commerce (8.3%),
and other services (7.0%).
The latest official statistics do not, however, cover 1987-88, which is when
the second DFI boom occurred. Nevertheless, the shift in the direction of
Japanese DFI has been reported in the newspapers. The Nikkei Data Base
was employed to provide reports on the overseas projects of Japanese firms
from the four newspapers of The Japan Economic Journal Group (including
those specializing in reporting on industry and distribution), which were then
used to analyze trends. Owing to the use of key words such as overseas ac­
tivity, international specialization and DFI, the data include other forms of
activities such as OEM (Original Equipment Manufacturing—that is, procur­
ing finished products from other firms and selling them labeled with the brand
names of the wholesaler or retailer) and production agreements, technical
cooperation, and other business cooperation, which are excluded from offi­
cial DFI statistics because there is no participating equity ownership.
Table 9.6, which is divided into two parts, summarizes the globalization
of Japanese firm activities by the number of projects. Part A describes
Japanese DFI by location and industry, and Part B by location and type of
business activity as well as periods (three months for the first period and six
months for those thereafter). The classification of the data into regions re­
veals that globalization has proceeded in all Japanese activities.
Part A in Table 9.6 shows that North America and Western Europe were
the focal points of Japanese activities, with shares of 42.5 percent and 21.6,
respectively. Pacific Asia's (Asian NIEs, ASEAN, and China) share was 33.6
percent. Japan's overseas activities were concentrated in North America,
Western Europe, and Pacific Asia. Overseas activities are concentrated in
manufacturing, in which electronics, automobiles, and chemicals have the larg­
est shares; textiles has only 5 percent. Nonmanufacturing includes distribu­
tion/trading companies, finance, information, and other services. The regional
composition of individual industries in manufacturing and nonmanufactur­
ing are similar, implying that Japanese commercial activities are distributed
in proportion to the size and prospects of the individual regions. However,
the industry composition from one region to the next sometimes differs,
reflecting the different emphasis given by Japanese firms to individual regions.
In North America, a much greater share of overseas projects is in informa­
tion and above average shares are in other services and automobile and chem-

Trade Conflicts and Structural Adjustment 211
icals production. In contrast, Western Europe has larger shares in textiles,
electronics, and distribution. The shares of the Asian NIEs are nearly equal
to those of Western Europe. All ASEAN members, China, and Oceania have
larger shares in food processing.
In Part B of Table 9.6, Japanese firms are shown to be as active in other
types of business as they are in DFI (i.e., local production), and their activi­
ties in other areas in later periods have tended to increase (except in technical
cooperation). Whereas North America had the largest shares in all types of
business, followed by Western Europe in earlier periods, the Asian NIEs and
ASEAN have increased their shares significantly. Since 1986, the Asia-Pacific
share exceeds North America and Western Europe in production coopera­
tion and DFI. It should be noted that the share of Japanese DFI for the Asian
NIEs is now stagnant and has been surpassed by that of ASEAN in early
1988. Both North America and Western Europe have maintained their shares
in technical and other business cooperation.
DFI by Japanese firms in North America and Western Europe is directly
affected by trade conflicts and restrictions, whereas Japanese DFI in Asia
is affected by increasing costs in Japan due to the rapid appreciation of the
yen. However, there seems to have emerged a clear change in the global strate­
gies of Japanese firms. They will not limit themselves to operating domesti­
cally as they have traditionally, even if the conflicts are resolved and the yen
depreciates.
Japanese manufacturers have moved simple, labor-intensive production and
production of standardized items to other countries in Asia. The products
are then exported directly to other countries or imported back to Japan. This
relocation of manufacturing to neighboring Asian countries has been made
possible by the specialization of products and production processes in elec­
tronics, precision instruments, and automobile parts. The manufacture of
sophisticated items in Japan and parts requiring specialized machines and
technology-intensive processes is now a priority, and production of such
products has been growing. The division in labor between Japan and its de­
veloping Asian neighbors has resulted in rapidly increasing trade in parts and
intermediate products at various production stages between Japan and the
Asian NIEs and ASEAN countries. For example, cheap clothing is now im­
ported from Asia while specialty fabrics manufactured by high-tech produc­
tion processes are supplied by Japan. While this has contributed to a rapid
net increase in manufactured imports to Japan, it also indicates an increas­
ing import dependence of Asia-Pacific countries on Japan. This network of
specialization has expanded not only as a result of Japanese firms, but also
as a consequence of CPC development and the establishment of an indus­
trial base in the Asia-Pacific region.
Overseas activity of Japanese firms in automobile assembly and electron-

Table 9.6
Globalization of Japanese Firm Activities (by number of projects)
A. INDUSTRY
Region
North Asian World
Industry America Europe NIEs ASEAN China Oceani a total 3
Manufacturing
Foodstuffs 136 62 64 43 45 13 375
(36.3) (16.5) (17.1) (11-4) (12.0) (3-5)
Textiles 71 112 70 16 48 4 329
(21.6) (34.0) (21.3) (4.9) (14.6) (1.2)
Chemicals 317 144 111 48 33 4 668
(47.5) (21.6) (16.6) (7.2) (4.9) (0.6)
Electronics 513 285 260 91 42 2 1,228
(41.8) (23.2) (21.2) (7.4) (3.4) (0.2)
Automobiles 337 115 116 45 20 16 698
(48.3) (16.5) (16.6) (6.4) (2.9) (2.2)
Subtotal 2,084 1,153 958 371 334 66 5,165
(40.3) (22.3) (18.5) (7.2) (6.5) (1.3)

Industry North
America
Nonmanufacturing
Distribution and trading 105
(30.2)
Finance 110
(39.7)
Information 175
(67.6)
Other services 208
(52.1)
Subtotal 846
(48.9)
Total 2,930 Region
Asian
Europe NIEs ASEAN
93 79 20
(26.7) (22.7) (5.7)
67 57 6
(24.2) (20.5) (2.1)
41 28 2
(15.8) (10.8) (0.8)
73 45 13
(18.3) (11.3) (3.3)
338 456 65
(19.5) (26.4) (3.8)
1,491 1,414 436 World
China Oceania total 3
26 8 348
(7.5) (2.3)
21 8 277
(7.5) (2.8)
11 1 259
(4.2) (0.3)
39 12 399
(9.8) (3.0)
135 66 1,729
(7.8) (3.8)
469 132 6,894

Table 9.6. Continued.
B. TYPE OF ACTIVITY
Activity/period 1* North
America Europe Asian
NIEs Region
ASEAN China Oceania World
total 3
Local production (DFI)
9-12/1985
1-6/1986
7-12/1986
1-6/1987
7-12/1987
1-6/1988
Production agreements
9-12/1985
1-6/1986
7-12/1986
1-6/1987
7-12/1987
1-6/1988 86
158
110
154
192
179
21
31
39
28
54
61 28
64
52
68
83
63
15
11
18
16
35
59 18
75
92
133
108
80
4
13
36
46
38
38 20
18
21
59
76
90
0
4
6
4
7
17 23
23
19
18
18
20
7
5
13
7
13
21 184
369
327
464
508
462
50
68
116
103
152
206

Region
Activity/period b North Asian World
Activity/period b America Europe NIEs ASEAN China Oceania total 3
Technical cooperation
9-12/1985 21 22 16 3 18 6 94
1-6/1986 50 33 26 14 38 3 180
7-12/1986 76 47 32 6 20 0 199
1-6/1987 61 20 50 10 23 5 181
7-12/1987 93 37 32 11 15 6 200
1-6/1988 82 51 25 6 17 11 185
Other business cooperation
9-12/1985 89 58 19 7 10 5 193
1-6/1986 165 102. 22 3 15 6 321
7-12/1986 185 95 34 3 16 10 357
1-6/1987 213 85 41 5 21 8 378
7-12/1987 187 126 53 7 27 10 415
1-6/1988 243 155 52 17 24 10 509
Notes:
a. Columns do not total due to the omission of rest of world.
b. For each activity, the first period covers three months; the remaining periods cover six months.
Source: Overseas Activities of Japanese Firms (OAJF) Project, Hitotsubashi University, Tokyo, compiled from Nikkei Data Base, September 1, 1985-June
30, 1988.

216 Role of Government in Industrial Growth
ics production began in the United States and the EEC countries with the
rising number of restrictions on Japanese exports of these goods. At present,
60-80 percent of the total value of the output of these Japanese subsidiaries
is produced in the host country to satisfy local production content and em­
ployment creation requests. Overseas production tends to incur additional
production costs, but it also enables Japanese firms to keep up with chang­
ing demand and to explore new frontiers in the American and European mar­
kets. A MITI survey on the motivations behind overseas operations by
Japanese firms endorsed the positive aspects of overseas production. 7 Accord­
ing to the survey, the most important incentives for Japanese firms engaging
in DFI in the United States and Europe are:
Degree of Importance of
Motivation Factor as
Measured by the Percentage
Motivation Factor of all Firms Responding 3
United States Europe
• Fear of trade conflicts 15.3 21.2
• Response to trade restrictions 12.5 18.2
• Sales expansion in local market 86.1 17.2
• Acquisition of technology and market information 12.5 21.2
a. Respondents were asked to give the three most important factors.
Seventy to eighty percent of Japanese firms leave marketing and procurement
of parts and materials in local production to their overseas subsidiaries, and
a few firms have established R&D facilities overseas. A deluxe television wall
set, developed at a Japanese subsidiary operating in the United States for
American consumers and now imported to Japan, exemplifies the globalized
activities of Japanese firms in recent years. Some major Japanese firms have
already divided the world into three regions—North America, Europe, and
Asia—and have established headquarters in each area.'
The globalization of activities is not confined to. Japanese firms. Ameri­
can firms began to globalize in the 1960s, and were followed by the Japanese.
Helleiner (1981) described the global operations of American firms as "in-
trafirm trade." The share of trade between U.S. parent firms and their sub­
sidiaries and between subsidiaries under a common parent firm (including
overseas subsidiaries), was 48.8 percent of total U.S. exports to the world in
1983. The share of intrafirm trade for Japan reached 34.0 percent in 1986
(MITI 1988c). The growing significance of intrafirm trade reflects the changing
nature of international trade from transactions between independent firms
to those within a single firm across borders. Diversified forms of interna­
tional specialization arise, with firms retaining technology and management

Trade Conflicts and Structural Adjustment 217
know-how within their organizations. The development of telecommunica­
tions has enhanced firms* abilities to adopt a global strategy and thereby
spread capital and technology worldwide. Whether trade restrictions urged
by lobbyists and politicians can continue to work alongside the globalization
of business activities is of significant interest.
Foreign Firms Operating in Japan
How has this universal trend of globalization affected the Japanese mar­
ket? That is, how have foreign firms entering the Japanese market affected
the restructuring of Japan's economy and industries?
A MITI survey of foreign firms operating in Japan (that is, firms with more
than 50% of its equity shares owned by foreigners) as of the end of March
1985 found that while DFI in Japan has risen much like Japanese DFI abroad,
the level of DFI in Japan remains less than Japanese DFI abroad (MITI 1988c).
From US$500 million in the 1970s, DFI in Japan has increased to US$1 bil­
lion in the 1980s. The main sources of DFI in Japan (in terms of the number
of firms as of March 1985) are the United States (48.5%), Europe (35.6%)—
which includes West Germany (7.7%), the United Kingdom (7,4%), Switzer­
land (5.9%), and France (4.7%)—and Asia (11.1%). The breakdown of DFI
in Japan by industry is 73.5 percent for manufacturing, with petroleum refining
(40.4%), chemicals (12.8%), and electrical machinery (9.8%) accounting for
the largest shares. Typically, foreign manufacturing firms import materials
and parts and sell their products in the Japanese market. This practice is
reflected in the high import/sales ratio of all foreign firms operating in Japan
(55.9%) and the low export/production ratio, which was 5.9 percent.
Although the restrictions on DFI in Japan were loosened during the pe­
riod 1968-73, foreign firms should be encouraged to continue operating in
Japan. With DFI, new technology and management are introduced and com­
petition is increased, thereby accelerating Japan's restructuring. In the fu­
ture, Japan should welcome the Asian NIEs' DFI in manufacturing/marketing
and more American and European investment in finance and information
services.
AN ASSESSMENT
OF GOVERNMENT INDUSTRIAL POLICIES
The Japanese government's policies have affected the restructuring of
Japanese industries and firms by providing the framework in which the re­
structuring has taken place This section reviews the industrial and trade poli­
cies that directly affect m/croeconomic behavior.

218 Role of Government in Industrial Growth
Adjustment Assistance to H&B Firms
The core of MITI's assistance after the first oil shock consisted of a series
of laws to assist depressed industries in their adjustment. For example, the
laws attempted to prevent the basic materials industries, which lagged behind
in the recovery after the first oil shock and whose output growth was less
than half that of the machinery industry, from impeding those sectors of the
economy that were performing well. The laws were enacted to assist depressed
industries in reducing their capacities to a manageable size. The 1978 Law
of Temporary Measures to Stabilize Specific Depressed Industries provided
for the implementation of an adjustment assistance program. 9 The program's
plans were as follows:
1. Fourteen industries were designated as qualified for adjustment assistance
for five years from 1978 to 1983. They included electric furnace steel mak­
ing, aluminum smelting, ferrosilicon manufacturing, shipbuilding, the
manufacture of four synthetic fibers, three chemical fertilizers, and the
-manufacture of cotton, wool, and paperboard.
2. For each industry, a Basic Stabilization Program was established jointly
by the industry, based on industry-wide consensus, and MITI to identify
the extent of excess capacity by forecasting demand and supply.
3. The actual decrease of excess capacity was left to individual firms, with
some administrative guidance probably coming from the industry's trade
association. Eight industries were instructed by MITI to form a cartel
{fukyo cartel) with the aim of reducing excess capacity. 10
4. Financial assistance was given to encourage the adjustment. A special loan
to aid the shift to other activities was provided by the Development Bank
of Japan. Subsidies were available to pay the cost of retraining surplus
workers in H&B production and of reemploying them in other production
sectors.
The Smaller Business Switchover Law was enacted to assist the adjustment
of industries composed primarily of small and medium-sized firms. Twenty-
one industries were designated as depressed and were provided with the finan­
cial adjustment assistance indicated in item 4 above for the ten-year period
1976-86.
The impact of these laws on the selected industries was dramatic For ex­
ample, excess capacity in aluminum production was reduced by 56.7 percent,
in urea production by 44.9 percent, in shipbuilding by 35 percent, and in am­
monia production by 26.1 percent. MITI reported that by 1980 most indus­
tries had achieved their goals, increased their capacity utilization rates, and
improved their management (Peck, Levin, and Goto 1987). However, the im-

Trade Conflicts and Structural Adjustment 219
provements were achieved mainly through the recovery of market demand
and not by any improvement in management structure as the law intended.
As a result, these industries suffered again from structural depression after
the second oil shock and appealed for additional help to further decrease
excess capacity.
Following the second oil shock, adjustment assistance continued under the
1983 Temporary Measures for the Structural Adjustment of Special Indus­
tries Law. Eleven out of the 14 industries that were protected under the 1978
Law of Temporary Measures to Stabilize Specific Depressed Industries and
15 new industries, including ethylene production, qualified for continued ad­
justment assistance under the 1983 law. Again, most industries achieved the
capacity reduction goal and many cartels were cancelled before the target year.
The 1978 and 1983 laws were generally successful in their employment ad­
justments, as few workers were displaced. Despite strong opposition to the
continuance of adjustment laws for depressed industries, the Special Mea­
sures for Facilitating Adjustment Law, the third in the series, was enacted
in 1987 to assist industries such as chemical fertilizers and ferro-alloys.
The adjustment laws for depressed industries have been both a success and
a failure. MITI claimed that the laws were consistent with the'OECD's criteria
of a positive adjustment policy in that the policy measures were well defined,
were transparent, and had a time limit for their implementation. In addition,
no trade restrictions were imposed. While the policies were consistent with
the OECD definition of a positive adjustment policy, their implementation
was not in several respects. First, administrative guidance did not clearly en­
courage individual firms to decrease their capacities, and repeated adjust­
ment assistance contradicted the idea of a time limit. Second, forecasts of
supply and demand, the main outcome of the Basic Stabilization Program,
often differed from actual figures during worldwide fluctuations. As a result,
individual firms modified their capacity decreases according to the actual
figures; thus, capacity reduction was inefficient. This argument is supported
by Peck, Levin, and Goto (1987) who found that market shares remained un­
changed in concentrated industries even after uniform capacity reduction
across industries was accomplished. Modifications to the 1987 law allowed
decisions on capacity decreases to be left to individual firms without the
benefit of a Basic Stabilization Program and cartels.
Nevertheless, the adjustment laws provided the framework for capacity
reduction. In fact, ethylene-producing firms continued to suffer from excess
capacity until 1983 when the industry was included under the 1983 law and
a cartel formed for capacity reduction. Employment adjustment was so suc­
cessful throughout the economy that only a limited number of surplus work­
ers were displaced in the H&B industries, despite major capacity adjustment.

220 Role of Government in Industrial Growth
Industry-Specific Adjustment
Along with the adjustment laws for depressed industries, industry-specific
adjustment assistance was provided for declining industries such as agricul­
ture, coal, and textiles. The ministries responsible for each particular indus­
try established advisory councils that included representatives from consumer
advocacy groups, the media, and academia, as well as from industry and trade
unions. Each of the advisory councils reviewed their industry's supply and
demand prospects, studied the adjustment difficulties of the industry, and
recommended the adjustment assistance needed for the industry. The as­
sistance was then implemented by the respective ministry in cooperation with
prefectural governments. The amount of public funds, the policy recommen­
dations for adjustment assistance, and the role played by government in the
adjustment process differed greatly between industries.
Coal and cereals have relied heavily on government assistance in their ad­
justment. Domestic prices for these commodities were maintained at a level
that was much higher than the prices of imports under the strict import quota
restriction. Domestic producers were provided with a variety of subsidies to
maintain or adjust production levels or cease production. Despite this heavy
protection, domestic production and employment have decreased dramati­
cally during the past quarter century, and further adjustments have recently
been needed as a result of the rapid appreciation of the yen. For example,
coal production and employment declined from 55 million tonnes per year
and 260,000 miners in the peak year of 1961 to 17 million tonnes and 26,000
miners in 1987. Another halving of production and employees is planned un­
der the current Eighth Coal Policy (1987-91). For cereals, price supports for
rice are being considered and it is widely anticipated that the current ban on
rice imports will be modified so that rice can be imported under quota re­
strictions, with the quotas increasing gradually.
The policy on textiles is a guidepost in the shift to L&S. In the textiles in­
dustry, the adjustments have been focused primarily on market mechanisms
(Yamazawa 1988a) and on small and medium-sized weaving mills. Since 1953,
equipment in these firms had to be registered, and new equipment could only
be installed when the old equipment became obsolete. The subsidized scrap­
ping of 10-20 percent of existing capacity was repeated several times. However,
complaints about excess capacity, which apparently resulted from the steady
increase in productivity per machine and a lack of voluntary (unsubsidized)
scrapping, are still being voiced.
Competent medium-sized firms are going ahead with the reorganization
of the industry toward L&S, as evidenced by the clothing manufacturers men­
tioned earlier. This adjustment is necessary because of increasing competi­
tion with imports in the domestic market and competition between foreign
and Japanese clothing producers in international markets. The increasing com-

Trade Conflicts and Structural Adjustment 221
petition has been intensified by the rapid appreciation of the yen and the liberal
import policy for textiles (without Milti-Fibre Arrangement quota restrictions).
Promotion of L&S
The structural shift to L&S is widely recommended in many of MITI's
reports, and competent firms in all industries have begun to respond. However,
in the absence of "picking the winner" policies, the shift to L&S at present
is not as easy as was the implementation of programs for steel in the 1950s
and automobiles in the 1960s.
MITI can play only a guiding role and can provide only moderate finan­
cial assistance to small and medium-sized firms. The general promotion of
R&D has supported the shift toward L&S. Subsidies to universities and
research institutes in Japan are almost identical to those of other industrial­
ized countries. The amount appropriated for R&D has increased steadily
despite general budgetary constraints, but R&D still occupies only 3 percent
of the central government's budget, which is below that of the United States
(5%) arid West Germany (5%).
Japan is unique, however, in that the Japanese public and private sectors
jointly promote applied research. The National Research and Development
Program (often called the Large-Scale R&D Project), which has been in oper­
ation since 1966, is run by the Agency of Industrial Science and Technology
(AIST) (Tamura and Urata 1988). The program is organized as follows:
• It has accepted only high-risk and lengthy research projects that would
not have been undertaken by private firms alone.
• For each project, public-sector research institutes and 10-20 private firms
jointly organize an R&D association in which a public-sector institute acts
as coordinator.
• A project is usually provided with 10-20 billion yen (approximately
USS70-150 million in 1987 dollars) and takes 5-10 years to complete. The
results are made public and are available to any interested party on a royalty
basis.
• So far, 24 projects undertaken since 1966 show a clear shift from resource
development and conservation (such as offshore oil drilling, steel produc­
tion, manganese nodule mining, and desulfurization) to projects related
to electronics (supercomputers, pattern information processing, automated
sewing systems, and advanced robotics).
The National Research and Development Program has been successful so
far because the research projects have been well defined, and the results of
the basic scientific studies have been published. AIST will find it increasingly

222 Role of Government in Industrial Growth
more difficult, however, to conduct basic studies without the help of other
institutes and the private sector. Its noncompetitiveness may be criticized, but
this is certainly more than offset by the advantages of avoiding a duplication
of R&D efforts when the participants act independently of each other.
Trade Policy
Japan has responded to trade conflicts throughout the 1970s and 1980s by
reforming its trade policy. Market liberalization measures and export restraints,
which have already been mentioned, are the chief changes. These are exam­
ined below in the context of restructuring.
Market liberalization measures include tariff reduction, the abolishment
or expansion of import quotas, the improvement of standards and procedures,
and the liberalization of services trade (construction, finance, and legal).
Tariffs on manufactured goods were reduced substantially through unilateral
reductions in the 1970s and previous GATT multilateral trade negotiations
(particularly the Tokyo Round). In addition, tariffs on products from develop­
ing countries were exempted or halved under the Generalized Scheme of
Preferences (Yamazawa 1988b).
Many import quotas on agricultural products have been in effect for a long
period of time, and proposals to remove them have met strong resistance from
vested interests. Nevertheless, some headway has been made and restrictions
on animal skins and leather were removed in 1985. It was also decided at the
1988 Japan-US. negotiations that restrictions on beef and oranges would be
removed by 1991. The current strict import ban on rice has been criticized
both in Japan and abroad, but, as noted earlier, it appears that an import
quota system will be in place in the near future.
Standards and procedures governing imports were improved substantially
by the Action Program of 1986, and this should help to promote imports sig­
nificantly. Services fall under domestic regulations, which have been modi­
fied to encourage non-Japanese competition. These liberalization measures,
which are largely a response to pressures from Japan's trading partners, will
contribute to Japan's restructuring, increase competition in the domestic mar­
ket in the long term, and improve the efficiency of the Japanese economy
overall.
Voluntary export restrictions (VERs) have been imposed on many machinery
exports, and the manufacturers' efforts to circumvent these barriers have af­
fected the restructuring of L&S industries. At the request of foreign govern­
ments the Japanese government coordinates VERs with domestic industries
and firms. The total quota is negotiated and distributed among exporting
firms. Some VERs were introduced by MITI based on the Trade Administra­
tion Law, but many VERs were also introduced through MITI's administra-

Trade Conflicts and Structural Adjustment 223
tive guidance and coordination of interests among related parties. VERs can
settle fierce trade conflicts in the short term but will cause distortions in the
long term. Rigid share agreements among rival exporters tend to depress com­
petition and keep prices high, thus injuring consumers and industries using
the protected good as an input (for example, steel).
The intended effect of VERs gradually erodes. Firms planning to extend
their operations overseas readily relocate their export production either to
the export market or to third countries from which they export the products
in order to avoid restrictions. In fact, even though restrictions on exports of
Japanese cars to the United States began in 1980 and exports from Japan
decreased to just short of the export quota of 2.3 million cars, production
outside Japan gradually increased to 1.6 million cars for 1988.
A new element complicating Japan's trade policy reform is the increasing
demand by Japanese producers to restrict imports from developing countries.
Imports of knitwear, steel plating, and cement from the Asian NIEs have in­
creased so rapidly that domestic producers fear that the imports may under­
mine their efforts to decrease capacity when domestic demand stagnates.
With persistent, huge trade surpluses, however, the Japanese government's
attempt to placate these demands is limited. Unlike other developed coun­
tries, Japan has not resorted to imposing quota restrictions on textiles and
clothing imports, which is allowed under the Multi-Fibre Arrangement, despite
repeated requests by the Textile Trade Association. However, when the Knit­
wear Association initiated an antidumping suit against Korean producers in
1988, MITI negotiated with the Korean producers to adopt VERs. By pursu­
ing these restrictive measures, Japan has, in effect, endorsed similar measures
that the United States and EC have sought to invoke against Japan. VERs,
however, are short-term solutions, as import competition will eventually re­
sume either through increasing imports of the same product from other Asian
countries or increasing imports of more sophisticated products not covered
by the VERs.
VERs are trade restriction measures outside the scope of GATT rules, and
their abolition is being discussed at the Uruguay Round of multilateral trade
negotiations. Instead of relying on VERs, Japan should follow GATT Arti­
cle 19 and implement temporary safeguard restrictions against excessive im­
port increases. This policy would also be consistent with Japan's restructuring
efforts.
Hollowing Out of the Industrial Structure?
The shift fom H&B to L&S is inevitably accompanied by substantial changes
in employment. As mentioned above, employment in Japan was reduced con­
siderably in the H&B industries. Are the unemployed being absorbed by the

224 Role of Government in Industrial Growth
expansion of L&S? If L&S expands less than H&B contracts, then structural
unemployment will occur. This argument is used to support the case against
"deindustrialization" or "the hollowing out of the industrial structure" and
has generated fierce policy debate in Japan and the United States.
Table 9.7 shows the changes in the employment structure by industry from
1975 to 1984 and the projected figures for 1990 and 1995. Both the primary
and consumer industries show substantial decreases in employment from 1975
to 1995, but the basic materials industry remains unchanged due to the growth
of other sectors. In 1995, the machinery and other services categories will
have increased by 1 million from their 1975 levels. Employment in the con-
Table 9.7
Changes in the Employment Structure (millions of persons, %)
1975 1980 1984 1990 1995
Primary
(agriculture, forestry, and fishery) 6.63 5.76 5.13 3.87 3.10
(12.7) (10.4) (8.9) (6.4) (5.0)
Consumer
(food processing, textiles and
clothing, paper, and 6.16 5.65 6.05 5.74 5.34
miscellaneous) (11.8) (10.8) (10.5) (9.5) (8.6)
Basic materials
(metals, metal products,
chemicals, coal and petroleum 2.92 3.04 2.94 2.96 2.98
products, .and ceramics) (5.6) (5.5) (5.1) (4.9) (4.8)
Machinery
(industrial, electrical, transporta­ 4.39 4.65 5.36 5.26 5.40
tion, and precision) (8.4) (8.4) (9.3) (8.7) (8.7)
Construction
(construction and mining) 4.91 5.59 5.36 7.13 7.88
(9.4) (10.1) (9.3) (11-8) (12.7)
Services
households and firms (advertis­
ing, information, and leasing)
and public services (education, 11.28 14.39 16.03 18.49 20.5
medical, and governmental) (21.6) (26.0) (27.8) (30.6) (33.1)
Other services
(utilities, transportation, 15.93 15.49 16.78 16.98 16.94
commerce, and finance) (30.5) (28.8) (29.1) (28.1) (27.3)
Note: Figures for 1990 and 1995 are forecasts based on an analysis of the trend of changes
described in the text. Numbers in parentheses are percentages of column total.
Source: MITI (1988a).

Trade Conflicts and Structural Adjustment 225
struction industry will have increased by almost 3 million (60%), while ser­
vices employment will have nearly doubled. Although the growth of employ­
ment in the expanding sectors will be smaller or even negative under slower
growth, the employment increase in the L&S industries will be large enough
to absorb employment cuts in the H&B industries.
More important is the problem of matching the skill requirements of one
group with another. A survey of positions requiring skilled labor conducted
by the Ministry of Labor in November 1987 reported different rates of un­
filled employment opportunities in L&S industries due to a shortage of quali­
fied skilled laborers and an excess supply in steel and other basic materials
industries (Nihon Keizai Shinbun [The Japan Economic Journal], October
1988). The shares of vacant positions by industry were 13.5 percent for con­
struction, 9.1 percent for services, and 5.2 percent for manufacturing as a
whole. Skilled labor is industry-specific and thus not easily transferred from
one industry to another. To bridge the gap, improvements in retraining and
placement are necessary.
Toward Institutional Reform
Because restructuring efforts by individual firms and industries are con­
fined by the institutional framework within which they operate, institutional
changes are sometimes required to promote further restructuring (OECD
1987). To this end, the Japanese government has been trying to change specific
sectors of the economy through policy reform and has been working to change
its economic institutions. The Maekawa Group (a group commissioned in
1985-86 and devoted to the study of economic structural adjustment) and
the Administrative Reform Council (1981-83 and 1987-) were formed to make
recommendations on how to meet these challenges. Both groups, which in­
cluded opinion leaders in the private sector and former government officials,
were formed at the request of the prime minister. After intensive study and
discussion, the two groups submitted their recommendations to the prime
minister, who in turn committed himself to incorporating them into actual
policies. The recommendations mainly advocated changes in macroeconomic
policies and economic institutions.
The Maekawa Report recommended the following:
• Improved access to the Japanese market and the promotion of manufac­
tured imports
• Separation of fiscal policy management from the strict balanced budget rule
• Further promotion of the adjustment of declining industries
• The liberalization of financial and capital markets
• Furthering development cooperation with developing countries

226 Role of Government in Industrial Growth
The Administrative Reform Council's first report recommended reducing
the size of the central government by consolidating or abolishing ministries
and agencies, and privatizing three state-run enterprises (tobacco, telephone
and telegraph, and the railroad). Both recommendations were implemented
by the government. The Administrative Reform Council's second report in
1988 proposed administrative deregulation in seven areas (the distribution sys­
tem, transportation, information, the financial markets, oil, rice and other
cereals, and civil aviation).
Other forms of institutional reform include tax reform, which was promoted
by the Ministry of Finance, A low consumer tax was introduced over a wide
range of commodities and services to correct for the overly heavy reliance
on income taxes. Moreover, it is widely anticipated that some form of land
reform to correct distorted land prices is imminent.
The reports by the Maekawa Group and the Administrative Reform Coun­
cil, which were partly intended to soften criticism of Japan's persistent trade
imbalances, have given Japan's trading partners a sense of the direction of
Japanese restructuring. Implementation of the institutional changes recom­
mended in these reports has not been easy, not only because the changes are
opposed by groups with vested interests, but also because radical changes are
often difficult to implement. Thus, although the United States has expressed
its dissatisfaction with the pace of implementation of the recommendations
and has criticized the reports as paying lip service to serious problems, it must
also be realized that institutional reforms take time. The slow pace may reflect
the cautious attitude toward institutional reform, which is irreversible, but,
in the long run institutional reform will promote the restructuring of the
Japanese economy.
JAPAN'S NEW INTERNATIONAL ROLE
As the Japanese economy grows and the trade imbalances continue, Japan
will increasingly be called upon to play a greater role in promoting growth
and stability in the world economy. It appears that national consensus has
been reached in Japan on the country's new international role.
It has often been pointed out that Japan should recycle its trade surplus
in the form of DFI and official direct assistance (ODA), and encourage im­
port expansion through boosting domestic demand and market liberalization
measures. This chapter has described the commitment and efforts of the
Japanese government in pursuing these goals. Although trade imbalances can­
not be attributed to the policies and performance of a single country alone,
Japan has stepped forward and committed itself to playing a more active
role—and one that is befitting of Japan's economic stature—in the global
economy.

Trade Conflicts and Structural Adjustment 227
But the restructuring of the Japanese economy, which is already under way
in the formulation of government policies and the private sector's perfor­
mance, is also needed to achieve this goal. However long the restructuring
takes, it will eventually resolve the trade imbalance and contribute to the fur­
ther development of the Japanese economy.

Part IV
Japan's Development Experience and
Contemporary Developing Countries

Ten
Is the Japanese Model Applicable?
Does the Japanese model of industrialization offer anything useful for con­
temporary developing countries? International market conditions and initial
requirements for industrialization now differ considerably from those of the
late nineteenth century. Nevertheless, pre-Second World War Japan and con­
temporary developing countries share basic similarities in their industrializa­
tion processes and mechanisms that go beyond the differences in their initial
conditions. Thus, the Japanese model may have relevance for developing coun­
tries. After reviewing the Japanese development experience on both the
microeconomic and macroeconomic levels, two topics from preceding chap­
ters—general trading companies and CPC development in modern indus­
tries—are examined for their applicability to developing countries. 1
UNIVERSALITY VERSUS UNIQUENESS
OF THE JAPANESE EXPERIENCE
The model presented in this book is based on the industrialization process
of a late-developing country—Japan. Contemporary developing economies
now face some of the same problems of late-developing industrialization that
Japan faced as a late-developing country. These problems and the similari­
ties between Japan and contemporary developing countries are described in
this section.
When a country begins to industrialize, there exists a huge technological
gap between the developing country and the advanced industrialized coun­
tries. The gap for Japan was almost of an incomparable magnitude, unlike
that of late-developing European countries. For example, in the initial stages
of industrialization in the nineteenth century, both Germany and Great Bri­
tain had at that time a few industries, such as the chemical industry, that were
among the most advanced in the world. Unlike these countries, Japan had

232 Japan's Experience and Contemporary Developing Countries
to import each new industry, which then had to follow the typical catching-
up process.
Pre-industrial Japan and the developing countries today have in common
large, indigenous, traditional sectors, from which the industrialization process
receives significant support. These sectors are what makes up what is often
collectively referred to as Japan's non-Western initial condition and account
for the so-called dualistic development pattern. As explained below, the suc­
cessful move from light to heavy industrialization in the production and trade
structures could not have occurred without a flexible indigenous sector. Thus,
the industrialization processes of pre-Second World War Japan and some
of the contemporary developing countries are not fundamentally different.
This is especially true for developing countries that, like Japan, are lacking
in resources and, therefore, are primarily dependent on processing trade.
The differences in initial conditions should not, however, be overlooked.
First, the development of science and technology in the first three quarters
of this century has given rise to many new industries in the advanced indus­
trialized countries. This development has created a gap between industrial­
ized countries and developing countries today that is even wider than that
which existed between the industrial world and Meiji Japan. As a consequence,
more external resources are now required for the establishment of new indus­
tries. The wider gap has, however, also created more opportunities to utilize
accumulated knowledge, which may account, in part, for the faster overall
growth of the developing countries. 1
Second, the backward economies of East and Southeast Asia provided a
market for Japan when its new industries reached the export stage. Today's
developing countries, in contrast, have difficulty finding similar economically
backward markets for their exports. It may be argued that this and the devel­
oped countries' protection of their less competitive industries leave develop­
ing countries in an unfavorable position to expand exports.
In order to overcome these formidable obstacles and establish new indus­
tries, many developing countries have to date relied heavily on DFI. Private
sector experience and existing channels in export marketing can, however, be
useful assets in export expansion. Heavy reliance on DFI, which basically
is a package transfer of such scarce resources as capital, technology, and
management, may be regarded as an institutional device by which a late-
developing country can overcome obstacles to industrialization, as suggested
by Gerschenkron (1966). Japan's general trading companies, for example, were
established to help late-developing Japan expand its trade in competition with
its developed rivals. But Japan's reliance on DFI was limited to a few indus­
tries within a certain period of time.
Third, contemporary developing countries suffer from their colonial her­
itage. Although colonial rule, or quasi-colonial rule, brought about some
modernization, such as railways, it obstructed or even repressed the emer-

Is the Japanese Model Applicable? 233
gence of socioeconomic factors necessary for the development of an indepen­
dent national economy. The legacy of this heritage can still be seen in the
international relations and domestic affairs of developing countries. Indeed,
a feeling of antineocolonialism prevails in the developing countries, and this
has bred mistrust of market economy solutions. That feeling is manifested
most strongly with respect to DFI and prevents developing countries from
taking full advantage of this type of investment in the industrialization process.
Because of these differences in initial conditions and the international en­
vironment, the Japanese model cannot be applied to developing countries
in the 1990s and beyond without modifications. Nevertheless, it can provide
useful insights into the mechanism and problems of their catching-up indus­
trialization process.
CPC DEVELOPMENT IN DEVELOPING COUNTRIES
IN THE 1990s AND BEYOND
CPC development is an appropriate strategy of industrial development for
a late-starting industrializing country with a domestic market of a certain
size and potential comparative advantage in industrialization. First, domes­
tic demand for a new product is created by imports; efficiency in domestic
production is then improved through import substitution; and finally, exports
increase as the product becomes competitive in foreign markets. If import
substitution is slower than expected, it is either because the country lacks com­
parative advantage in the industry or because improvements in production
efficiency are obstructed. The CPC development model provides a framework
by which the development performance of individual industries can be as­
sessed. Of course, it should take into account different international market
conditions and greater participation of multinational corporations (MNCs)
in the developing countries.
CPC Development in Asian Developing Countries
The promotion of industrialization and the introduction of modern indus­
tries began in East and Southeast Asia during the 1950s and 1960s. Due to
their potential comparative advantage in labor-intensive textile production
and their high domestic demand for imported textiles, the countries of the
two regions introduced the textile industry as their first modern industry. In
Taiwan and Korea, for example, the textile industries were introduced in the
1950s and by 1970 the industry in these two countries had achieved import
substitution and were beginning export expansion. In the 1970s, they proceeded
to adopt import substitution and export expansion in the heavy industries
such as shipbuilding and steel making.

234 Japan's Experience and Contemporary Developing Countries
The ASEAN member countries began industrializing in the 1960s. In
Thailand, the modern textile industry began with cotton and then synthetics
in the mid-1960s, and achieved import substitution and began export expan­
sion in the 1970s. Development of Indonesia's modern textile industry began
in the late 1960s and the industry achieved import substitution in the 1970s.
Figure 10.1 illustrates the CPC development of both synthetic fabric and
crude steel production for four Asian countries. Production/demand ratios
are indicated on the vertical axis. The increase in these ratios beyond unity
shows the attainment of import substitution and export expansion for in­
dividual industries (also see Figure 2.1, Panels B and C). Rising production/de­
mand ratios were observed for the two industries and in all four countries,
which supports the role of CPC development of the two major industries
in these countries. But CPC development occurred at different times for the
two industries in all four countries and among the four countries for each
industry.
Synthetic fabric weaving was introduced and developed after cotton fabric
weaving in all four Asian countries. Taiwan was the first to begin production
and began to export before 1970. Korea followed closely and reached Tai­
wan's level of export expansion by the mid-1970s. Thailand's production/de­
mand ratio exceeded unity in the early 1970s, and Thailand followed both
Taiwan and Korea in export expansion by partly taking advantage of the earlier
development of the cotton textile industry. For Indonesia, only a long-run
trend is shown in Figure 10.1 due to the lack of consistent time series data
for the industry's development. The development of the Indonesian synthetic
fabric industry was delayed until the mid-1970s and exceeded unity only after
1980.
Two characteristics of production/demand ratios of the four countries
should be noted. First, domestic demand for textile goods (estimated by
production plus imports minus exports) was largely made up of goods in the
intermediate stage of production. The intermediate goods were consumed by
the textile industry to produce finished textile products such as clothing, bags,
and other accessories, most of which were directly exported. This characteristic
implies that the production/demand ratio for intermediate goods would tend
to be very low relative to the same ratio for finished textile products. Second,
the share of indirect exports was relatively high at 70 to 80 percent in Korea
and Taiwan. This tended to raise the overall production/demand ratio for the
synthetic textile industries (including unfinished fabrics as well as finished
products) higher than it would be otherwise by reducing the apparent domestic
demand estimate without increasing production. In Thailand and Indonesia
the indirect export of synthetic fabric began in the 1980s and has reached
around 15 percent only recently.
Development of the steel industry, which lagged behind synthetic fabric
production by five to ten years, occurred in the mid-1970s in all four coun-

Notes: Product ion /demand ratios tended to increase for these industries. The import sub­
stitution stage is reflected as the ratio approaches unity; the export expansion stage,
as the ratio continues to increase above unity. Although the general trend is depicted,
there were observed time lags between the two industries and among the four coun­
tries, which indicates time differentials in CPC development.
Sources: Calculated from statistics obtained from the Japan Chemical Fiber Association
and the Japan Iron and Steel Federation. Because of the unavailability of consistent
time series data, the curve for Indonesian synthetic fabrics (top) depicts only a long-
run trend based on ad hoc information of the import substitution and export expan­
sion stages of the industry's development.

236 Japan's Experience and Contemporary Developing Countries
tries (see Figure 10.1). For steel, the production figures measure domestic
production of crude steel, whereas the demand figures include finished steel
products converted to a crude steel basis. The production/demand ratios in­
clude indirect exports in the form of steel products. Furthermore, the ratios
fluctuated in response to both domestic and world market business cycles.
Korea quickly developed its steel industry and succeeded in reaching the ex­
port expansion stage by 1980, whereas Taiwan still remained below unity
throughout the 1980s, due, in part, to being handicapped by the small size
of its domestic market. The production/demand ratios of both Thailand and
Indonesia stayed below 0.3 until the mid-1980s but have recently begun to
increase rapidly.
Both the textile and steel industries have typically followed CPC develop­
ment in countries with abundant labor and a sizable domestic market.
Although MNCs participated at the early stage of development, with local
firms entering at a later stage, both groups aimed at import substitution from
the beginning. In contrast, finished clothing production and the electrical
machinery industry, although labor-intensive and introduced earlier, relied
on export expansion from the beginning without experiencing import substi­
tution. There are also many such cases in Hong Kong and the export-
processing zones of Taiwan and the Philippines.
This particular pattern is made possible by MNCs, which import capital,
technology, and packaged parts and materials, use local labor for sewing and
assembling, and export abroad through their own distribution networks. Ex­
port expansion is possible from the beginning because the production tech­
nology of MNCs is standardized, so that skilled labor, continuous learning,
and the adoption of new technology are not required. This situation describes
a typical process in the product cycle (PC) theory, by which technology and
production are transferred to developing countries through DFI (Vernon 1964).
Industrial development of the PC type provides developing countries with
labor employment and foreign exchange, but both the educational effects on
local firms and laborers and the linkage effects on other industries are in­
evitably limited.
Kinked Efficiency Improvement in CPC Development
The primary force underlying CPC development in Japan was improved
production efficiency—that is, better utilization of labor, a larger scale of
operation, advanced technology and equipment, and a general improvement
in production and management techniques. Both cost reduction and quality
improvement occurred, which allowed for the substitution of domestic
products for foreign ones, first domestically (import substitution) and later
abroad (export expansion). Figure 10.2, Panel 1, illustrates the declining im­
port/demand ratio (import substitution) and the increasing export/produc-

Figure 10.2
Alternative Mechanisms Underlying CPC Development
Panel
A

t2 h
Time
Notes: Panel 1 illustrates the process of CPC development—import substitution followed by
export expansion, by means of a decreasing import/demand ratio and an increasing ex­
port/production ratio. Panel 2 illustrates alternative mechanisms underlying CPC develop­
ment—the ABC curve illustrates a continuous reduction of domestic production costs relative
to foreign production costs, whereas the ADC curve illustrates a kinked path of relative costs.
Under the latter mechanism, import substitution is achieved not so much by small reduc­
tions of relative costs as under import restrictions (the AD curve), whereas export expan­
sion, which is subsidized only in the beginning, is mainly achieved through rapid reduction
of relative costs (the DC curve).
Source: Reproduced from Yamazawa and lambunlertchai (1985: figure 16.1).

238 Japan's Experience and Contemporary Developing Countries
tion ratio (export expansion) in CPC development, whereas Figure 10.2, Panel
2, charts the reduction of the underlying relative costs. The AC curve in Panel
2 shows the continuous improvements in efficiency and decreasing domestic
costs relative to foreign costs that are required for successful import substi­
tution and export expansion in the marketplace. Quality improvements may
be included as a factor contributing to the nonprice competitiveness of domes­
tic products. It was observed in Chapters 4 and 5 that both a productivity
increase and a relative price decline accompanied the successful import
substitution and export expansion of cotton textiles and steel products in
Japan.
In contrast, in contemporary developing countries, industrial development
has been supported by protection rather than by improvements in efficiency.
Once imports have created domestic demand, domestic production is en­
couraged to meet that demand through restrictions on imports. Import sub­
stitution is achieved even without sufficient productivity improvement and
cost reduction (see the AD curve in Figure 10.2, Panel 2). That is, import
substitution and export expansion are possible with limited improvements in
efficiency if high tariffs, import bans, and export subsidies are effectively
imposed.
This kind of government intervention is common in many developing coun­
tries, and it has been widely observed that these countries continue to im­
pose high tariffs even after import substitution has apparently been achieved
and export expansion has begun. The ADE curve in Figure 10.2, Panel 2, il­
lustrates the effects of this strategy. It is too costly in the long run, however,
to continue export expansion without improvements in efficiency, but it may
also be an exaggeration to say that no improvements in efficiency are achieved
in this process. It is plausible to contend that import substitution can be
achieved under tariff protection and that export expansion can begin with
the introduction of export subsidies, but that improvements in efficiency and
cost reduction are only brought about by the competition of exports in for­
eign markets.
The ADC curve illustrates the kinked path of improvements in efficiency.
The domestic product becomes competitive at point C, so that true import
substitution is achieved (assuming there is no rush to import in the absence
of import restrictions), and export expansion can continue without subsidi­
zation. As Figure 10.2, Panel 2 shows, kinked rather than linear improvements
in efficiency seem to better explain the import substitution and export ex­
pansion of textiles in Asian developing countries.
Changes in Business Behavior at the Export Stage
It is not easy to substantiate the kinked path that results from improve­
ments in efficiency as found in textile industries in Asia because of the lack

Is the Japanese Model Applicable? 239
of reliable data on productivity and costs throughout the import substitu­
tion and export expansion stages. However, Yamazawa and Tambunlertchai
(1985) provide evidence to support the belief that Thai textile firms under­
took behavioral changes as they began the process of export expansion.
During August and September 1979, the authors conducted interviews in
major textile firms in Thailand to examine the relationship between the ex­
port performance (measured by the percentage of production exported) of
individual firms and their production, management, and marketing charac­
teristics. All the sample firms began with import substitution, and none ex­
ported before 1971. Although the export/production ratio for all firms
combined had increased to more than 30 percent in 1978, the ratios for the
individual firms differed greatly.
Product type is the crucial factor in the export/production ratio, a finding
that is consistent with the difference in comparative advantage among product
categories. The export/production ratio is larger for cotton and polyester-
cotton blend fabrics because of the large domestic market, making produc­
tion on a larger scale practical. The ratio is smaller for polyester-rayon blends
and synthetic yarn fabrics because of the small domestic demand and the
resulting smaller scale of production and the increased difficulty in transfer­
ring technology.
Among firms with the same products, however, the principal determinant
in making behavioral changes is the ability to make changes in production
and management techniques. The shift from domestic sales to exports and
the establishment of marketing and distribution channels abroad require such
changes, but not every firm can make them. Standards of quality, packaging,
and timely delivery are usually higher for exports than for domestic sales.
Textile finish, for example, is classified into categories—A+, A, B, and C—
according to the number of frays and streaks in the goods. Normally only
A+ and A goods can be exported; B goods are sold locally, and C goods
are rejected. Quality control inspection is required for exports, but the dis­
tinction between A and B goods is less important when production is geared
to the domestic market. Those firms that exported established stricter inter­
nal quality control measures when they began exporting regularly. Changes
in the product mix and concentration on fewer items enabled them to realize
economies of scale, and marketing and production controls were integrated
more closely to permit quicker responses to changes in world demand.
All these changes in production and management strategies were introduced
regularly, and the firms that achieved a greater export/production ratio were
the firms that made greater efforts in these areas. Changes in production and
management strategies seem to have contributed to improvements in efficiency
at the export expansion stage, although this is difficult to quantify. Neverthe­
less, the strategy of kinked improvements in efficiency is a rational one for
developing countries grappling to achieve CPC development.

240 Japan's Experience and Contemporary Developing Countries
Not every industry can pass through import substitution to export expan­
sion. Many, such as the metals, machinery, and chemical industries, have not
been able to export after achieving import substitution under protection. Be­
cause they are protected and because they are also constrained by small-scale
production due to the small size of domestic markets and the greater need
for technology transfers, these industries can only meet domestic demand.
The formation of larger regional markets through regional integration agree­
ments will help promote the CPC development of these industries.
APPLICABILITY OF JAPANESE
GENERAL TRADING COMPANIES
Motivated by the need to increase foreign exchange earnings, many develop­
ing countries are eager to develop new exports and explore new markets for
their traditional exports. Aware of the difficulties involved, they have become
interested in Japanese general trading companies (GTCs) because the GTCs
significantly contributed to the expansion of Japan's foreign trade (see Chapter
6). Some developing countries, including Korea and Thailand, implemented
GTC promotion policies with the expectation that sales of their exports would
rapidly increase. Because their GTCs have only recently been set up, it is too
soon to fully evaluate their performance or to forecast their future with any
degree of confidence. However, it is possible to assess their early develop­
ment and, because they are modeled after Japan's GTCs, to compare develop­
ment experiences.
Some insist that GTCs evolved out of Japan's social, economic, and cul­
tural background and that Japanese-style GTCs cannot be introduced into
many developing countries. They argue that entrepreneurship is present only
in countries with a Confucian tradition and that sociocultural factors or differ­
ent environmental factors in countries without a Confucian tradition can de­
lay industrialization; they suggest that developing countries fitting this
description ought to utilize Japanese GTCs rather than try to create their own.
However, despite questions regarding the applicability of the GTCs' functions
under different social, economic, and cultural conditions, a case can be made
for promoting indigenous GTCs. This is discussed below in the context of
Korean GTCs.
Korean Promotion of GTCs
In 1963, Korea adopted an export-oriented growth policy, under which
manufacturing firms, not intermediary firms, became eligible for preferen­
tial export financing from banks at half the prime rate. Because of this in­
centive, the export of labor-intensive manufactures expanded rapidly in the

fs the Japanese Model Applicable? 241
1960s. By presidential decree in 1972, Korea began to promote its heavy and
chemical industries and used monetary and fiscal incentives to encourage
domestic firms to explore new areas such as shipbuilding, steel manufactur­
ing, and chemicals production. These Korean domestic products eventually
would be exported but to do so meant finding new distribution channels. To
this end, another presidential decree promoting GTCs was handed down in
1975.
Two types of trading channels existed in Korea before 1975. One was made
up of small and medium-sized trading companies, which generally special­
ized in a limited number of commodities. Because their major markets were
in North America and Western Europe and because they could not afford
to maintain their own information and marketing networks in those distant
markets, they exported their goods through foreign trading companies (in­
cluding Japanese GTCs). Large manufacturers exporting directly represented
the other channel. For instance, large textile or electrical manufacturers es­
tablished branches or sales agents in major markets and exported products
through them. Non-Korean trading companies were not interested in the new
exports, and Korean manufacturer traders lacked the skills required to de­
velop new markets. Thus, large, effective, domestically owned trading corn-
Table 10.1
Eligibility Requirements and Preferential Treatment for Korean GTCs
Qualification criteria Eligibility requirements
Capital 2 billion won (minimum total assets)
Annual export performance 2 percent increase in total exports over previous
year
Number of export items 5 (US$1 million or more in total sales per year)
Number of overseas branches 20
Preferential treatment benefits
Preferential treatment in international tender
GTCs can import major raw materials even if they serve as intermediary
GTCs can use more than two banks for letters of credit for importing major raw
materials
Overseas branches of GTCs can hold more foreign exchange than other trading
companies
GTCs can have more than two branches in one country
Tax reduction and exemption for GTC export and import commissions
Source: Presidential Decree on the Implementation of the Foreign Trade Transaction Act, Republic
of Korea, 1975. Seoul.

242 Japan's Experience and Contemporary Developing Countries
panies were needed, which led the Korean government to hand down the 1975
presidential decree to begin promoting GTCs.
The eligibility requirements and types of preferential treatment accorded
GTCs in Korea are listed in Table 10.1. These GTCs were also eligible for general
benefits for export promotion. The requirements listed in Table 10.1 reveal
the Korean government's intention to promote trading companies that ex­
ported many types of products and had global information networks. In this
manner, the Korean government could encourage the development of heavy
manufactures for export.
By 1978, 13 companies had been declared eligible for preferential treatment.
They were affiliated with major manufacturing groups, such as Sam Sung,
Dae Woo, and Hyun Dai, and 60 percent of their exports came from within
their own groups. They were required to renew their status every year. Three
firms eventually failed to meet the eligibility requirements and withdrew their
applications. In 1982, ten companies continued to maintain eligibility, and
one of these, which had been created especially for small and medium-sized
manufacturing firms, was operated by the government.
The expansion of the designated Korean GTCs was remarkable. Their com­
bined share of total Korean exports was 26.4 percent in 1977, 33.9 percent
in 1979, and more than 50 percent by 1982. GTC exports of heavy industrial
goods also increased remarkably and represented 47 percent of GTC exports
in 1979, compared to the share of 21 percent of such exports by Japanese
trading companies in Korea.
Korean GTCs have also been active in exploring new markets, particularly
in Latin America and Africa. Their performance so far has justified their
promotion by the Korean government.
Problems with Korean GTCs
Although it is too soon to fully assess the effects of the Korean GTCs, a
comparison with Japanese GTCs is helpful in identifying problems. 1
By 1979, Korean GTCs had reached a scale of operations comparable to
that of the Japanese GTCs in 1960, when the latter began to diversify (see
Table 10.2). For this reason, Korean GTCs are compared with the Japanese
GTCs of FY1960 rather than with the Japanese GTCs of the late 1970s.
Japanese figures for FY1960 in Table 10.2 are converted at 360 yen to the U.S.
dollar, whereas, in 1979, the rate was 219 yen and the export price index had
risen by 38 percent. This means the Japanese Figures for FY1960 must be mul­
tiplied by 2.27 to correspond with the FY1979 Korean data. Total Japanese
sales in 1960 adjusted in this manner are US$933 million for the average of
12 GTCs and approximately US$1,977 million for Mitsubishi; sales per em­
ployee are US$5.0 million and US$6.4 million, respectively, all of which are

Table 10.2
Comparison of Korean and Japanese GTCs
Korea (FY1979) Japan (FY1960)
Average
(9 GTCs) Dae Woo a Average
(12 GTCs) Mitsubishi 3 Japan (FY1978)
Average
(9 GTCs) Mitsubishi
Overseas branches (number)
Overseas staff*
Total sales c
Sales per branch c
Sales per employee c,cl
GTC share of total exports (%)
GTC share of total imports (°?o) 29
51
561
19.3
5.2 56
116
1,120
20.4
4.3
Korea (FY 1979)
33.9
4.7 41
189
411
10.0
2.2 55
308
871
15.8
2.8
Japan (FY 1960)
47.6
63.9 103
647
11,969
116.1
18.5 138
873
21,314
154.4
2.2
Japan (FY 1978)
48.1
51.2
Notes:
a. Dae Woo Industrial and Mitsubishi Corporation are the largest GTCs in their respective countries.
b. Number of expatriate employees only; Korean data are for 1977.
c. In millions of U.S. dollars. Korean figures include only exports and imports; Japanese data also include offshore trade and domestic (within
Japan) transactions.
d. Total sales divided by the worldwide total number of employees.
Sources: Financial statements of the companies.

244 Japan's Experience and Contemporary Developing Countries
comparable to the Korean figures (see Table 10.2). Thus the Korean GTCs
are not very different from the Japanese GTCs at the start of rapid economic
growth.
The Korean GTCs still need to resolve the following problems:
1. Despite their rapid expansion in recent years, Korean GTCs act as trading
arms of parent manufacturing groups, and the GTC figures, which include
existing exports by parent manufacturers within the respective groups, can­
not be expected to increase at the same rate in the future. Expansion of
their activities outside their own groups is the biggest task confronting
Korean GTCs.
2. Korean GTCs specialize in export trade. Although their share of total ex­
ports has become more similar to that of the Japanese GTCs, their share
of imports remains much smaller than that of their Japanese counterparts.
Domestic sales represent only a small share of Korean GTC earnings. Dae
Woo's exports, for example, were 528 billion won in 1979 and represented
85 percent of Dae Woo's total sales. In contrast, the shares of exports, im­
ports, and domestic and intra-country trade in the total sales of the Mit­
subishi Corporation in FY1960 were 17.4 percent, 29.9 percent, and 51.3
percent and 1.4 percent, respectively. These shares show that Mitsubishi's
reliance on domestic trade was over 50 percent. Taking into account the
commission fees, risks, and handling costs of each type of transaction,
it is often said that the order of profitability is domestic, import, export,
and intra-country trade. During the pre-Second World War period,
Japanese GTCs mainly profited from domestic and import trade and bore
the costs incurred in export trade. Since only export values are considered
under the Korean eligibility requirements, a Korean GTC tends to handle
only the export trade for other members of the same group, and other mem­
bers retain their own import and domestic trade capabilities. Thus, in ex­
panding their business, Korean GTCs are restricted to the less profitable
export transactions.
3. Large Korean manufacturers tend to export their products through their
own channels. Before the Second World War, Japanese manufacturers
lacked personnel skilled in foreign transactions and thus had to rely on
trading companies. But today in Korea, manufacturers have this type of
personnel and can engage in direct exports independently. This situation
resembles the practices of Japanese automobile manufacturers and elec­
tronics firms exporting either directly or through affiliated special trading
companies. Large manufacturers today will continue to export and import
directly as long as there is no incentive to do otherwise.
4. The promotion of GTCs has been implemented on a competitive basis,
and little preferential treatment is shown. Ten GTCs are too many for the

Is the Japanese Model Applicable? 245
Korean economy given its present size. Three or four GTCs would be closer
to the optimum number. However, the promotion policy in effect at present
is consistent with the promotion of efficient GTCs; that is, the policy it­
self does not attempt to limit the number admitted. Instead, it allows com­
petition to narrow the field.
Indigenous Korean trading companies have closer contact with domestic
producers and consumers. Considering the quick shift to export expansion
of heavy industrial products, they have been as eager to develop new export
items as the Japanese GTCs were during the high-growth period of their econ­
omy. Korean GTCs are credited with their contribution to the diversification
of export products and expansion of markets. The promotion of indigenous
GTCs is thus justified despite the high costs to maintain overseas networks
at present.
Japanese industrial policy and technological borrowing as well as the CPC
model and GTCs have attracted the interest of foreign economists. An at­
tempt by developing countries to copy Japan's GTCs is not recommended.
Nevertheless, analysis of the Japanese experience can aid in the development
of appropriate policy measures and the identification of rational responses
consistent with accepted economic principles.

Notes to Chapters
CHAPTER ONE
1. Toyo Keizai Shinposha's Long-Term Economic Statistics series on Japan's eco­
nomic development enables this overview. Volume 14 (Yamazawa and Yamamoto
1979a) especially provides time series statistics of manufacturing production and
exports and imports classified by common industry groups, which allows for the
analysis of the interaction between production and trade.
2. It should be noted that exports and imports include only commodities and that
exports are FOB and imports .OF.
3. Kuznets (1967) gives a higher foreign trade ratio for Japan than the estimates
presented here.
4. The distinction between light and heavy industries is based on factor propor­
tions of production and end use, but it has a weak theoretical basis. It does,
however, fit well into the sequence of industrial development and is often used
in Japan and centrally planned economies.
5. This expansion of the export market by stages is discussed in Nawa (1937), who
hypothesizes it as a "three-stage development" of Japan's foreign trade, and
in Kojima (1965), who describes it as an "expanding export frontier." For cross-
classification of commodity by geographical region, see Yukizawa and Maeda
(1978).
6. See Yamazawa and Yamamoto (1979a, Pt. 1, chap. 1, and Pt. 3) for detailed
statistics and an analysis of the geographical structure of Japan's foreign trade.
7. This phasing approximately coincides with that described by Fei, Ohkawa, and
Ranis (1985) in their analysis of Japan's economic development.
8. Elasticities are computed from an estimation of raw material import functions
of the log-linear type.

248 Notes
CHAPTER TWO
1. Akamatsu apparently obtained the idea from W. Hoffman (1931) during his stay
at the Kiel Institute of World Economics in the late 1920s but assimilated, ex­
panded, and gave the concept a poetic name. Readers having difficulty trying
to imagine a flock formation of flying wild geese may refer to the illustration
in Laura Ingalls Wilder, Little House on the Prairie, Harper and Row, 1953, p. 263.
2. See the estimation of the constant price series of production, imports, and ex­
ports of cotton fabrics in Chapter 4, pp. 71-73.
3. Changes in the organizational structure of the expanding industry certainly af­
fect its CPC performance. Owing to the relatively large size of Japan's domestic
market, major industries had at least a few competing firms. Competition among
domestic producers brought about rapidly decreasing costs, and accelerated im­
port substitution and export expansion. Because of the small domestic markets
and lack of competition among domestic rivals in the Asian developing coun­
tries, there is typically a lack of incentive to reduce costs, which tends to delay
CPC development. However, competition in the domestic market can be strength­
ened through government policies that encourage more competition with the in­
troduction of imports. This topic presents an intriguing question for the study
of industrial organization but is mentioned only occasionally in the following
chapters.
4. Despite the decline in world prices of rayon and some chemicals, which was due
to rapid expansion and greater economies of scale, Japanese prices of these
products declined at an even faster rate.
The differences in domestic and export prices do not indicate dumping prac­
tices but reflect product differentiation within manufacturing industries such that
productivity increases and decreasing costs are greater in the export-producing
sector than in the rest.
5. Shinohara (1972:30) was the first to analyze the correspondence between the growth
cycle and balance of payments. He concludes that export booms, through an im­
proved balance of payments and favorable demand conditions, generated long
swings in pre-Second World War manufacturing. His explanation is consistent
with the second type of correspondence in the analysis presented here, but it does
not explain the interaction between production and the balance of payments over
the entire period.
CHAPTER THREE
1. Myint (1958), Caves (1965, 1971), and Meier (1976) review studies on the role of
primary goods exports in generating economic growth in North America and
Southeast Asia.
2. Fujino (1965) and Fujino, Fujino, and Ono (1979) analyze the development of
Japan's silk-reeling industries, focusing on productivity improvement, employ­
ment creation, and other domestic economic impacts.

Notes 249
3. The constant market share (CMS) approach, which allows for the identification
of the sources of export growth, has frequently been used by analysts since the
1950s. Richardson (1971) summarizes the formulation of the CMS analysis and
its economic implications. Empirical applications of the analysis are also reviewed.
4. Time series data of Japanese and Chinese exports of hand and steam filature
to the United States were constructed by Niyomka (1975) under the assumptions
(hat (1) both Japan and China exported all production of steam filature before
1910, (2) the two countries exported hand and steam filatures to the United States
in the same proportions as their total exports to the world, and (3) the United
States imported hand filature from these two countries alone.
5. Lockwood (1954), Baba and Tatemoto (1968), and Shinohara (1962) attributed
the expansion of Japan's exports to the depreciation of silver. But as shown in
the following analysis the effects of silver depreciation need to be examined with
care.
6. A similar point was made by Fujino (1965).
7. Equation 3.2 was estimated alternatively by applying individual consumer ex­
penditures of the United States to y.
log, A*=0.O003+ 1.7930 log, .y-0.2976 log, L-0.8085 log, P
(8.34) (1.36) (1.92)
ft 2 = 0.9371, DW = 1.93
Cumulative consumption of silk fabric (L) was included in order to measure the
learning effect of silk fabric consumption, which increased rapidly after 1870 but
proved to be statistically insignificant.
8. See Kajinishi (1964) for an analysis of the financing of the silk-reeling industry
in Meiji Japan, and Fujino (1965, chap. 19) for the impact of the industry on
the development of Japan's banking system.
CHAPTER FOUR
1. The analysis in (his chapter is based on Kajinishi (1964), but the periodization
presented here is the work of this author.
2. Conventional tariffs are those that are agreed upon with trading partners.
3. The most severe competition in the commercialization of acrylic took place among
five firms because of the absence of a basic patent and possible independent de­
velopment.
CHAPTER FIVE
1. Zaibatsu were powerful family-controlled business conglomerates involved in
manufacturing, trading, and financial activities. Each zaibatsu was centered around

250 Notes
a holding company (under family control), which in turn controlled the main
operating companies of the group, which in turn controlled smaller companies.
2. Fogel and Engerman (1969) presented a short-run equilibrium model in which
equilibrium price and quantity are determined by the intersection of annual short-
run demand and supply curves. The short-run demand curves were moved to the
right by domestic income increases and supply curves were shifted in the same
direction by increases in productive capacities. A similar approach was tried in
an earlier study on which this chapter is based, but the present model was even­
tually adopted because of data requirements. Reliable data on productive capac­
ity over the whole period are not available for Japan's steel industry.
3. The price was deflated by the GDE deflator to adjust for the inflationary trend
that characterized Japan's economic growth (Yamazawa 1972).
4. Differentiation of equations (1) through (5) with respect to time / enables a sys­
tem of log-linear equations in growth rates of variables to be derived. The linear
equation system is solved for the equilibrium growth rate for each endogenous
variable. It can be proven, making a few plausible assumptions, that the model
generates a path of successful import substitution and export expansion but tends
to move in the other direction for small values of S/D.
5. Yamazawa (1972) gives the details of final tests and simulations.
6. This section is based on MITI (1970).
7. Tecson (1985) gives a detailed analysis of the first and second reorganization pro­
grams and an appraisal of their contribution to the development of industry.
CHAPTER SIX
Chapter 6 is based on two studies jointly undertaken by Ippei Yamazawa and Hiro-
hisa Kohama (Yamazawa and Kohama 1984).
1. In 1888, the number of foreign merchant houses in Yokohama totalled 127, which,
by represented nationality, included English (41), Chinese (25), American (18),
German (17), French (10), Swiss (8), other European (5), and Indian (3). Among
the largest foreign merchant houses were Jardine, Matheson and Co. (English),
Walsh, Hall and Co. (American), and Siber Brenwald and Co. (Swiss).
2. The term direct (i.e., direct trade, and direct imports and exports) is used here
to describe trade conducted by Japanese nationals rather than foreigners. In other
words, direct trade excludes foreign merchants as go-betweens.
3. Extraterritoriality was admitted in foreign settlements, a situation described by
Takahashi as "quasi-colonial" (Takahashi 1968: vol. 2, pp. 167-193).
4. See Takahashi (1973: vol. 3, chap. 2) for a description of the disadvantages to
Japanese merchants. Takahashi also cites an editorial that appeared in Tokyo Keizai
Zasshi (Tokyo Economic Journal), 10 January 1881, in which direct trade policy

Notes 251
was criticized by free traders on the ground that the government was driven too
much by nationalism when it forced, at great cost to the national economy, the
substitution of Japanese for foreign merchants.
5. Old exports and imports are those that had already been traded before 1875 when
direct trade promotion was announced. New exports and imports are those goods
that began to be traded only after 1875. Indigenous goods were produced and
consumed domestically before the opening of trade, whereas modern goods were
introduced to Japan through trade, with production in Japan beginning later.
Imports are all modern goods by definition and are classified into six groups
by end use.
6. G. C. Allen contended that Western merchants and Japanese branches of for­
eign banks contributed greatly to the expansion of Japan's foreign trade in the
Meiji period (1868-1912). See Allen and Donnithorne 1954:196. Data in this chap­
ter, without refuting the contribution of Allen and Donnithorne, substantiate the
proposition that Japanese merchants expanded so as to replace foreign competi­
tors and accelerate the expansion of Japan's foreign trade.
7. See Takahashi 1973: vol. 3, p. 138.
8. This is consistent with Allen's explanation of the competence of Japanese mer­
chants in comparison with Chinese merchants (see Allen and Donnithorne
1954:204). He pointed out that Japan's mercantile activity rested on a secure in­
stitutional basis and an industrial hinterland that had been developed mainly by
native initiatives.
9. The author is indebted to Yasukichi Yasuba for pointing out the significance of
this regulation to this discussion. However, because of the lack of documenta­
tion in either Japanese or foreign sources, the restrictive effect of the regulation
cannot be accurately assessed. But this author disagrees with the view that this
regulation was the driving force in increases in the direct trade ratio.
10. See Takahashi 1973: vol. 3, chap. 2.
11. Differences in average values between major pairs of categories turned out to
be statistically significant.
) 2. /?,=/?, = 0 for exports to or imports from Europe and North America; R,
= I and R 2 = 0 for exports to and imports from East Asia; Ri = 0 and R 7 =»
1 otherwise.
13. In order to avoid the effect of extreme values in log-linear formulation, zero and
near-zero values of the DER and DIR are replaced uniformly by 0.001 through­
out the estimation.
14. In the absence of interaction terms, the coefficients of both N and M are posi­
tive and the coefficient of N is statistically significant.
15. In other regressions not shown here, however, the coefficients of R u the category
Old and /?,, and the category Indigenous had a negative effect on DER, which
partly refutes hypothesis 1 (part 3).

252 Notes
16. See Chap. 4, page 78 of this volume; also see Takahashi 1973: vol. 3, pp. 173-197.
17. See Yasuba (1979) for details on the development of ocean transportation.
18. See Takahashi 1973: vol. 3, p. 63.
19. Until a merger reduced the number to nine, the major GTCs in Japan had been
referred to as "the ten GTCs." The companies include Mitsubishi, Mitsui, C. Itoh,
Marubeni, Sumitomo, Nissho-Iwai (formed by a merger in 1968), Toyo Menka,
Kanematsu-Gosho (merged in 1967), Nichimen, and Ataka (which was absorbed
by C. Itoh in 1977).
CHAPTER SEVEN
1. Although the Special Emergency Tariffs (1904-05), as well as the Luxury Tariffs
(1924), were motivated by the need to increase tariff revenue, these laws were ex­
ceptional in the history of Japan's tariff laws.
2. Excerpt from the speech delivered by Minister of Finance Osachi Hamaguchi on
the proposal for the revision of the Tariff Law at the 51st session of the Japanese
Diet (Ministry of Finance 1963: Vol. 13, p. 46).
3. The protection effects of tariffs on domestic manufacturing activities are properly
measured not in terms of nominal rates, but in terms of their effective protection
rates (ad valorem equivalents). These rates, however, are not available for the prewar
years because of the prohibition on the release of data for security reasons. Chap­
ter 8 provides estimates of the effective protection rate for the years after the
Second World War.
4. The data presented here are based on data originally published in "Quantity and
Value of Merchandise Imported by Each Port," Dainihon gaikoku boeki nem-
pyo (Foreign Trade Annual of the Empire of Japan), Ministry of Finance, vari­
ous years.
5. Note that when the ratio of tariffs collected on import value is calculated for
an aggregated commodity group, such as total commodities or manufactures,
the arithmetic mean of the tariffs collected on individual commodities in the group
is weighted by the value of imports. But very high tariffs on individual commodities
tend to restrict imports and lower the weighted arithmetic mean of tariffs. Re­
strictive tariffs are therefore given low weight. Data for tariffs on individual com­
modities, on the other hand, are fairly free from this bias because simple arithmetic
means are calculated as average tariffs for groups classified by industry and eco­
nomic use.
6. Japan lagged behind the other developed countries in the adoption of restric­
tions. Quantitative restrictions had not been adopted as a permanent feature of
government policy by any country until 1931 when the world began to see the
rise of protectionism. In 1931, France adopted quantitative restrictions, and the
main European countries followed suit.

Notes 253
7. Great Britain, Department of Overseas Trade (1930).
8. Before 1899, although DFI was allowed outside the foreign settlements, it did
not significantly affect industrial development. Within the foreign settlements,
however, DFI was active in such activities as tea processing, ship-repairing, iron-
making, paper manufacturing, and the installment of gas lighting immediately
after the opening of trade.
9. Although Hitachi independently developed the technology to produce electric
dynamos, it purchased the technology for producing related machinery, such as
boilers and turbines, under patent contracts (see Aihara 1973).
CHAPTER EIGHT
1. Japan joined the International Monetary Fund (IMF) as an Article 14 country
in August 1952 (which entitled Japan to impose foreign exchange controls on
transactions). Japan became an Article 8 country in April 1964.
2. For a detailed discussion on these measures see Komiya et al., eds. (1988).
3. "Excessive competition" has a special meaning in Japan. See Komiya et al., eds.
(1988).
4. The Foreign Trade and Foreign Exchange Control Law was enacted in December
1949 and continued unchanged until January 1982.
5. The Foreign Exchange Concentration Rule was eliminated in 1969. See MITI
(1982).
6. AIQ is not different from AA in practice, but is still an interim form between
IQ and AA and may be changed back to IQ if necessary.
7. The unsuccessful passage of the Machinery Industry Encouragement Act in 1963
and MITI's failure to strengthen its leadership over business are often cited as
indicators of changes in MITI's industrial policy. See Tsuruta (1982) for details.
8. Economic Planning Agency, Master Plan for Trade and Foreign Exchange Liber­
alization (June 1960), reprinted in Kanamori, ed. (1974).
9. These import quota restrictions remained despite their prohibition under the GATT
rule. Both rice and barley are categorized as state trading items and are exempted
from this rule.
10. The tariff on cheese and butter was raised from 35 to 45 percent, that on boilers
and electric generators from 15 to 20 percent, and those on lathes and other
machinery from 15 to 25 percent.
11. The government classified liberalized industries as either category 1 or category
2 and allowed different degrees of foreign participation.
12. A similar rediscounting facility for import financing was provided by the Bank
of Japan.

254 Notes
13. The numerator does not include foreign exchange fund loans, whereas the denomi­
nator includes export financing in the form of overdrafts provided to trading com­
panies by foreign exchange banks. See MITI, Tsusho Hakusho (1972:191, 339).
14. Rediscounting at the basic official rate by the Bank of Japan is still available
for qualified export bills.
15. The quantitative appraisal of the export privileges provided by the Korean govern­
ment is based on Yen (1983: table 3).
CHAPTER NINE
1. Thailand's growth was set back much less than other ASEAN countries in 1985
and 1986, and it recovered quickly to 6.6 percent in 1987 and 9.0 percent in 1988.
Although other ASEAN countries, with the exception of Singapore, recovered,
their growth rates were 2-3 percent lower than Thailand's.
2. Y=(C + I + G) + {X-M) = D + B,
where Y denotes national income, C denotes consumption, / denotes investment,
G denotes government spending, (X – M) denotes net exports, D denotes na­
tional expenditure, and B denotes the current external surplus in the national ac­
count (that is, exports minus imports of goods and services), which is the current
account balance minus the net transfer payment. Changes in B nearly reflect the
changes in the current account balance.
dY/Y = (D/Y)dD/D + (B/Y)dB/B
Since the ratio (D/Y) is around 1, the growth rate of national expenditure in Ta­
ble 9.2 closely reflects the changes in the contribution of national expenditure
to overall growth.
3. Figures for 1978 and 1986 are not consistent with this relationship. This is be­
cause both prices and the exchange rate changed drastically in these years so that
the current account balance, expressed in billions of current dollars, departed
substantially from the contribution of foreign demand.
4. Komiya (1988) provides a detailed analysis of Japan's macroeconomic policies
during the period covering the two oil shocks.
5. Due to the continued boom in 1987-88, the steel- and ethylene-producing indus­
tries exceeded their existing capacities and this has led to persistent excess de­
mand for iron. While some iron-producing firms are said to be planning expansion
capacity, many are exercising extreme caution in resuming capacity expansion at
the same rapid pace that occurred before the oil shocks.
6. Leasing machinery and equipment is new to many firms. Firms now prefer to
rent instead of purchase machinery and equipment to avoid depreciation of the
value of the machinery and equipment. This practice has become popular with
factory and office equipment, especially because technical progress is so rapid
that purchased machines become obsolete much earlier than their designated life

Notes 255
period. Factory and office equipment valued at about US$30 billion comprised
three quarters of the total amount of lease contracts in 1986 (MITI 1988a).
7. Cheap labor costs were cited as a prime motivation for DFI by 60.9 percent of
the firms operating in the Asian NIEs and 46.1 percent of firms in ASEAN.
However, the number of firms motivated by the desire to be present in local mar­
kets for these areas is also large at 58.3 and 51.7 percent, respectively.
8. Nihon Keizai Shinbun (The Japan Economic Journal), 11 November 1988. The
MITI White Paper (MITI 1988c) was based on a questionnaire survey conducted
in September 1988 on the globalization of Japanese firms' R&D activities.
9. Peck, Levin, and Goto (1987) provide the most detailed analysis of the industry
adjustment laws.
10. Peck, Levin, and Goto (1987) suggest that these eight industries were highly con­
centrated, and that it was anticipated that capacity reduction would be difficult
to achieve if left to individual firms.
CHAPTER TEN
1. The basis for this chapter began with the author's participation in the Compara­
tive Analysis Project, which was jointly coordinated by Kazushi Ohkawa and
Gustav Ranis at the International Development Center of Japan from 1975 to
1981. The studies by Yamazawa and Hi rata (1978), Yamazawa and Kohama (1984),
and Yamazawa and Tambunlertchai (1985) all originated from this project, and
both Chapter 6 and this chapter were based on these studies. The comparative
analysis methodology used by Fei, Ohkawa, and Ranis (1985) influenced that used
by the author when comparing the economic history of Japan with the recent
economic performances of various developing countries.
2. This is what Ohkawa (1976) called the "compressed process" of catching-up de­
velopment.
3. This section utilizes information from discussions with Rhee Chong Inn, and
from his doctoral dissertation (Rhee 1984).

Bibliography
Japanese statistics sources often include English explanations and are acces­
sible to English-language readers.
Aihara, H. 1973. "Nihon niokeru Gaikokushihon no Yakuwari [Role of Foreign Capital
in Japan]" Keizai to Boeki [Economy and Trade], No. 110 (September). Yokohama-
shiritsu Daigaku (Yokohama City University).
Allen, G. C, and A. G. Donnithorne. 1954. Western Enterprise in Far Eastern Eco­
nomic Development: China and Japan. Winchester, MA: Allen and Unwin.
Akamatsu, Kaname. 1943. "Shinkou Kogyokoku no Sangyou Hatten [Industrial De­
velopment in Newly Industrializing Countries]," in Veda Teijiro Hakushi Kinen
Ronbunshu [Essays in Honor of Dr. Teijiro Ueda], Vol. 4. Tokyo: Kagagushugi
Kogyosha.
. 1961. "A Theory of Unbalanced Growth in the World Economy;' Weltwirt-
schaftliches Archiv, 86(2):196-217.
Asahi Shinbunsha. 1930. Nihon Keizai Tokei Sokan [Statistical Abstracts of the
Japanese Economy]. Tokyo: Asahi Shinbunsha.
Baba, M., and M. Tatemoto. 1968. "Foreign Trade and Economic Growth in Japan:
1868-1937." In L, Klein and K. Ohkawa, eds.. Economic Growth: The Japanese
Experience Since the Meiji Era. Homewood, IL: Richard D. Irwin.
Balassa, B. 1965. "The Tariff Protection in Industrial Countries: An Evaluation," J.
Political Economy 73(6):573-594.
Bank of Japan [Nihon Ginko]. 1956- . Keizai Tokei Nempo [Economics Statistics An­
nual]. Tokyo: Bank of Japan.
Bensusan-Butt, D M. 1954. "A Model of Trade and Accumulation," American Eco­
nomic Review 44(4):511-529.

258 Bibliography
Bergsten, Fred, and W. R. Cline. 1987. The United States-Japan Economic Problem.
Rev. ed. Washington, DC: Institute for International Economics.
Caves, R. E. 1965. "Vent for Surplus Models of Trade and Accumulation." In R. E.
Baldwin et al., eds., TYade, Growth, and the Balance of Payments, pp. 95-115.
Chicago and Amsterdam: North Holland.
. 1971. "Export-led Growth and the New Economic History?' In J. Bhagwati
et al., eds., Trade, Balance of Payments and Growth, pp. 403-442. New York:
North Holland.
Caves, R. E., and R. W. Jones. 1973. World Trade and Payments, 2d ed. Boston, MA:
Little.
Chen, E. K. Y. 1989. "The Changing Role of the ANICs in the Asian Pacific Region
Towards the Year 2000." In M. Shinohara and Fu-chen Lo, eds., Global Adjust­
ment and Future of Asia-Pacific Economies, pp. 207-231. Tokyo: Institute of De­
veloping Economies and Asia Pacific Development Centre.
Destler, I. M., H. Fukui, and H. Sato. 1979. 77ie Textile Wrangle: Conflict in Japanese-
American Relations 1969-1971. Ithaca, NY: Cornell University Press.
Fei, J. C. H., K. Ohkawa, and G. Ranis. 1985. "Economic Development of Korea,
Taiwan and Japan in Historical Perspective." In K..Ohkawa and G. Ranis, eds.,
Japan and the Developing Countries: A Comparative Historical Analysis, pp.
35-64. New York: Basil Blackwell.
Fogel, R. W., and S. L. Engerman. 1969. "A Model for the Explanation of Industrial
Expansion During the Nineteenth Century, With an Application to the Ameri­
can Iron Industry?' J. Political Economy 77(3):306-328.
Fujino, Shozaburo. 1965. Nihon no Keiki Junkan [Japan's Economic Cycle]. Tokyo:
Keiso-shobo.
Fujino, Shozaburo, Shiro Fujino, and Akira Ono. 1979. Sen-i Kogyo [Textile Indus­
try]. Choki Keizai Tokei [Long-Term Economic Statistics of Japan], Vol. 11. Tokyo:
Toyo Keizai Shinposha.
Gerschenkron, A. 1966. Economic Backwardness in Historical Perspective. Cambridge,
MA: The Belknap Press of Harvard University Press.
Great Britain, Department of Overseas Trade. 1930. Economic Conditions in Japan.
London: His Majesty's Stationary Office.
Haberler, G. 1964. "Integration and Growth of the World Economy in Historical Per­
spective," American Economic Review 54(2):l-22.
Hayami, Y. 1972. "Rice Policy in Japan's Economic Development," American J.
Agricultural Economics 54(1):19-31.
Helleiner, G. K. 1981. Intra-ftrm Trade and the Developing Countries. New York: Mac-
millan.

Bibliography 259
Henmi, K. 1969. "Primary Product Exports and Economic Development: The Case
of Silk!' In K. Ohkawa, B. F. Johnston, and H. Kaneda, eds., Agriculture and
Economic Growth- Japan's Experience. Tokyo: University of Tokyo Press.
Hoffman, W. 1931. Stadien und Typen der Industrialisierung. Jena, Germany: G.
Fischer.
Inada, K., S. Sekiguchi, and Y. Shouda. 1972. Keizai Hatten no Mekanizumu-Sono
Riron to Jisho [Mechanism of Economic Development: Its Theory and Practice).
Tokyo: Sobunsha.
International Monetary Fund (IMF). 1988. International Financial Statistics. Vari­
ous issues.
Japan Tariff Association. 1962. The Customs Tariff of Japan. Tokyo: Japan Tariff
Association.
Johnson, H. G. 1968. Comparative Cost and Commercial Policy Theory for a De­
veloping World Economy. Stockholm: Almqvist and Wiksell.
Kajinishi, M. 1964. Gendai Nihon Sangyo Hattensht Sen-i [History of Industry De­
velopment in Modern Japan: Textiles]. Tokyo: Kojunsha.
Kanamori, Hisao, ed. 1970. Boeki to Kokusai Shushi (Trade and Balance of Payments].
Tokyo: Nihon Keizai Shinbunsha.
Key, B. 1970. The Role of Foreign Contribution in Prewar Japanese Capital
Formation—with Special Reference to the Period 1904-14. Unpublished Ph.D.
dissertation. University of California, Berkeley.
Kobayashi, U. 1930. The Basic Industries and Social History of Japan: I9I4-I918. New
Haven, CT: Yale University Press.
Kojima, K. 1958. Nihon Boeki to Keizai Hatten IJapan's Trade and Economic De­
velopment]. Tokyo: Kunimoto-shobo.
. 1965. Seka't Keizai to Nihon Boeki [World Economy and Japan's Trade]. Tokyo:
Keiso-shobo.
. 1973. "Reorganization of North-South Trade: Japan's Foreign Economic Policy
for the 1970s;' Hitotsubashi J. Economics 13(2):l-28.
Komiya, R. 1988. Gendai Nihon Keizai [Modern Japanese Economy]. Tokyo: Tokyo
Daigaku Shuppan Kai.
Komiya, R., M. Okuno, and K. Suzumura, eds. 1988. Industrial Policy of Japan. Tokyo:
Academic Press Japan Inc.
Kuznets, S. 1967. "Quantitative Aspects of the Economic Growth of Nations: X Level
and Structure of Foreign Trade: Long-Term Trends," Economic Development and
Cultural Change 15(2), Part 11:1-140.
Linder, S. B. 1961. An Essay on Trade and Transformation. New York: Wiley.

260 Bibliography
Lockwood, W. W. 1954. The Economic Development of Japan- 1968-1938. Prince­
ton, NJ: Princeton University Press.
Mason, F. R. 1910. "American Industry and Tariff," American Economic Associa­
tion Quarterly 11(4):1-182.
Meier, G. M. 1976. Leading Issues in Economic Development, 3d ed. New York: Ox­
ford University Press.
Ministry of Finance [Okurashol. 1882-1943. Dainihon Gaikoku Boeki Nempyo [For­
eign Trade Annual of the Empire of Japan]. Tokyo: Ministry of Finance.
. 1938. Meiji Taisho Zaiseishi [Fiscal History of the Meiji and Taisho Periods].
Tokyo: Zaisei Keizai Gakkai.
. 1950. Zaisei Kinyu Tokei Geppo [Fiscal and Financial Statistics Monthly],
5 (May).
. 1951- . Nihon Gaikoku Boeki Nempyo [Foreign Trade Annual of Japan].
Tokyo: Ministry of Finance.
. 1963. Showa Zaiseishi [Fiscal History of the Showa Period], 13. Kokusai Kinyu
to Boeki [International Finance and Trade]. Tokyo: Toyo Keizai Shinposha.
. 1982. Zaisei Kinyu Tokei Geppo (Fiscal and Financial Statistics Monthly],
No. 365 (September).
Ministry of International Trade and Industry (MITI) [Tsusho Sangyo-shol. 1951- .
Tsusho Hakusho [White Paper on International Trade and Industry]. Tokyo:
Tsusho Sangyo Chosakai.
. 1952-60. Boeki Gyotai Tokei [Trade Business Statistics, by Industry]. Tokyo:
Tsusho Sangyo Chosakai.
. 1970. Shokou Seisakushi [History of Commercial and Industrial Policy].
Tokyo: Tsusho Sangyo Chosakai.
. 1982. Tsusan 30 Nenshi (Thirty-Year History of MITI]. Tokyo: Tsusho San­
gyo Chosakai.
, Industrial Policy Bureau [Sangyo Seisaku-kyoku]. 1988a. Susumu Kozouchosei
to Sangyo Kozou no Tembow Kozou Chosei Bijyon [Progress of Structural Ad­
justment and Prospect for Industrial Structure: Vision for Structural Adjustment].
Tokyo: Tsusho Sangyo Chosakai.
. 1988b. Tsusan Tbkei Handobukku (Handbook of Industrial and Trade Statis­
tics]. Tokyo: Tsusan Tokei Kyokai.
. 1988c. Gaishikei Kigyou no Doukou [Activities of Foreign Corporations in
Japan], Vol. 19. Tokyo: Keibun Shuppan.
Morikawa, H. 1976. "Sogoshosha no Seiritsu to Ronri [Evolution of General Trad­
ing Companies and Their Business Principles]." In M. Miyamoto, ed., Sogoshosha
no Keiseishi [Business History of General Trading Companies]. Tokyo: Toyo Keizai
Shinposha.

Bibliography 261
Myint, H. 1958. "The Classical Theory of International Trade and the Underdevel­
oped Countries," The Economic Journal 68(270):317-337.
Nakagawa, K. 1967. "Nihon Kogyo-ka Katei Niokeru Soshikikasareta Kigyouka Kat-
sudo [Organized Entrepreneurial Activities in the Process of Japan's Industrial­
ization]," Keizai-Shigaku [J. Business History], 2(3)(November).
Nawa, T. 1937. Nippon Bosekigyo to Genmen Mondai [Japanese Cotton Spinning
Industry and the Raw Cotton Issue]. Osaka: Daido-shoin.
Nihon Boseki Kyokai [Japan Spinners' Association]. 1982. Bokyo Hyakunenshi [A
Century of History of the Japan Spinners' Association]. Osaka: Nihon Boseki
Kyokai.
Nihon Kagaku Sen-i Kyokai [Japan Chemical Fiber Manufacturers' Association]. 1974.
Nihon Kagaku Sen-i Sangyoshi (History of Japan's Chemical Fiber Industry].
Tokyo: Nihon Kagaku Sen-i Kyokai.
Nihon Kanzei Kyokai. 1965. Boeki Nenkan [Trade Annual]. Tokyo: Nihon Kanzei
Kyokai.
Nihon Tekko Renmei. 1957- . Tekko Tokei Yoran [Iron and Steel Statistics Abstract].
Tokyo: Nihon Tekko Renmi.
Niyomka, P. 1975. Japanese Silk Exports and Economic Development: 1860-1910.
Unpublished master's thesis. Hitotsubashi University, Tokyo.
OECD. 1978. The Impact of Newly Industrializing Countries on the Pattern of World
Trade and Production in Manufacturing. Paris: OECD.
1979. The Case for Positive Adjustment Policies. Paris: OECD.
. 1981. Structural Problems and Policies Relating to the OECD Textile and Cloth­
ing Industries. Paris: OECD.
. 1987. Structural Adjustment and Economic Performance. Paris: OECD.
1988. Economic Survey: Japan 1987/1988. Paris: OECD.
Ohkawa, K. 1976. Keizai Hatten to Nihon no Keiken (Economic Development and
Japan's Experience]. Tokyo: Taimeido.
Ohkawa, K., and H. Rosovsky. 1973. Japanese Economic Growth: Trend Accelera­
tion in the Twentieth Century. Stanford, CA: Stanford University Press.
Ohkawa, K., N. Takamatsu, and Y. Yamamoto. 1974. Kokumin Shotoku [National
Income]. Choki Keizai Tokei 1 Long-Term Economic Statistics of Japan], Vol. 1.
Tokyo: Toyo Keizai Shinposha.
Ohkita, S. 1987. "The Outlook for Pacific Co-operation and the Role of Japan," The
Indonesian Quarterly 15(3). Jakarta: CSIS.
Ono, A. 1968. "Gijutsu Shimpo to 'Borrowed Technology' no Ruikei [Technological
Progress and Typology of Borrowed Technology]." In J. Tsukui and Y. Murakami,

262 Bibliography
eds., Keizai Seicho Riron no Tenbo [Prospects of Economic Growth Theories].
Tokyo: Iwanami Shoten.
Pasha, W. 1987. On the Contribution of Phasing to the Analysis of East Asian Eco­
nomic Development. Discussion Paper, Nagoya University, Nagoya, Japan.
Patrick, H., and H. Rosovsky. 1976. Asia's New Giant: How the Japanese Economy
Works. Washington, DC: The Brookings Institution.
Peck, M. J., R. C. Levin, and A. Goto. 1987. "Picking Losers: Public Policy Toward
Declining Industries in Japan," J. Japanese Studies 13(1):79-123.
Rhee, C 1. 1984. Kankoku no Boeki Hatten to Sogoshosha Katsudo [Korea's Trade
Development and General Trading Companies]. Unpublished doctoral thesis
presented to Hitotsubashi University and later published in Korean language un­
der the same title in 1987.
Richardson, J. D. 1971. "Constant Market Share Analysis of Export Growth," / In­
ternational Economics l(2):227-239.
Sautter, C. 1973. Le Prix de la Puissance [The Price of Power]. Paris. Le Sevil.
Shinohara, M. 1962. Growth annd Cycles in the Japanese Economy. Tokyo:
Kinokuniya.
. 1972. Ko-Kogyo [Mining and Manufacturing Industries]. Choki Keizai Tokei
[Long-Term Economic Statistics of Japan], Vol. 10. Tokyo: Toyo Keizai Shinposha.
1982. Industrial Growth, Trade and Dynamic Patterns in the Japanese Econ­
omy. Tokyo: University of Tokyo Press.
Shionoya, Y., and I. Yamazawa. 1973. "Industrial Growth and Foreign Trade in Prewar
Japan." In K. Ohkawa and Y. Hayami, eds., Economic Growth- The Japanese
Experience Since the Meiji Era, pp. 515-555. Tokyo: Japan Economic Research
Centre.
Takahashi, K. 1968. Nihon Kindai Keizai Keiseishi [History of the Formation of the
Modern Japanese Economy]. Tokyo: Toyo Keizai Shinposha.
. 1973. Nihon Kindai Keizai Hattatsushi [History of the Development of the
Modern Japanese Economy]. Tokyo: Toyo Keizai Shinposha.
Takeuchi, H. 1972. "Denki Kogyo [Electrical Industry]." In K. Kojima, ed., Nihon
Boeki no kozo to Hatten [Structure and Development of Japanese Trade]. Tokyo:
Shiseido.
Tamura, S., and S. Urata. 1988. "Changing Pattern of Technology Policy in Japan."
Paper presented at the 17th Pacific Trade and Development Conference, Bali, In­
donesia.
Tecson, G. R. 1985. Industrial Policy for International Competitiveness: A Case Study
of Industrial Policy in the Iron and Steel Industry of Japan. Unpublished Ph.D.
dissertation, Hitotsubashi University. Summary published in Hitotsubashi Review
94(2):113-118 (August 1985).

Bibliography 263
Toyo Keizai Shinposha. 1935. Nihon Boeki Seiran [Foreign Trade of Japan: Compre­
hensive Statistical Survey]. Tokyo: Toyo Keizai Shinposha.
Tran, V. T. 1985. "Vertical Integration of Japanese Firms in Southeast Asia: Case
Study of the Synthetic Fiber Industry [Nihon kigyo no Tonan Ajia deno
keiretsuka-gosen kogyo no kesu]," Nihon Keizai Kenkyu [J. Japanese Economic
Research!, No. 14, March.
. 1988. "Foreign Capital and Technology in the Catching-up Process of De­
veloping Countries: The Experience of South Korea's Synthetic Industry;' The
Developing Economies 26(4):386-402.
Tsuruta, Toshimasa. 1982. Sengo Nihon no Sangyo Seisaku [Japan's Industrial Pol­
icy after the Second World War]. Tokyo: Nihon Keizai Shinbunsha.
United Nations. 1947- . Monthly Bulletin of Statistics. New York: United Nations
Statistical Office.
Vernon, R. 1964. "International Investment and International Trade in the Product
Cycle," Quarterly J. Economics 80(2):190-207.
Yamaguchi, K., et a)., eds. 1974. "Mitsui Bussan Kaisha Hyakunenshi" [Hundred-
Year History of Mitsui and Co.). Unpublished draft.
Yamamoto, Y. 1969. "Kokusai Shushi Tokei no Choki Sogoka Ni Tsuite [Long-term
Estimate of Japan's Balance of Payments Statistics]," Jinbun Gakuho, Kyoto
University, 28 (March).
Yamamura, K. 1976. "General Trading Companies in Japan: Their Origins and
Growth." In H. Patrick, ed., Japanese Industrialization and Its Social Conse­
quences, pp. 161-199. Berkeley: University of California Press.
Yamazawa, I. 1969. "Kanzeikozo to boekihogo [Tariff Structure and Trade Protec­
tion]." In I. Yamada, K. Emi, and T. Mizoguchi, eds., Nihonkeizai no Kozohen-
do to Yosoku (Structural Changes and Forecast of the Japanese Economy!. Tbkyo:
Shunjusha.
. 1972. "Industry Growth and Foreign Trade: A Study of Japan's Steel Indus­
try;* Hitotsubashi J. Economics 12(2):41-59.
. 1975. "Industrial Growth and Trade Policy in Prewar Japan," The Develop­
ing Economies 13(l):38-65.
. 1984. Nihon no Keizai Hatten to Kokusai Bungyo [Japan's Economic De­
velopment and International Trade]. Tbkyo: Toyo Keizai Shinposha.
1988a. "The Textile Industry?' In R. Komiya et al., eds., Industrial Policy
of Japan, pp. 345-367. Tokyo: Academic Press Japan, Inc.
. 1988b. "Japan-U.S. Economic Conflicts and Their Impacts on the Pacific-
Asian Region" In R. A. Scalapino, ed., Pacific-Asian Economic Policies and
Regional Interdependence, pp. 219-239. Berkeley: University of California Press.

264 Bibliography
Yamazawa, I., and A. Hi rata. 1978. "Industrialization and External Relations: Com­
parative Analysis of Japan's Historical Experience and Contemporary Develop­
ing Countries* Performance," Hitotsubashi J. Economics, Vol. 18 (February).
Yamazawa, I., and H. Kohama. 1984. "Trading Companies and the Expansion of For­
eign Trade: Japan, Korea and Thailand." In K. Ohkawa and G. Ranis, eds., Japan
and the Developing Countries: A Comparative Analysis, pp. 426-446. New York:
Basil Blackwell.
Yamazawa, 1., and S. Tambunlertchai. 1985. "Manufactured Exports from Develop­
ing Countries: The Thai Textile Industry and Japan's Historical Experience." In
K. Ohkawa and G. Ranis, eds., Japan and the Developing Countries: A Com­
parative Analysis, pp. 369-388. New York: Basil Blackwell.
Yamazawa, I., and Y. Yamamoto. 1979a. Boeki to Kokusai Shushi [Trade and Balance
of Payments]. Choki Keizai Tokei [Long-Term Economic Statistics of Japan], Vol.
14. Tokyo: Toyo Keizai Shinposha.
. 1979b. "Trade and Balance of Payments." In K. Ohkawa and M. Shinohara,
eds., Patterns of Japanese Economic Development: A Quantitative Appraisal,
pp. 134-156. New Haven, CT Yale University Press.
Yasuba, Y. 1979. "Meijiki Kaiun Niokeru Unchin to Seisansei [Wages and Produc­
tivity of the Marine Transport Industry in the Meiji PeriodJ." In H. Shimpo and
Y. Yasuba, eds., Kindai-Ikoukino Nihon Keizai [Transition of the Japanese Econ­
omy in the Modern Period]. Tokyo: Nihon Keizai Shinbunsha.
Yen, K. W. 1983. "Export Assistance Regimes in the Pacific Asian Developing Coun­
tries: The Case of South Korea, Taiwan, the Philippines, and Thailand." Unpub­
lished paper presented to the Pacific Economic Cooperation Conference Task
Force on Trade in Manufactured Goods. Seoul: Korea Development Institute.
Yukizawa, K., and S. Maeda. 1978. Nihon Boeki no Choki Tokei: Boeki Kozoushi
Kenkyu no Kiso Sagyo [Long-Term Statistics of Japan's Trade: A Groundwork
for the Historical Study of Trade Structure]. Osaka: Dohosha.

Index
Action Program (1986), 222
Adjustment assistance. See Industrial
structure
Administrative Reform Council, 22S, 226
Agricultural protection. See Tariffs
Akamatsu, Kaname, 28, 33, 248nl; and the
"flying wild geese pattern," 13
Allen, G. C, 25ln6
Asahi Kenshoku. 80
Azuma Kogyo, 80
Baba, M, 28, 52, 249n5
Backward and forward linkage effects, in
industrial diversification, 107
Balance of payments: constraint on indus­
trial growth, 43-45; constraints, 25; of
Japan, 46/; long-term swings in, 23yig.
Bank of Japan, 182, 183, 199
Bensusan-Butt, D. M, 34
Brussels Tariff Nomenclature (BTN), 148,
175
"Buy-Japanese" movement, 158, 159
Capital inflows, 47
Capital outflows, long-term, 24
"Catching-up product cycle," (CPC), 13
Catching-up product cycle (CPC) develop­
ment: accumulation of managerial
resources, 31; in Asian developing coun­
tries, 233-236; automobile industry, 106,
107; basic type, 27-32; and capital ac­
cumulation, 34; changes in business be­
havior, 238; cotton textiles, reverse
import stage in. 75; crude steel, 234,
235//g.; decreasing unit costs in, 31; de­
fined, 30; and developing countries, 231; development stages, 29/ig., 30; direct
foreign investment (DFI) and technology
transfer, 35, 36; econometric analysis of,
98, 103; effect of terms of trade on, 36;
effects of, 23; entrepreneurship in, 34;
export stage of, 30; government policies,
effect of, 32; import substitution stage
of, 30; and industrial self-sufficiency, 48;
industrial machinery industry, 106; in­
troductory stage of, 30; iron and steel
industry, 87-108; kinked efficiency im­
provement in, 236; learning effect, 31,
103; the mature stage of, 30; mechanism
of, 92-96; model of, 95f; price equation,
95, 100, 104; and production efficiency,
236; reverse import stage of, 30; and
S/D curve, 105, 107; steel products,
104/; synthetic fabrics, 234, 235/ig.; tex­
tile industry, 69-86; variation I, 33; vari­
ation 2: international transfer of CPC,
35
Caves, R. E., 34, 248nl
Chen, E. K. Y., 28, 36
Chinese compradors, 119
Cobden-Chevalier Treaty (1860), 146
Commodity terms of trade, defined, 37,
38/fc
Comparative Analysis Project, 255nl
Constant market share (CMS) analysis, 60,
249«3; and commodity-composition ef­
fect, 56, 57; and competition effect, 56,
59; and market-demand-growth effect,
56, 57
Constant price series, 4
Cotton fabrics, 7, 71; export expansion of,
74; import substitution of, 74
Cotton industry, development of, 71-78

266 Index
Cotton industry development, linkage ef­
fects, 82-83
Cotton textile industry: catching-up product
cycle (CPC) development of, 73/ig.; stra­
tegic role of, 70-71
Cotton textiles: labor productivity in, 76;
and long-run cost reduction, 76; reverse
import stage in, 75
Cotton yarn, 71, 72/fg.; export expansion
of, 74; import substitution in, 74
Current account balance, components in, 24
"Deindustrialization." See Industrial struc­
ture, hollowing out of
Demonstration effect, 98
Destler, I. M., 195
Developing countries, 6, 238
Developing economies, 231-233
Development Bank of Japan, 218
Direct foreign investment (DFI). See under
Foreign investment
Direct trade, 117; defined, 250n2; expansion
of, 112; ratio of. 124/
Dualistic development pattern, 232
Dupont, 80
East Asiatic Company, 136
Economic Planning Agency, 173
Eighth Coal Policy (1987-91), 220
Electrification, 161
Employment adjustment, 219
Engerman, S. L., 250/12
European Atomic Energy Community (EU-
RATOM), 166, 168/
European Coal and Steel Community
(ECSC), 166, 168/
European Economic Community (EEC),
166, 168/
European Free-Trade Association (EFTA),
166, 168/
Export financing, preferential, 181-184
Export-oriented industrialization strategy,
silk industry as forerunner, 53
Export-promoting institutions, 186-188
Exports: commodity composition of, 7;
composition of industrial goods, 16/;
direct, 113/; direct, defined, 250n2;
direct, expansion of, 110, IHfig.; direct,
by manufacturers, 136; direct, ratio of,
120, 121/; expansion, 17, \9fig ., 31 (see
also Catching-up product cycle (CPC)
development); expansion in CPC de­
velopment, 30; expansion of cotton
fabrics, 74; expansion of cotton yarn,
74; foreign-exchange-earnings effect of
raw silk, 64; function, 100; growth of, and constant market share (CMS) analy­
sis, 56; to Manchuria, 128; ratio to
GNP, 6; ratio to production, 20; ratio to
production of pig iron and steel product
imports, 93/ig.; ratio to production, sig­
nificance of, 18; raw silk, 51-68; staple
goods, 51-52; vent for surplus theory, 52
External economies, 97
Extraterritorial privilege, 250n3; abolition
of, 110
Factor proportion theory, 33
Factoral terms of trade, defined, 37, 38/ig.
Factories, government-managed, 68
Fei, J. C. H., 247/T7, 255*1
Flying wild geese pattern, 13, 28, 248nl
Fogel. R. W., 250n2
Ford (Motor Co.), 164
Foreign exchange, allocation, 102
Foreign exchange management, 167-173
Foreign investment, direct (DFI), 28, 226,
253n8; and globalization of firm activi­
ties, 209; and heavy industry, 160; in
Japan, 160, 217; Japanese, 209; in
Japanese manufacturing, 162-163/; and
joint ventures, 160; liberalization of, in
Japan, 179; of textile industry, 84
Foreign Trade and Foreign Exchange Con­
trol Law (1949), 253n4; Automatic Ap­
proval (AA), 172, 174/, 175; Automatic
Import Quota (AIQ), 172, 174/, 175;
Import Quota (IQ), 172, 174/, 175
Forward and backward linkage, effects of,
34
Fujino, Shozaburo, 76, 249/i8; and Japan's
silk-reeling industries, 248/i2
General Agreement on Tariffs and Trade
(GATT), 166, 168/; Article 24, European
trade discrimination under, 167; Kenne­
dy Round, 169/; tariff principles, 167;
Tokyo Round, 177, 179; Uruguay
Round, 190, 223
General Motors, 164
General trading companies (GTCs). See
Trading companies, general (GTCs)
Generalized Scheme of Preferences (GSP),
169, 177, 197. 222
Gerschenkron, A., 136, 232
Global adjustment, 190-94
Gold standard, 45; Japan's adoption of, 59
Goto, A., 218, 219, 255n9
Government: assistance, 32; effect of pro­
tection, 108; market intervention, 171
Greater Japanese Association of Spinning
Firms, 77

Index 267
Hamaguchi, Osachi: Minister of Finance,
252n2; Prime Minister, 147
Havana Convention, 166
"Heavy and big" industries. See Industrial
structure
Heavy industrial goods, 13; imports of, 10
Heavy industry, 247«4
Henmi, K., 52, 66/
High-growth period, 165, 199
Hirata, A., 255/il
Hitachi. 160
Hong Kong and Shanghai Banking Corpo­
ration, 116
Imports: basic food, II; commodity compo­
sition of, 9/; composition of industrial
goods, 15/; direct, 114/; direct, defined,
250n2; direct, expansion of, 110, \l\fig .;
direct, ratio of, 120, 122/; function, 98;
growth in raw materials inputs, 11;
heavy industrial goods, 10; light indus­
trial goods, 10; "negative-list" formula,
168/; ratio to demand, 20; ratio to de­
mand, of pig iron and steel product
imports, 93fig.\ ratio to demand, sig­
nificance of, 18; ratio to GNP, 6; struc­
ture of, 8-9; substitution, 17, 19/fg., 25,
31 (see also Catching-up product cycle
(CPC) development); substitution in
catching-up product cycle (CPC) de­
velopment, 30; substitution of cotton
fabrics, 74; substitution in cotton yarn,
74
"Import-led structural change," 18
Inada, K., 34
Income Doubling Plan (1961-70), 171
India Cotton Transportation Arrangement,
77
Industrial adjustment. See Industrial
structure
Industrial diversification, mechanism of,
107
Industrial output, composition of, 14/
Industrial rationalization, 171
Industrial structure: adjustment assistance,
industry-specific, 220; Basic Stabilization
Program, 218; capacity reduction, 219;
changes in, 203; diversification of, 33,
105; "heavy and big" (H&B) industry,
202, 204; H&B firms, adjustment as­
sistance to, 218-219; H&B industries and
employment adjustment, 219; H&B in­
dustries, retrenchment of, 205-207; hol­
lowing out of, 223, 224; Law of
Temporary Measures to Stabilize Specif­
ic Depressed Industries, 218; "light and
small" (L&S) industry, 202, 204; L&S industries, growth potential of, 207-209;
L&S industries, and the Japanese Article
Number (JAN) code, 208; L&S indus­
tries, and point of sales (POS), 208;
L&S industries, promotion of, 221; re­
structuring, 202-217; Smaller Business
Switchover Law, 218; Special Measures
(of Facilitating Adjustment Law (1987),
219; Temporary Measures for the Struc­
tural Adjustment of Special Industries
(1983), 219
Industrial Structure Council, 171
Industry: government protection, 97; long-
run decreasing costs, 96, 97
Inflation, 102
International division of labor, between
Japan and its Asian neighbors, 211
International Monetary Fund (IMF), 168/;
Article 8, classification of Japan under,
169/; and Article 8 countries, 166; Japan
as Article 14 country, 253nl
International monetary system, role of, 45
International Trade Organization (ITO), 166
Iron-Manufacturing Encouragement Act
(1917), 88, 91/
Iron-Manufacturing Enterprise Act (1937),
91/, 92
Iron and steel industry, and capacity expan­
sion, 254n5
Japan Export-Import Bank, 181/, 186, 187
Japan Export Trade Organization, 181/, 186
"Japan, Inc.," 170
Japanese firm activities, globalization of,
212-215/
"Japanese Model" 3
Japanese trading companies. See Trading
companies
Japan's colonies, 90
Japan's manufacturing output, growth cycle
of. 41, 42fig., 43
Japan's manufacturing output coefficients
of structural change, 41, 42fig., 43
Jardine, Matheson and Company, 136
Johnson, H. G., 33
Jones, R. W., 34
Kagaku Sen-i Kyokai, 81
Kajinishi, M., 77, 249n8
Kansai Gomen, 132
Key, R, 45
Kobayashi, U, 158
Kohama, Hirohisa, 250, 255nl
Kojima, K., 28, 34. 37, 76. 78, 247/T5; and
the "catching-up product cycle," 13
Komiya, R., 170, 254n4
Korea, 92; general trading companies

268 Index
(GTCs), eligibility requirements and
preferential treatment, 241/; GTCs,
problems with, 242; GTCs, promotion
of, 240; relationship to Japan, 12
Korean War, 103, 180
Kuznets, S., 247*3; and foreign trade ratio
of Great Britain, France, Germany, and
Japan,6
Learning effect, in CPC development, 31
Levin, R. C, 218, 219, 255*9
Liberalization: Action Program, 195;
Market-Oriented Sector-Specific (MOSS)
talks, 196; "negative-list system" 180;
trade, phased implementation, 173-180;
trade and foreign exchange. Master Plan
for, 168/
Light industrial goods, imports of, 10
Light industry, 247n4
"Light and small (L&S)" industries. See
Industrial structure
Linder, S. B., 31
Lockwood, W. W., 249/i5
Long-Term Economic Statistics Series, 247nl
Machinery and equipment, renting vs. pur­
chasing, 254/i6
Macroeconomic performance, of Japan,
197, 198/
Maeda, S., 247/i5
Maekawa Group, 225
Maekawa Report, 197
Manchuria. 92, 128
Mason, F. R., 54
Master Plan for Trade and Foreign Ex­
change Liberalization, 173, 253n8
Matsushita, 136
Meier, G. M., 248nl
Merchant house trade, Meiji period, 109
MITI, 172
Mitsubishi, 125, 128
Mitsui Bussan, 78
Mitsui and Company, 117, 125, 128
Morikawa, H., 136
Multi-factor, multi-commodity model, 34
Multi-Fibre Arrangement (MFA) (1974),
195, 223
Multinational corporations (MNCs), 236
Myint, H., 248nl
Nakagawa, K., 109, 136
Nawa, T., 76
Newly industrializing economies (NIEs), 27
Nihon Boseki Kyokai, 77
Nihon Keizai Shinbun (The Japan Eco­
nomic Journal), 255n8 Nihon Mempu Yushutsu Kumiai, 78
Nihon Yusen Kaisha, 110
Nippon Steel Company, 90, 91/, 92
Nissan, 136
Niyomka, P., 66/, 249n4
Official direct assistance (ODA), 226
Ohkawa, K., 247n7, 255nl, 255n2
Ohkita, S., 28
Oil shocks: first, 197; first, effect of, 200;
second, 197, 219; second, effect of, and
disinflationary response, 200; second,
and the Temporary Measures for the
Structural Adjustment of Special Indus­
tries Law (1983), 219
Ono, Akira, 68, 76; and Japan's silk-reeling
industries, 248/i2
Organisation of Economic Co-operation
and Development (OECD), 166, 168/,
225; and complaints of Japanese export
strategy, 195
Organisation of European Economic Co­
operation (OEEC), 166, 168/
Oriental Banking Corporation, 116
Original Equipment Manufacturing (OEM),
210
Osaka Boseki, 74
Pasha, W., 28
Patrick, H., 167
Peck, M. J., 218, 219, 255n9
Policy: direct trade, 109; Eighth Coal Policy
(1987-91), 220; export promotion, 159,
180-186; export promotion, special tax
measures for, 181/; fiscal, for export pro­
motion, 184-186; foreign exchange allo­
cation, 102; foreign investment, 159-164;
government protection, 108; import
tariffs, effect on import substitution,
100; industry rationalization, 102; locali­
zation, 161, 164; macroeconomic 197;
macroeconomic policy mix, redirection
of, 201-202; "picking the winners," 171;
preferential purchasing, effect on import
substitution, 100-101; tariff, principles
of, 146; tariff reform, 175
Postwar reconstruction, and government
guidance, 171
Priority Production Policy, 101
Priority Production Project, 91/
Product cycle (PC) model, 28, 236
Protectionism: "buy-Japanese" movement,
158, 159; by government, 97; govern­
ment aid, 157; government purchases,
158; import quotas, 156; production
subsidies, 156. See also Tariffs

Index 269
Ranis, C, 247n7, 255«l
Raw materials, growth in imports of, II
Reconstruction Finance Corporation, 101
Reform, institutional, 225-226; Administra­
tive Reform Council, 225, 226; Maeka­
wa Group, 225; and restructuring,
225-226
Regression analysis, of direct trade ratios
and trade expansion, 119-125
Relative export price (REP), 40/, 41/; de­
fined, 37, 3Bfig.
Relative import price (RIP), 39, 40/, 41/;
defined. 37, 38/fe.
Remittance Control Law (1939), 161
Research and development (R&D): Agency
of Industrial Science and Technology
(AIST), 221; National Research and De­
velopment Program (Large-Scale R&D
Project), 221; promotion of, 221
Rhee Chong Inn, 255n3
Rosovsky, H., 170
Sanei Mempu Yushutsu Kumiai, 78
Sautter, C. 28
Saving, domestic, 45
Sericulture. 52; expansion of, 65; techno­
logical innovations in, 55
Services economy, 204
Shibusawa Eiichi, Baron, 158
Shinohara, M., 12-13, 28, 37, 170, 180,
248n5, 249/i5; and the "boomerang ef­
fect," 86
Shipbuilding Encouragement Act (1896),
157
Shoho Koshujo (Hitotsubashi University),
117
Silk: China's filature exports, 59; composi­
tion of U.S. import market, 54; ex­
ports, capital accumulation effect of,
67; exports, employment effect of, 66;
exports, and the income-and-output-
generation effect, 65; exports, linkage
effect of, 67; exports to the United
Stales, 59/; and foreign exchange earn­
ings, 64, 65/; foreign-exchange-earnings
effect of raw-silk exports, 64; hand fila­
ture, 58/; price elasticity of Japanese
silk, 64; raw-silk exports, 51-68; raw-
silk exports to the U.S. market, 57; role
of, 7-8; silk-reeling industry, 52; steam
filature, 58/
Silk industry: as forerunner of export-
oriented industrialization strategy, 53;
steam filature vs. hand filature, 55-56
Silk-reeling industry, expansion of, 65
Silver depreciation, 59, 62, 62fig.; effect
of, 60-61, 63; income effect of, 64; price effect of, 64
Smithsonian Agreement (1971), 187, 199
Special Measures for Facilitating Adjust­
ment Law (1987). 219
Specie: changes in, 46/, 47; movement of,
46/, 47
Staple theory of growth, 51
Steel: first reorganization program
(1951-55), 102; second rationalization
program (1956-60), 102
Steel-Manufacturing Encouragement Act
(1917), 157
Structural adjustment, 189-227 passim. See
also Industrial structure; Oil shocks
Subsidies, and inflation, 102
Supreme Export Council, 181/, 187
Susuki, 125, 128
Synthetic dyestuffs, 83
Taiwan, relationship to Japan, 12
Takahashi, K.. 250n4
Takashi, Masuda, 116
Takeuchi, H., 158
Tambunlertchai, S., 239, 255nl
Tamura, S., 221
Tariff Factory Law (1927), 150
Tariff Law, 141-142
Tariff rates, 150-153
Tariffs: ad valorem vs. specific duties, 149;
and agricultural protection, 144; an­
tidumping, 149; autonomy, 141, 146;
autonomy, recovery of, 77; Brussels
Tariff Nomenclature (BTN), 148; con­
ventional, 142; conventional, defined,
249n2; countervailing, 149; by economic
use, 154; effective rates, 177; emergency
surcharges, 146; First Emergency Spe­
cial Tariff law (1904), 142; general
tariff revision of 1926, 147; general
tariff schedule, disaggregation in, 148;
and industrial protection, 142-146; prin­
ciples of, 146; protection, history of,
141-146; retaliatory, 149; Second Emer­
gency Special Tariff law (1905), 142;
special, 142; Special Emergency Tariffs
(1904-05), 252/rl; structure, escalation
of, 148, 154, 177; structure of, nominal
vs. effective protective rates, 176/;
Tariff Factory Law (1927), 150; Tariff
Law (1906), 149; Temporary Storage
Yard Law (1900). 150
Tatemoto, M., 28. 52, 249/i5
Technology transfer, 28
Tecson, G. R.. 103, 185, 250n7
Teijin, 160
Temporary Storage Yard Law (1900), 150
Textile industry: catching-up product cycle

270 Index
(CPC) development in, 69 r86; direct for­
eign investment (DFI) in East Asia and
ASEAN, 86
Textiles: and chemical and synthetic fiber
development, 79; export/production ra­
tio, 80/ifg.; import/demand ratio, SO/ig.;
raw materials, diversification of, 79-82;
raw materials, export-production ratio
of, 79; raw materials, import-demand ra­
tio of, 79; synthetic fibers, 81; viscose
rayon, 80
Tokyo Marine Insurance Company, 110, 129
Toyo Rayon, 80
Toyota, 136
Trade conflict, 189, 194-197; with ASEAN,
197; with Asian countries, 197; factors
causing, 195-197; Kansai International
Airport project,' 196; and market liber­
alization, 222; with newly industrializing
economies (NIEs), 197; sectoral and
overall, 194-195; between United States
and Japan, 196
Trade-promoting services, 116, 128
Trading companies: comparative advantage,
117; European and American merchants,
117, 118; factors of trade potential, 116;
function of, 115; general (GTCs), 109,
240; GTCs, and developing countries,
231; GTCs, diversification strategy of,
132; GTCs, emergence of, 126-128;
GTCs. Japanese, 130/; GTCs, Japanese,
overseas branches of, 134-135/; GTCs,
Korea, eligibility requirements and
preferential treatment, 241/; GTCs,
Korean, problems with, 242; GTCs, and
Korean promotion of, 240; GTCs, or­
ganizing capability of, 133; GTCs, post­
war evolution of, 129; role of, 109; role
of Japanese, 115 Tran, V. T, 36, 85/
Treaty of Amity and Commerce, 117
Tsuruta, Toshima, 185
Two-factor, two-commodity model, 34
Urata, S., 221
U.S. Rice Millers Association, 196
US. Trade Act, Article 301, 196
U.S. Trade Expansion Act, 169/
Vent for surplus theory, 52
Vernon, R., 236; product cycle (PC) model
of, 28, 36
Voluntary export restraints (VERs), 195,
196, 222
World industrial production and trade in
industrial products, growth rate of, 170/
World trade, growth rate of, 5
Yamamoto, Y, 37, 45, 46/, 247n6
Yamamura, K., 109
Yamazawa, I., 28, 195, 239, 247n6
Yasuba, Y., 252nl7
Yawata Iron Works, 88, 91, 157
Yokohama Foreign Exchange Company, 67
Yokohama Specie Bank, 110, 116, 129
Yukizawa, K., 247n5
Zaibatsu, 88, 129, 137; defined, 249nl;
growth of, 90, 92

DISTRIBUTED BY
UNIVERSITY OF HAWAII PRESS
HONOLULU, HAWAII 96822
ISBN 0-86638-121-X