(Advances in Japanese Business and Economics 2) Takashi Negishi (Auth.) - Developments of International Trade Theory-Springer Japan (2014)

Advances in Japanese Business and Economics 2
Takashi Negishi
Developments
of International
Trade Theory
Second Enhanced Edition
Advances in Japanese Business and Economics 2
For further volumes:

http://www.springer.com/series/11682
Advances in Japanese Business and Economics
Editor in Chief:
RYUZO SATO
C.V. Starr Professor Emeritus of Economics, Stern School of Business,
New York University
Managing Editors:
HAJIME HORI, Professor Emeritus, Tohoku University
KAZUO MINO, Professor, Kyoto University
MARIKO FUJII, Professor, The University of Tokyo
Editorial Board Members:

TAKAHIRO FUJIMOTO MASAHIRO MATSUSHITA
Professor, The University of Tokyo Professor Emeritus, Aoyama Gakuin
University
YUZO HONDA
Professor Emeritus, Osaka University TAKASHI NEGISHI
Professor, Kansai University Professor Emeritus, The University of Tokyo
The Japan Academy
TOSHIHIRO IHORI
Professor, The University of Tokyo KIYOHIKO NISHIMURA
Professor, The University of Tokyo
TAKENORI INOKI
Professor Emeritus, Osaka University TETSUJI OKAZAKI
Special University Professor, Professor, The University of Tokyo
Aoyama Gakuin University
YOSHIYASU ONO
JOTA ISHIKAWA Professor, Osaka University
Professor, Hitotsubashi University
KOTARO SUZUMURA
KUNIO ITO Professor Emeritus, Hitotsubashi University
Professor, Hitotsubashi University The Japan Academy
KATSUHITO IWAI HIROSHI YOSHIKAWA
Professor Emeritus, The University of Tokyo Professor, The University of Tokyo
Visiting Professor, International
Christian University
Advances in Japanese Business and Economics showcases the research of Japanese scholars.
Published in English, the series highlights for a global readership the unique perspectives of
Japan’s most distinguished and emerging scholars of business and economics. It covers research of
either theoretical or empirical nature, in both authored and edited volumes, regardless of the sub-
discipline or geographical coverage, including, but not limited to, such topics as macroeconomics,
microeconomics, industrial relations, innovation, regional development, entrepreneurship, interna-
tional trade, globalization, financial markets, technology management, and business strategy. At
the same time, as a series of volumes written by Japanese scholars, it includes research on the
issues of the Japanese economy, industry, management practice and policy, such as the economic
policies and business innovations before and after the Japanese "bubble" burst in the 1990s.
Overseen by a panel of renowned scholars led by Editor-in-Chief Professor Ryuzo Sato, the series
endeavors to overcome a historical deficit in the dissemination of Japanese economic theory,
research methodology, and analysis. The volumes in the series contribute not only to a deeper
understanding of Japanese business and economics but to revealing underlying universal principles.
Takashi Negishi
Developments of
International Trade Theory
Second Enhanced Edition
123
Takashi Negishi
Member
The Japan Academy
Japan
Professor Emeritus
The University of Tokyo
Japan
ISSN 2197-8859 ISSN 2197-8867 (electronic)

ISBN 978-4-431-54432-6 ISBN 978-4-431-54433-3 (eBook)
DOI 10.1007/978-4-431-54433-3
Springer Tokyo Heidelberg New York Dordrecht London
© Springer Japan 2001, 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered
and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of
this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from Springer.
Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations
are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

To my dear Aiko, for our golden wedding
anniversary.
Preface
This book is based on my lectures on international trade in 1993–2000 at the School

of International Politics, Economics and Business, Aoyama Gakuin University,
Tokyo. According to the late Professor M. Bronfenbrenner, who did the same lec-
tures in 1982–1990, this school was established in 1982 based on the late Chancellor
Kinjiro Ohki’s vision of a new college of foreign studies, whose graduates should
be competent simultaneously in the English language, oral and written, and in some
branch of international affairs, international politics, international economics, or
international business. Aoyama Gakuin University, which was established in 1949,
originated in schools started in 1874 and 1879 by missionaries sent to Japan by the
Methodist Church of America to propagate Christianity through education.
To generate more interest in the subject, I started with the exposition of the
historical developments of the trade theory in Part I. Then, Part II deals with
the basic neoclassical theory of international trade and some of the more recent
developments. Since the school offered other lectures on international finance, I
could concentrate on the nonmonetary real problems of the international economics.
Some of the exercises given in each chapter and the appendices attached to several
chapters deal with more advanced graduate-level materials and some new research
results. So as to encourage students, most of whom were Japanese, I particularly
tried to discuss some of the recent contributions made by Japanese scholars.
It is my pleasant duty to thank, firstly, the board of editors, Professors Ryuzo
Sato, Rama V. Ramachandran, and Kazuo Mino, for their decision to include this
book in the series of Research Monographs in Japan–US Business and Economics,
and secondly, the Aoyama Gakuin University Society of International Politics,
Economics, and Business, headed by Professor Shigemi Honda, for the grant which
made the publication of the book possible.
January 2001 Takashi Negishi
vii
Preface to the Second Enhanced Edition
It is my pleasure to be able to publish the second enhanced edition of this book.

The first edition was published in 2001 by Kluwer Academic Publishers. Taking
advantage of this new edition, I decided to add the new Part III to the original two
parts, which is entitled as Historical Appendix and consists of two articles of mine,
“Adam Smith and Disequilibrium Economic Theory” and “Complete Specialization
in Classical Economics.” The first article was originally published in The Adam
Smith Review (2004), and the second in Economic Theory and Economic Thought
(edited by P. A. Samuelson and others (2010)). I am highly grateful to the editors of
these literatures for their permission to include these articles of mine into this book.
It is expected that readers who mastered Part I and Part II of this book can enjoy
reading these professional literature in Part III.
Tokyo, Japan Takashi Negishi

June 2013
ix
Contents
Part I Historical Progress of International Trade Theory
1 Mercantilism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 3
1.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 7
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 8
2 Specie-Flow Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 9
2.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 13
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 13
3 Adam Smith and Division of Labor . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 15
3.1 Problem.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 20
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 20
4 Ricardo and Comparative Costs . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 21
4.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 26
4.2 Appendix: Kojima on Ricardo . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 27
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 32
5 J. S. Mill and Reciprocal Demand . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 33
5.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 37
5.2 Appendix: Mill and Thornton . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 38
5.3 Problems for Appendix . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 42
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 42
6 Mill and Infant Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 43
6.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 48
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 48
7 Marx and International Exploitation.. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 51
7.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 58
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 59
xi
xii Contents
8 Marshall and Offer Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 61

8.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 66
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 66
Part II Modern Theory and Recent Developments

of International Trade
9 Theory of Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 69
9.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 73
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 74
10 Heckscher–Ohlin Theory (1) . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 75
10.1 Problem.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 80
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 80
11 Heckscher–Ohlin Theory (2) . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 81
11.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 85
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 86
12 Leontief Paradox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 87
12.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 91
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 91
13 Domestic Distortions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 93
13.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 97
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 98
14 Export Promotion and Welfare . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 99
14.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 103
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 103
15 Oligopoly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 105
15.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 108
15.2 Appendix: Tariffs Versus Quotas .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 108
15.3 Problems for Appendix . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 114
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 114
16 Immiserizing Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 115
16.1 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 119
16.2 Appendix: Second Best Problems .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 119
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 126
17 External Economies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 127
17.1 Problem.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 131
17.2 Appendix: Internal Economies . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 131
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 136
Contents xiii
Part III Historical Appendix
18 Adam Smith and Disequilibrium Economic Theory.. . . . . . . . . . . . . . . . . . . 141

18.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 141
18.2 Equilibrium Theory and Disequilibrium Theory . . . . . . . . . . . . . . . . . . 142
18.3 Smith’s Theory of International Trade .. . . . . . . .. . . . . . . . . . . . . . . . . . . . 143
18.4 The Division of Labor and the Extent of the Market .. . . . . . . . . . . . . 146
18.5 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 148
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 149
19 Complete Specialization in Classical Economics . . . .. . . . . . . . . . . . . . . . . . . . 151
19.1 Interpretations of Specialization . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 151
19.2 Ricardian Theory of Competitive Advantage... . . . . . . . . . . . . . . . . . . . 151
19.3 Ricardian Model of the Economy .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 152
19.4 Ricardo’s Theory of Gains from Foreign Trade . . . . . . . . . . . . . . . . . . . 154
19.5 Mill’s Assumption of the Single Factor of Production .. . . . . . . . . . . 155
19.6 The Two-Country, Two-Good Case . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 156
19.7 Bastable on Reciprocal Demand . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 158
19.8 Thornton on Supply and Demand Theory . . . . .. . . . . . . . . . . . . . . . . . . . 159
19.9 Pareto on Complete Specialization .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 161
19.10 Pareto’s Two-Country, Two-Good, One-Factor (Labor) Model . . 162
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 164
About the Author.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 167
Name Index .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 169
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 171

Part I
Historical Progress of International Trade
Theory
Chapter 1
Mercantilism
According to Schumpeter,1 the construction of scientific economics was started in

the late eighteenth century on two different foundations made in earlier periods.
The first one is the ancient and medieval economic thought of philosophers,
while the second one is popular arguments of current practical economic problems
in the sixteenth to eighteenth centuries (Schumpeter 1954, pp. 9–10). The medieval
theories of just price and usuary can be considered as representative examples of
the former. The latter is, of course, related to what is now called mercantilism.
Since it deeply concerned with problems of international trade, our explanation
of the development of the theory of international trade is also to start with the
consideration of mercantilism.
Mercantilism is a system of policies which aimed to create strong and centralized
nation states in European countries after the break-up of the medieval organization
of industry and commerce. Mercantilism literature, which discussed such policies
and went beyond it to consider economic principles, is diversified highly and
heterogeneous so that it is not possible to call mercantilism a school in the history of
economics. “Mercantilism was neither a scientific school nor a scientific theory—
there were then no schools at all in our sense of the word—and we distort the picture
if we seek already in this period what was in fact the consequence of a specialized
discipline after it had properly constituted itself” (Schumpeter 1954, p. 39). We
cannot discuss here the details of such diversified literature. From our point of view,
however, we may raise, at least, the following three points.
Firstly, Adam Smith,2 who criticized mercantilism systematically in his Wealth
of Nations (1776), defined it as a system of commerce based on then popular but
wrong concept of the wealth of a nation.
1 Joseph A. Schumpeter (1883–1950) was a professor at the University of Bonn and Harvard
University. He emphasized the role of the dynamic, innovating entrepreneurs in the development
of capitalism.
2 Adam Smith (1723–1790) was a professor of moral philosophy at the University of Glasgow and
founded the classical school of economics. See Chap. 3 for his Wealth of Nations.
T. Negishi, Developments of International Trade Theory, Advances in Japanese Business 3

and Economics 2, DOI 10.1007/978-4-431-54433-3__1, © Springer Japan 2014
4 1 Mercantilism
“That wealth consists in money, or in gold and silver, is a popular notion which
naturally arises from the double function of money, as the instrument of commerce
and as the measure of value.—A rich country, in the same manner as a rich man, is
supposed to be a country abounding in money; and to heap up gold and silver in any
country is supposed to be the readiest way to enrich it” (Smith 1776, p. 139).
“Some of the best English writers upon commerce set out with observing, that the
wealth of a country consists, not in its gold and silver only, but in its lands, houses,
and consumable goods of all different kinds. In the course of their reasonings,
however, the lands, houses, and consumable goods seem to slip out of their memory,
and the strain of their argument frequently supposes that all wealth consists in
gold and silver, and that to multiply those metals is the great object of national
industry and commerce.”
“The two principles being established, however, that wealth consisted in gold and
silver, and that those metals could be brought into a country which had no mines
only by the balance of trade, or by exporting to a greater value than it imported; it
necessarily became the great object of politicaloeconomy to diminish as much as
possible the importation of foreign goods for home-consumption, and to increase as
much as possible the exportation of the produce of domestic industry. Its two great
engines for enriching the country, therefore, were restraints upon importation, and
encouragements to exportation” (Smith 1776, p. 450).
According to the theory of price-specie-flow mechanism,3 however, it is impos-
sible to keep the result of such a favorable balance of trade in a country. If a country,
as a result of an export surplus, gains specie, her price level will rise, while the
opposite effect will take place in the rest of the world, which has lost specie on
account of its import surplus. Prices in the country are now too high to enable the
country to maintain her export surplus. Her high price level attracts imports from the
rest of the world while reducing her exports. The opposite will happen in the rest of
the world, to which there will be a reversal of the flow of specie. Even Thomas Man
(1571–1641), a representative mercantilist thinker, realized that “all men do consent
that plenty of money in a Kingdom doth make the native commodities dearer,—so it
is directly against the benefit of the Publique in the quantity of the trade; for as plenty
of money makes wares dearer, so dear wares decline their use and consumption”
(McCulloch 1954, p. 138). However, he did not hesitate to advocate the indefinite
accumulation of hard money. This is the so-called mercantilist dilemma. Even if
mercantilists thought that an increase in the supply of money is attended by a rise
in the demand for money and hence the volume of trade and not prices would be
directly affected by a specie inflow (Blaug 1985, p. 18), specie-flow mechanism
still works, since the import of a country depends positively on the level of her real
national product.
3 The theory of specie-flow mechanism was formulated by Richard Cantillon (1680–1734), an Irish
merchant banker in Paris, and David Hume (1711–1776), a Scottish philosopher. For Adam Smith
and this theory, see Petrella (1968). See also Chap. 2 for the further developments of the theory.
1 Mercantilism 5
If we emphasize mercantilists’ wrong view that the wealth of a country consists

in money, therefore, we have to reject mercantilism as a wrong economics. Policies
suggested by mercantilists are useless, since, even if we admit that it is nice to
increase the country’s stock of gold and silver, it cannot be increased through
international trade.
The second issue is the so-called primitive accumulation. Marx4 defines it as the
accumulation “preceding capitalistic accumulation; an accumulation not the result
of the capitalist mode of production, but its starting-point” (Marx 1954, p. 667). It is
“the historical process of divorcing the producer from the means of production. It
appears as primitive, because it forms the pre-historic stage of capital and of the
mode of production corresponding with it. The economic structure of capitalistic
society has grown out of the economic structure of feudal society. The dissolution
of the latter set free the element of the former” (Marx 1954, p. 668).
In other words, it is necessary to set up by exogenous forces a starting point
of a market economy which, once set up, can grow endogenously. It is called
the primitive accumulation, since the dissolution of pre-capitalistic self-sufficing
economy is necessary for the autonomous development of the capitalistic market
economy through the accumulation of capital. Many Marxian economists believe
that the absence of primitive accumulation is a principal cause of the current
difficulties of many underdeveloped countries.
After his life-long studies of mercantilism, a Japanese historian of economic
thoughts, Kobayashi5 (1976, pp. 335–425), concluded that mercantilism is to be
defined as economic theories of primitive accumulation. As a matter of fact,
Marx himself called James Steuart (1713–1780), the author of An Inquiry into the
Principles of Political Oeconomy (1767), the rational expression of the monetary
and mercantile system and argued as follows.
“His service to the theory of capital is that he shows how the process of
separation takes place between the conditions of production, as the property of a
definite class, and labour-power. He gives a great deal of attention to this genesis
of capital—without as yet seeing it directly as the genesis of capital, although he
sees it as a condition for large-scale industry. He examines the process particularly
in agriculture; and he rightly considers that manufacturing industry proper only
came into being through this process of separation in agriculture. In Adam Smith’s
writings this process of separation is assumed to be already completed” (Marx 1963,
p. 43).
Steuart entertains a strong view of the influence which statesmen can exercise
on the process of primitive accumulation. His strong influence cannot be denied
to the theory of economic policies of the historical school of economics. This
school was formed, against the classical school founded by Adam Smith, in the
nineteenth century in Germany, which was then a less developed country. From the
4 K. Marx(1818–1883) was the well-known leader of the socialism and the founder of the so-called
Marxian economics. See Chap. 7 for his own theory of international trade.
5 N. Kobayashi is a member of Japan Academy and a leading historian of economics in Japan.
6 1 Mercantilism
point of view of primitive accumulation, therefore, we can say that the economics of
mercantilism played significant historical roles in the development of the capitalism.
The final issue is Keynes’s view of mercantilism as a monetary economics.6
In his General Theory, he considered that mercantilism is “a doctrine which the
classical school has repudiated as childish but which deserves rehabilitation and
honour” (Keynes 1936, p. 351). For a country which has no mines, it is impossible
to increase the supply of money so that the economy can grow without deficiency
of the effective demand, which causes unemployment, unless she has the positive
balance of trade, which causes a specie inflow.
The balance of the aggregate demand and the aggregate supply in a country may
be described as
Y +M =C+I+X (1.1)
where Y and M denote, respectively, the domestic production of commodities and

the import of commodities (the supply from foreign countries) in the left-hand
side, while C, I, and X denote, respectively, the domestic consumption demand for
consumers’ goods, the domestic investment demand for producers’ goods, and the
export (the demand from foreign countries) in the right-hand side. If we assume that
C = cY , where 0 < c < 1 is the given propensity to consume, we have from (1.1)
I + (X − M)
Y= (1.2)
1−c
so that we can see that the effective demand for Y can be increased either by
increasing the domestic investment I or by increasing the trade surplus (the foreign
investment) (X − M) > 0. The investment I is a decreasing function of the rate of
interest r which is determined by the supply and demand of money,
L(r) = G (1.3)
where L is the liquidity preference of the people (demand for money) which is a
decreasing function of r, and G is the supply of money which is increased by the
trade surplus.
“At a time when the authorities had no direct control over the domestic rate
of interest or the other inducements to home investment, measures to increase
the favourable balance of trade were the only direct means at their disposal for
increasing foreign investment; and, at the same time, the effect of a favourable
balance of trade on the influx of the precious metals was their only indirect means
of reducing the domestic rate of interest and so increasing the inducement to home
investment” (Keynes 1936, p. 336).
6 J.
M. Keynes (1883–1946) initiated the so-called Keynesian Revolution, by his new theory of the
aggregate income determination in his General Theory.
1.1 Problems 7
“It is impossible to study the notions to which the mercantilists were led by
their actual experiences, without perceiving that there has been a chronic tendency
throughout human history for the propensity to save to be stronger than the
inducement to invest. The weakness of the inducement to invest has been at all
times the key to the economic problem. Today the explanation of the weakness
of this inducement may chiefly lie in the extent of existing accumulations; whereas,
formerly, risks and hazards of all kinds may have played a large part” (Keynes 1936,
pp. 347–348).
Keynes’s exposition of the mercantilist interpretation of economic phenomena
was, to a very large extent, based on the first edition of E. F. Heckscher’s
Mercantilism. In an appendix in the new edition (1955), however, Heckscher
criticized Keynes’s view on mercantilism. For example, Heckscher pointed out
that the basic flaw in Keynes’s interpretation is the belief that unemployment in
the mercantilist era was similar in character to unemployment recurring in the
industrialized economies. Unemployment caused by a fall in fixed investment
was virtually unknown before the Industrial Revolution. The predominant type
of industrial unemployment before the Industrial Revolution was mainly, if not
wholly, of the classical type which Keynes called voluntary or frictional (see
Heckscher 1955, pp. 340–358, especially, 354–356, and Blaug 1985, pp. 14–16,
31). The issue is, however, not completely settled yet, since there are a number of
economic historians who maintain that involuntary industrial unemployment, due
to the deficiency of effective demand, which was to a serious extent of monetary
origin, was a major problem in the seventeenth century England (Hutchison 1978,
p. 130).
1.1 Problems
1.1. Mercantilists considered that the supply of money in a country is directly

related to her stock of gold and therefore to her balance of trade. Is this still true
in our present world? Why?
1.2. According to Keynes, in the era of mercantilism, “the authorities had no
direct control over the domestic rate of interest or the other inducements to home
investment.” In the case of the authorities in our present world, what is the difference
from them? Explain.
1.3. By considering the balance of the aggregate demand and the aggregate supply
of a country, explain that the trade surplus of the country is equal to the difference
between the domestic income from production and the absorption (the domestic
consumption demand + the domestic investment demand).
8 1 Mercantilism
Bibliography
Blaug, M. (1985). Economic theory in retrospect. Cambridge: Cambridge University Press.

Heckscher, E. F. (1955). In E. F. Söderlund (Ed.), Mercantilism, vol. 2. London: George Allen and
Unwin.
Hutchison, T. W. (1978). On revolutions and progress in economic knowledge. Cambridge:
Cambridge University Press.
Keynes, J. M. (1936). The general theory of employment, interest and money. London: Mcmillan.
Kobayashi, N. (1976). Keizaigakushi chosakushu (Collected works on the history of economics),
vol. 1. Tokyo: Miraisha.
McCulloch, J. R. (Ed.). (1954). Early english tracts on commerce. Cambridge: Cambridge
University Press.
Marx, K. (1954). Capital, vol. I. Moscow: Progress Publishers.
Marx, K. (1963). Theories of surplus value, vol. I. Moscow: Foreign Language Publishing House.
Petrella, F. (1968). Adam Smith’s rejection of Hume’s price-specie-flow mechanism: a minor
mystery resolved. Southern Economic Journal, 34, 365–374.
Schumpeter, J. A. (1954). Economic doctrines and method (R. Aris, Trans.). London: George Allen
and Unwin.
Smith, A. (1776). An Inquiry into the nature and causes of the wealth of nations. Oxford: Oxford
University Press.
Chapter 2
Specie-Flow Mechanism
The classical economists, who followed Adam Smith, did not doubt that the
arguments of their predecessors, the mercantilists, in favor of a chronic export
surplus were based on an intellectual confusion. The classical refutation of the
mercantilist principle is derived from the so-called Cantillon–Hume price-specie-
flow mechanism. By this mechanism an inflow of bullion raises domestic prices,
and selling dear and buying cheap tends to turn the balance of trade against the
country. Purely automatic forces tend, therefore, to establish a natural distribution
of specie between the trading countries of the world and there is a level of domestic
prices such that each country’s value of exports equals that of imports.
The crux of the classical price-specie-flow mechanism is thus the change in
prices caused by redistribution of specie due to the trade imbalance. The famous
and concise statement in Hume’s essay Of the Balance of Trade runs as follows:
“Suppose four-fifths of all the money in Great Britain to be annihirated in one
night and the nation reduced to the same conclusion, with regard to specie, as
in the reigns of the Harrys and Edwards. What would be the consequence? Must
not the price of all labour and commodities sink in proportion, and everything be
sold as cheap as they were in those ages? What nation could then dispute with us
in any foreign market, or pretend to navigate or to sell manufactures at the same
price, which to us would afford sufficient profit? In how little time, therefore, must
this bring back the money which we had lost, and raise us to the level of all the
neighbouring nations?—Again, suppose that all the money of Great Britain were
multiplied fivefold in a night, must not the contrary effect follow? Must not all
labour and commodities rise to such an exorbitant height, that no neighbouring
nations could afford to buy from us, while their commodities, on the other hand,
became comparatively so cheap, that, in spite of all laws which could be formed,
they would be run in upon us, and our money flows out, till we fall to a level with
foreigners, and lose that great superiority of riches, which had laid us under such
disadvantages?”1
1 See Hume (1955, pp. 62–63). For Cantillon’s version, see Cantillon (1931, pp. 167–169).

10 2 Specie-Flow Mechanism
One might wonder why “the price of all the labour and commodities” rise in a
country which gained money and sink in a country which lost money, since the same
good has always the same gold price in different countries, if it is internationally
traded in the absence of obstacles. Such is the law of indifference. Staley (1976)
rightly argued that what Hume had in mind is a model of an economy in which
international trade takes place not continuously but discretely, so that the same good
can have different prices in different countries unless the international distribution
of gold has already settled in equilibrium.
In Hume’s day, it is certain that arbitrage took time to establish the law of
indifference internationally. If international trade does not take place quickly and
continuously, certainly prices rise temporarily not only for exportables and domestic
goods but also for importables in the gold gaining country. There is no reason
to assume that the adjustment process in international trade to establish uniform
prices is much quicker than the process of the specie-flow mechanism to achieve the
balance of trade equilibrium.
The traditional interpretation which follows Viner (1937, pp. 313–317) considers,
however, that uniform gold prices always prevail for identical commodities in
different countries. Since it is insisted on as the interpretation of the classical specie-
flow mechanism in general, to consider it is worthwhile, as Staley himself admitted,
independently of one’s view about the nature of the price changes envisioned by
Hume. As the same price change is now assumed to occur in all countries at
the same time, the price variations responsible for adjustment in the balance of
trade are changes in terms of trade, i.e., the relative price of the exportables and
the importables for countries. The price of the exportables must rise relative to that
of the importables in the gold-gaining country, and vice versa, if the classical price-
specie-flow mechanism works successfully.
Modern literature on international transfer has made it clear, however, that the
resultant changes in prices can be in either direction, depending on the international
difference in demand patterns, and are not necessarily in the direction suggested
by the classical price-specie-flow theory, that is, the terms of trade rise in the
surplus country and fall in the deficit country (Kemp 1964, pp. 79–81). If, for
example, two countries are identical in taste which can be expressed by a homothetic
social indifference map, so that Engel curves are identical straight lines through
the origin,2 the equilibrium prices are independent of the distribution of income
between the two countries, including the distribution of specie. In this case, as is
pointed out by Dornbusch, Fischer, and Samuelson (1976) , there is no price effect
associated with a redistribution of the world money supply and therefore no effects
on real variables in the adjustment process for monetary disequilibrium, contrary to
the classical price-specie-flow mechanism.
2 Indifferencemap is homothetic, when the slope of indifference curves remain unchanged by any
proportional changes in the quantities of all the commodities. Income consumption curve, which
shows how consumption varies if income increases and prices remain unchanged, is called Engel
curve, since a German statistician Engel (1821–1896) studied it originally.
2 Specie-Flow Mechanism 11
Let us construct a drastically simplified version of the model used by Dornbusch,

Fischer, and Samuelson. For the sake of simplicity, we consider the case of a
two-good, two-country model, in which each country completely specializes in the
production of the exportables.3 The production is of constant returns to scale with
respect to the sole factor of production, called labor. As for the demand side, it is
assumed that the level of aggregate expenditure of each country is proportional to
the supply of money in the country4 and that the ratio of expenditure on each good
to the aggregate expenditure is a given constant.5 The sum of supplies of money in
the two countries is assumed to be constant.
The condition for the equilibrium of demand and supply of labor in the home
country is then
M M
wL = aV G + a∗V ∗ 1 − G (2.1)
G G
where L is the given supply of labor, w is the money rate of wage, a is the given
ratio of expenditure on the exportables of the home country, V is the constant
velocity of the circulation of money, M is the domestic money supply, G is the
given world money supply, variables and parameters with(out) asterisk are those
of foreign(home) country, and the rate of foreign exchange is assumed to be 1.
Similarly, for the labor in the foreign country, we have
M M
w∗ L∗ = (1 − a)V G + (1 − a∗)V ∗ 1 − G. (2.2)
G G
If the distribution of specie, M, is given, we can solve (2.1) and (2.2) for w and w∗ .
If two countries have identical taste, such that a = a∗ and V = V ∗ , furthermore,
it is easily seen that equilibrium w and w∗ are independent of the distribution of
specie, M.
The specie-flow mechanism is given as
dM
= wL − V M (2.3)
dt
where t denotes time and dM/dt signifies the instantaneous rate of change in M.
The supply of money M is increased as a result of the trade surplus that is equal to
the difference of income wL and absorption V M (see Problem 1.3). Since w remains
unchanged when M is changed, if two countries are identical in taste, it can easily
be seen that the solution of (2.3), M(t) is stable in the sense that it approaches to
wL/V and the trade balance is eventually established. In Fig. 2.1, M is measured
horizontally, and dM/dt, vertically. A downwardly sloping line AA signifies the
3 Dornbusch, Fischer, and Samuelson 1976 considered the case with infinitely many goods.
4 See Dornbusch and Mussa (1975), where such a behavior of expenditure is explained by
intertemporal optimization.
5 See Problem 2.1.
12 2 Specie-Flow Mechanism
Fig. 2.1 The specie-flow mechanism
solution of (2.3). If M > B = wL/V , M decreases through time and if M < B =

wL/V , it increases, as the arrows indicate, so that it always approaches to B where
the balance of trade is established.
Since the price of each good is completely determined by the wage cost in
our model, there is no price effect of specie flow in this special case. Something
must be done to explain the changes in prices in the direction suggested by the
classical price-specie-flow theory. Dornbusch, Fischer, and Samuelson showed that
even in this special case the introduction of non-traded domestic goods revitalizes
the classical conclusion that in the adjustment process prices decline along with
the money stock in the deficit country while both rise in the surplus country. Let
us therefore introduce non-traded goods in our model and assume that the ratio
of expenditure on non-traded goods in each country is constant; that is, (1 − k).
Non-traded goods and the exportables are produced in each country but there is still
no import competing production. In view of identical taste, then, (2.1) and (2.2) are,
respectively, modified into
M
W L = aV G + (1 − k)V G (2.4)
G
and
M
W ∗ L∗ = (k − a)V G + (1 − k)V 1 − G (2.5)
G
from which w and w∗ are obtained.6 Now equilibrium wages are no longer
independent of the distribution of specie. An increase in M increases w and
reduces w∗ . The prices of goods produced in a country change in the same direction
as the supply of money in the country. Since we have from (2.4)
6 Numerical values of a in (2.4) and (2.5) are different from those in (2.1) and (2.2).
Bibliography 13
dw V
= (1 − k) (2.6)
dM L
the right-hand side of (2.3) is decreasing with respect to M and therefore the price-
specie-flow mechanism is stable.
2.1 Problems
2.1. Suppose the utility function of a consumer is
U = X aY b
where X and Y are, respectively, the quantity of commodities X and Y , and a and b
are positive constants such that a + b = 1. When U is maximized, being subject to
the budget constraint,
pX + qY = Z
where p, q, and Z signify, respectively, the price of X, the price of Y , and the given
income of the consumer, express X and Y as functions of p, q, and Z. Show that
“the ratio of expenditure on each commodity to the aggregate expenditure is a given
constant.”
2.2. Draw the indifference map between X and Y in the case of the utility function
given in Problem 2.1, and show that it is a homothetic case where “Engel curve is a
straight line through the origin.”
2.3. What are typical examples of non-traded goods?
2.4. Solve (2.3) explicitly for M as a function of t and discuss its stability.
Bibliography
Cantillon, R. (1931). In H. Higgs (Ed.) Essai sur la nature du commerce en general. London:
Macmillan.
Dornbusch, R. S., Fischer, S., & Samuelson, P. A. (1976). Comparative advantage, trade, and
payments in a Ricardian model with a continuum of goods. American Economic Review, 67,
823–839.
Dornbusch, R. S., & Mussa, M. (1975). Consumption, real balances and the hoarding function.
International Economic Review, 16, 415–421.
Hume, D. (1955). In E. Rotwein (Ed.), Writings on economics. London: Nelson.
Kemp, M. C. (1964). The pure theory of international trade. Engelwood Cliffs, NJ: Prentice-Hall.
Staley, C. E. (1976). Hume and Viner on international adjustment mechanism. History of Political
Economy, 8, 252–265.
Viner, J. (1937). Studies in the theory of international trade. New York: Harper.
Chapter 3
Adam Smith and Division of Labor
Adam Smith, who criticized mercantilists for their wrong concept of the wealth
of nations, considered that the real wealth is the annual produce of the land and
labor of the society (Smith 1776, p. 12). According to Smith, furthermore, the high
productivity of labor in civilized and thriving nations seems to have been the effects
of the division of labor. International trade certainly pre-supposes the division of
labor among different countries. As for the nature and causes of international trade,
therefore, we can expect to learn very much from Smith’s theory of the divisions
of labor. For this purpose, let us have a glance of the content of his The Wealth of
Nations (Smith 1776), which consists of five Books, i.e.,
Book 1: Of the Causes of Improvement in the productive Powers of Labour,
and of the Order according to which its Produce is naturally
distributed among the different Ranks of the People.
Book 2: Of the Nature, Accumulation, and Employment of Stock.
Book 3: Of the different Progress of Opulence in different Nations.
Book 4: Of Systems of political Oeconomy.
Book 5: Of the Revenue of the Sovereign or Commonwealth.
An outline of each Book now follows.
Since “the greatest improvement in the productive powers of labour and the
greater part of the skill, dexterity, and judgement with which it is any where directed,
or applied, seem to have been the effects of the division of labour” (Smith 1776,
p. 13), Book 1 starts with the famous exposition of the division of labor in the
production of pins and that of nails in Chap. 1. It is insisted “that the Division of
Labour is limited by the Extent of the Market” in Chap. 3, and Chap. 4 discusses
“the origin and use of money,” since people have to exchange their products as
the division of labor established and a commodity is chosen as money to avoid the
inconvenience of barter trades.
If we consider a modern society where all land is private property and stocks
are used in production, profits should be left to those employers who use their

16 3 Adam Smith and Division of Labor
own stocks after wages are paid to employees and rents are paid to landowners.1
Chapter 6 insists that the price of a commodity consists of three parts: wages, rents,
and profits. Chapter 7, then, discusses the determination of the price of commodities.
There are an ordinary or average rate of wages, rents, and profits which are regulated
by the general circumstances of society. Smith calls these rates the natural rates.2
“[W]hen the price of any commodity is neither more nor less than what is
sufficient to pay the rent of the land, the wages of the labour, and the profits of
the stock employed in raising, preparing, and bringing it to market, according to
their natural rates, the commodity is then sold for what may be called its natural
price” (Smith 1776, p. 72).
The effectual demand is then defined as the demand from those who are willing
to pay the natural price, i.e., the central price to which actual market prices are
attracted. When the quantity of a commodity which is brought to market falls short
of the effectual demand, the market price will rise more or less above the natural
price, each component part of the price must rise above its natural rate, and more
land, labor, and stock are used in raising, preparing, and bringing to market the
commodity so that the quantity brought to market is sufficient to supply the effectual
demand. This implies, then, that the market price falls to the natural price and each
component part to its natural rate, respectively. Similarly, when the quantity brought
to market exceeds the effectual demand, the market price sinks below the natural
price and each component part falls below its respective natural rate, with the result
that the quantity of labor, land, and stock used are diminished, the quantity of the
commodity is equalized to the effectual demand, and the natural price is regained
with the natural rates of its component parts. This equilibrium theory of Chap. 7
was evaluated very high by Schumpeter (1954, p. 189) as the one which points to
the modern theory of general equilibrium.
While Book 1 of The Wealth of Nations is a theory of prices and distributions,
Book 2 is a theory of capital and development. In the Introduction of Book 2, it is
emphasized that the accumulation of stock is necessary for the development of the
division of labor. The problematic chapter in Book 2 is Chap. 5, “Of the different
Employment of Capitals” where Smith insists on the natural order of investment.
It is the order of agriculture, manufacture, domestic trade, and foreign trade, which
constitutes the theoretical foundation of Smith’s arguments on economic history
and economic policy in Book 3 and Book 4 of The Wealth of Nations. Smith’s own
explanation of this important theory is, however, very much confused as is judged
by Kobayashi (1977, chapter 7) that “it is almost entirely bankrupt.”3 In view of the
important role assigned to this theory in the system of The Wealth of Nations, we
1 Smith defined capital as the stock which is used to obtain profit.

2 The natural rates of wages, rents, and profits are considered to reflect, respectively, the demand
and supply conditions of labor, land, and stock, since the natural rate of wages is, for example,
defined as high in a progressive society and low in a declining one.
3 For example, Smith argued that returns to capital is higher in agriculture than in manufacture,
since the nature works along with men in the former (Smith 1776, p. 363).
3 Adam Smith and Division of Labor 17
have to reconstruct it on the basis of Smith’s theorem of division of labor that capital
accumulation is the prerequisite to the development of the division of labor (Negishi
1985, Chapter 3 and Vassilakis 1987).
“[W]hen the division of labour has once been thoroughly introduced, the produce
of a man’s own labour can supply but a very small part of his occasional wants. The
far greater part of them are supplied by the produce of other men’s labour, which
he purchases with the produce—of his own. But this purchase cannot be made till
such time as the produce of his own labour has not only been completed, but sold. A
stock of goods of different kinds, therefore, must be stored up somewhere sufficient
to maintain him, and to supply him with the materials and tools of his stock till such
time, at least, as both these events can be brought about” (Smith 1776, p. 276).
Because the natural order of investment is concerned with how the accumulation
of capital leads to improvement in productivity due to the division of labor, it is
clear that investment must start in a most unspecialized, self-sufficing industry and
gradually proceed so that industries are more and more subdivided and specialized.
Agriculture can be regarded as such a most inclusive and most unspecialized
industry if we include household and coarser manufactures into agriculture. This
all-inclusive agriculture cannot be divided into an independent manufacture and
agriculture unless a certain stock of capital is accumulated. When enough capital
is accumulated to support a manufacture as an independent, specialized industry,
however, investment should be and actually is made to develop the occasional jobs
in the neighborhood of artificers into a regular manufacture for more distant sale.
As capital accumulates more, the division of labor advances to interdistrict
specialization of local agricultures and manufactures. Interdistrict specialization
requires still larger capital accumulation, since returns from such specialized
industries for distant sale are very slow. Investment in home trade should be done,
therefore, only when the accumulation of capital has already reached the stage when
interdistrict specialization is possible. The highest stage of the division of labor is
that of international trade based on the international division of labor. Only in this
last stage is investment in foreign trade relevant. Since the returns of foreign trade
are very seldom so quick as those of the home trade, the required domestic capital
accumulation is larger than in the case of home trade.
Following static and dynamic economic theories in Book 1 and Book 2, Book
3 explains the economic history of various nations. Its basic point of view is, of
course, the natural order of things, agriculture, manufacture, and foreign commerce,
and it is concluded as follows. “Though this natural order of things must have taken
place in some degree in every such society, it has, in all the modern state of Europe,
been, in many respects, entirely inverted” (Smith 1776, p. 380).
Book 4 is Smith’s theory of economic policies. On the basis of the arguments
in Books 1–3, Smith insists on laissez-faire, and discusses mercantilism (the com-
mercial System) and physiocracy (the agricultural System). The criticism against
the former is particularly thorough and is concluded as follows. “It cannot be very
difficult to determine who have been the contrivers of this whole mercantile system;
not the consumers, we may believe, whose interest has been entirely neglected;
but the producers whose interest has been so carefully attended to—”(Smith 1776,
p. 661).
Incidentally, the famous “invisible hand” is discussed in Chap. 2 of Book 4.
“Every individual necessarily labours to render the annual revenue of the society
as great as he can. He generally, indeed, neither intends to promote the public
interest, nor knows how much he is promoting it.—by directing that industry in
such a manner as its produce may be of the greatest value, he intends only his own
gain, and he is in this, as in many other cases, led by an invisible hand to promote
an end which was no part of his intention” (Smith 1776, p. 456).
In the beginning of Book 4, Smith argued that the object of political economy
is, firstly, to provide a plentiful revenue for the people, and secondly, to supply the
state with sufficient revenue. Book 5 is concerned with this second object of political
economy, i.e., public finance. The expenses of the state, the revenue of the state, and
public debts are discussed. Cheap government is insisted on and the role of the state
is limited to defense, justice, public works such as road construction and education.
Smith’s theory of gains from trade is two-sided, i.e., the so-called vent for surplus
theory and the theory of increasing returns due to further division of labor.
“Between whatever places foreign trade is carried on, they all of them derive two
distinct benefits from it. It carries out that surplus part of the produce of their land
and labour for which there is no demand among them, and bring back in return for
it something else for which there is a demand. It gives a value to their superfluities,
by exchanging them for something else, which may satisfy a part of their wants
and increase their enjoyments. By means of it, the narrowness of the home market
does not hinder the division of labour in any particular branch of art or manufacture
from being carried to the highest perfection. By opening a more extensive market
for whatever part of the produce of their labour may exceed the home consumption,
it encourages them to improve its productive powers, and to augment its annual
produce to the utmost, and thereby to increase the real revenue and wealth of the
society” (Smith 1776, pp. 446–477).
As for the theory of vent for surplus, it seems that the existence of such a surplus
itself is inconsistent with the economy Smith considered, where we can rely on the
invisible hand of the efficiently working well-developed markets (Bloomfield 1975;
Elmslie 1998). As Myint (1958) rightly argued, of course, the theory is informative
in analyzing underdeveloped economies in which capital and labor cannot move
from sector to sector so easily.
From our point of view, however, what is more interesting is Smith’s theory of
economies of scale or increasing returns based on the division of labor, which he
emphasized is limited by the extent of the market. This is because the modern
theory of international trade has been developed, until quite recently, with the
assumption of diminishing returns. With such an assumption, we can explain trade
between dissimilar countries, or trade in primary products, for which differences
in exogenously given conditions like climate, production functions (technology),
3 Adam Smith and Division of Labor 19
or factor endowments are important to explain the comparative advantage.4 Trade

between similar countries, or trade in manufactures, however, cannot be explained
in terms of such differences. It is the increasing returns suggested by Smith which
we hope can explain such trades by explaining endogenously the comparative
advantages. A country can export a commodity simply because the scale of her
domestic industry is so large that it can supply the commodity cheaply, i.e.,
increasing returns.
Smith gives two different kinds of illustrations of the divisions of labor; one is
concerned with the subdivision of different operations to produce a given product,
the extent of which is limited by the demand for output of a firm or a plant, while the
other is concerned with an inter-firm division of labor or the specialization of firms
in the same industry, the extent of which is limited by the demand for the industry.
The illustration of the former division of labor is drawn from pin making in The
Wealth of Nations (Smith 1776, pp. 14–15).5 Let us consider a simple model of a
firm which can explain such a division of labor (Negishi 2000).
Consider a firm in an industry producing a homogeneous product. For the sake
of simplicity, let us suppose that only the labor is necessary to produce the product.
The entrepreneur can divide the production process into many operations so that
a laborer can specialize in a limited number of them. Given the total number of
laborers to be employed, m, the entrepreneur decides the degree of the division
of labor (number of operations to be assigned to a laborer) so as to maximize
the average productivity of labor, a(m), which is an increasing function of m, i.e.,
a < 0. Then, the level of the output of the firm is x = a(m)m and the average cost
of production is wm/a(m)m = w/a(m), where w is the given rate of wage.
The average cost of the firm is a decreasing function of the level of output, since
dm 1
= (3.1)
dx a + a m
and
w/a(m) wa
=− 2 < 0. (3.2)
dx a (a + am)
With respect to the division of labor within a firm, modern economic theory’s
evaluation of Smith was not very high (see for example Richardson 1975). While
diminishing cost caused by such division of labor must produce concentration
and, in the end, monopoly, Smith was not troubled by this inconsistency between
competition and increasing returns. This evaluation was based, however, on the
Walrasian view of competition in the neo-classical general equilibrium theory. In
4 Comparative advantage determines which country can export which commodity in international
trade. See Chap. 4.
5 As for the latter division of labor, the explanation of the increasing returns is not so simple, as is
argued by Negishi (2000).

this view, a firm’s scale of production should be limited by the increasing cost of
production, since it can sell whatever amount of product at the given unchanged
market price. Smith should be evaluated higher, however, if we follow a more
recent view of competition, i.e., Sraffian view, in which a firm’s scale of production
is limited, not by the cost, but by the deficiency of demand (see Negishi 1985,
Chapter 2, Negishi 1998, and Whitaker 1990).6
As far as an increase in the demand for the industry, not only induces the entry
of new firms, but also expands the demand for and therefore the scale of production
of the existing firms, it reduces the cost and price of the product (see Chap. 17 for
the modern theory of international trade based on increasing returns to scale).
3.1 Problem
3.1. By using some proper statistics on world trade, demonstrate that most interna-
tional trade is carried on between large industrial areas that are very similar.
Bibliography
Bloomfield, A. I. (1975). Adam Smith and the theory of international trade. In A. S. Skinner, & T.
Wilson (Eds.), Essays on Adam Smith (pp. 455–481). Oxford: Oxford University Press.
Elmslie, B. T. (1998). Vent for surplus. In H. D. Kurz, & N. Salvadori (Eds.), The Elgar companion
to classical economics (pp. 504–507). L-Z, Cheltenham: Edward Elgar.
Kobayashi, N. (1977). Kokuhurontaikei no seiritsu (Formation of the system of wealth of nations).
Tokyo: Miraisha.
Myint, H. (1958). The “Classical Theory” of international trade theory and the underdeveloped
countries. Economic Journal, 68, 317–337.
Negishi, T. (1985). Economic theories in a non-Walrasian tradition. Cambridge: Cambridge
University Press.
Negishi, T. (1998). Sraffa and the microfoundations of Keynes. European Journal of the History of
Economic Thought, 5, 452–457.
Negishi, T. (2000). Adam Smith’s division of labor and structural changes of an industry. Structural
Change and Economic Dynamics, 11, 5–11.
Richardson, G. B. (1975). Adam Smith on competition and increasing returns. In A. S. Skinner, &
T. Wilson (Eds.), Essays on Adam Smith (pp. 350–360). Oxford: Oxford University Press.
Schumpeter, J. A. (1954). History of economic analysis. Oxford: Oxford University Press.
Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations. Oxford: Oxford
University Press.
Vassilakis, S. (1987). Increasing returns to scale. In J. Eatwell, M. Milgate, & P. Newman (Eds.),
The new palgrave dictionary of economics, vol. 2 (pp. 761–765). London: Macmillan.
Whitaker, J. K. (1990). Marshall’s theories of competitive price. In R. M. Tullberg (Ed.), Alfred
Marshall in retrospect (pp. 29–48). Hants: Edward Elgar.
6 Walras(1834–1910) is the founder of the neo-classical general equilibrium theory, while Sraffa
(1898–1983) was critical to the mainstream neo-classical economics and a leader of the so-called
neo-Ricardian school of economics.
Chapter 4
Ricardo and Comparative Costs
David Ricardo (1772–1823) was the representative theorist of the classical school
of economics which was initiated by Adam Smith. Ricardo’s theory of comparative
advantage developed in Chap. 7 of his Principles (1817) has been one of the few
theories that economists of all the different schools understand and agree with.
Although the current mainstream economics, the neo-classical school, has been
developed from the marginal revolution1 against the classical school, it cannot
be denied that Ricardo’s theory of comparative advantage is still an important
cornerstone of the modern theory of international trade and has been studied
intensively by many leading scholars of the neo-classical economics.2 Such modern
interpretations of Ricardo are, however, sometimes very much different from what
Ricardo really meant, as will be seen below.
Ricardo’s exposition of his theory of comparative costs is given by the use of his
famous numerical example. Consider an international economy composed of two
countries, England and Portugal, and two commodities, cloth and wine. Suppose
one unit of cloth made in England is being exchanged against one unit of wine
made in Portugal.
“England may be so circumstanced, that to produce the cloth may require labour
of 100 men for one year, and if she attempted to make wine, it might require the
labour of 120 men for the same time. England would therefore find it her interest
to import wine, and to purchase it by the exportation of cloth. To produce the wine
in Portugal might require only the labour of 80 men for one year, and to produce
the cloth in the same country might require the labour of 90 men for the same time.
It would therefore be advantageous for her to export wine in exchange for cloth”
(Ricardo 1951a, p. 135).
1 Revolution made in 1870s by W. S. Jevons (1835–1882) in Manchester, C. Menger (1840–1921)

in Wien and L. Walras (1834–1910) in Lausanne by their systematic use of marginal concepts, like
marginal utilities, marginal costs, etc.
2 A recent example is Dornbusch, Fischer, and Samuelson (1977).

22 4 Ricardo and Comparative Costs
Fig. 4.1 The situation of

England
England has the comparative advantage in cloth, and Portugal, in wine. This
exchange takes place notwithstanding that the commodity imported by Portugal
could be produced there with less labor than in England and that England gives
the produce of the labor of 100 men for the produce of the labor of 80. What does
matter is the comparative costs. England can export cloth since she can produce it
with the relative cost (to wine) cheaper than Portugal, i.e., 100/120 < 90/80.
What should we imagine as the model of an international economy, which is
hidden behind Ricardo’s famous four numbers, 100, 120, 90, and 80? The modern
standard interpretation is the so-called Ricardian model, where labor is assumed to
be only factor of production, which requires remuneration, and the input-coefficients
of labor are given constants, i.e., Ricardo’s four numbers. Labor is assumed to
be immobile internationally and the size of labor population is given constant for
each country. Figure 4.1 shows the situation of England. The quantity of cloth
is measured vertically, and that of wine, horizontally. The point A signifies the
maximum volume of production of cloth, which can be obtained by using all the
labor in the cloth production. In other words, OA = L/100 in Ricardo’s numerical
example, when the labor population in England is L. Similarly, the point B shows
the maximum quantity of wine, which can be obtained by using all the labor in its
production, i.e., OB = L/120. Then, any point on the line segment AB can show
the maximum possible combination of cloth and wine possible to produce under
the given labor population, L. In autarky, i.e., if there is no international trade, AB
also shows the maximum possible combination of cloth and wine which can be
consumed in England.
Suppose that England is specialized completely in the production of cloth for
which she has the comparative advantage and is situated at A in Fig. 4.1. If she
exports cloth in return for the import of wine of the same units, as is suggested by
Ricardo in the above quotation, England can start from the point A in Fig. 4.1 and
proceed on the downwardly sloping broken line with the slope of 45◦ . Since this
line is located above and to the right of the line AB, England can consume larger
4 Ricardo and Comparative Costs 23
amount of both cloth and wine with international trade than in autarky. There exist
gains from international trade. This is also true in the case of Portugal.
In the so-called Ricardian model of the modern standard interpretation of the
comparative costs theory of Ricardo, only the labor is the factor of production and
there exists neither land nor capital. Therefore, the only cost of production is the
labor cost, and all the commodities produced are distributed among laborers so that
the G.N.P. is composed only of the wage income. According to Ricardo’s numerical
example, however, the labor productivity is higher in Portugal than in England not
only in the production of cloth, but also in that of wine. Then, Samuelson (1972,
pp. 678–9) is quite right to accuse Ricardo for his odd economic geography. In the
heyday of England’s industrial revolution, Ricardo selected Portugal as the superior
of England in every respect, having a real per-capita G.N.P. that is somewhere
between one-ninth and one half greater depending upon whether you are a drunkard
or a dandy. Why this odd economic geography? Was it noblesse oblige, to give the
other fellow the advantage?
What is odd is, however, not Ricardo’s economic geography, but the so-called
Ricardian model of the modern interpretation of his comparative cost theory.
Ricardo declared in Preface of his Principles that the principal problem of the
political economy is to determine the laws which regulate the distribution of G.N.P.
among landowners, capitalists, and laborers under the names of rent, profit, and
wages (Ricardo 1951a, p. 5). In the true Ricardian model of the economy which
exists behind the four numbers of Ricardo’s theory of comparative costs, therefore,
there must exist land and capital, in addition to labor, as the factors of production
which require the remuneration.
The economic growth implies, for Ricardo, the accumulation of capital and the
increase of labor population. Since land is given, however, the marginal productivity
of capital and labor declines as the result of the economic growth. In other words,
these marginal productivities are lower than the average ones. Now Ricardo’s four
numbers must be interpreted to show the level of the marginal productivity of labor,
rather than the average productivity. As the result of the economic growth, the
marginal productivity of labor becomes lower in England than in Portugal, but the
average productivity is high. Since the population of a country is largely dominated
by the labor population, therefore, per-capita G.N.P. is higher in England than in
Portugal. This is because the large land rent income, which results, as land becomes
more scarce, from the difference between the average and marginal productivities
of labor. G.N.P. consists, not only from the wage income, but also from profit and
rent incomes. Ricardo’s economic geography is not odd, therefore, from the point
of view of the true Ricardian model of the economy.
There is another difficulty with the so-called Ricardian model of the modern
interpretation of Ricardo. Ricardo simply considered the terms of trade, i.e., the
exchange ratio of cloth and wine, is equal to one. The terms of trade is, however,
dependent in general on the demand conditions, as, for example, the relative price
of cloth against wine rises when the demand for cloth increases relatively both in
England and Portugal against wine. There exists, however, no consideration of the
demand side conditions in the so-called Ricardian model. Therefore, the modern
interpreters conclude, Ricardian theory of international trade is an incomplete theory

which cannot determine the term of trade.3 Let us show that this is a wrong
conclusion.
In the so-called Ricardian model of the modern interpretation, wages of labor in
England (specialized in cloth production) rise when the demand for the labor, i.e.,
the demand for cloth both in England and Portugal increases. This is, however, a
quite un-Ricardian result and clearly shows how this model is different from the
true Ricardian model. Ricardo considered that the equilibrium wage is the subsistent
wage, which is just sufficient to purchase a given collection of goods, and is
independent of the demand for labor (Ricardo 1951a, p. 93). An increase in demand
for labor is considered to be absorbed in the increase in the labor population
(Malthusian principle of population). An increase in the world demand for the
product to which she is specialized, therefore, is absorbed in the increase in the
labor population of a country. Wages in each country is independent of demand and
are given by the historical and social factors. To make the story simple, let us assume
that such subsistence wages are identical in England and Portugal and just sufficient
to purchase the given amount of each good, i.e., cloth and wine.
Ricardo’s four numbers are considered to show the marginal productivities of
labor so that we can consider only the production on the marginal land (the least
productive land used, i.e., land with no rent).4 The price of a good, then, can be
explained by the wage cost and profit on it, without consideration of land rent. The
wage cost of a unit of cloth produced in England is 100W, where W is the rate of
wage. Similarly, the wage cost of a unit of wine produced in Portugal is 80W . This is
because, if we disregard transportation cost, prices of cloth and wine are equalized
in both countries, and therefore, the subsistence wage in each country is equalized
to the same W .
Wage cost is advanced by capitalists. Let us denote the rate of net profit to be
paid to the advanced capital by R in England and by S in Portugal. Since the price of
a good to be sold is equal to the sum of the wage cost and profit, the price P of a unit
of cloth to be produced in England is equal to (1 + R)100W and similarly the price
Q of a unit of wine to be produced in Portugal is equal to (1 + S)80W. The terms
of trade is the ratio of P and Q, which is now equal to the ratio of 100(1 + R) and
80(1 + S). To determine the terms of trade, therefore, we have to know the relation
between the rate of profit of each country.
If the capital is perfectly immobile internationally, there exists no relation
between the rates of profit of two countries. If goods are traded between them,
however, there must exist some mobility of capital between countries. This is
because capital is also necessary to do export and import business. Suppose that
the rate of profit is lower in England than in Portugal as a result of the accumulation
3 In Japan, however, Kojima has insisted on to the contrary, by considering a three-good (cloth,
wine, and gold) model. See Appendix to this chapter.
4 Ricardo explained land rent by the difference in productivity of land. With equal wages and the
rate of profit, surplus left in a more productive land must be paid to the landowner as the rent.
4 Ricardo and Comparative Costs 25
of capital in the former. Competition between capitals induces capitals in England

to move from the production of cloth to the export business of cloth and import
business of wine, capitals in Portugal, from the import of cloth and the export of
wine to the production of wine. Then, the rate of profit in the production of cloth in
England rises, and the rate of profit in the production of wine in Portugal, falls.
If the capital mobility is perfect between countries, then, the rate of profit is
equalized everywhere so that we have R = S. The mobility is, however, not so
perfect, as Ricardo emphasized.
“Experiences, however, shows, that the fancied or real insecurity of capital, when
not under the immediate control of its owner, together with the natural disinclination
which every man has to quit the country of his birth and connexions, and intrust
himself with all his habits fixed, to a strange government and new laws, check the
emigration of capital. These feelings—induce most men of property to be satisfied
with a low rate of profits in their own country, rather than seek a more advantageous
employment for their wealth in foreign nations” (Ricardo 1951a, pp. 136–137).
Capitalists prefer the safe investment at home with a low rate of profit to the risky
investment abroad with a higher rate. In other words, high rates of profit abroad
must be discounted by a risk premium to be compared with the rate of profit at
home. Suppose that capitalists in England where the rate of profit is lower decide
whether to move their capital by the comparison between the low gross rate of profit
(1 + R) in England with the high gross rate (1 + S) in Portugal multiplied by the risk
premium A < 1, i.e., A(1 + S), then, capital moves until the condition (1 + R) =
A(1 + S) is established.
The terms of trade is the ratio of the price of cloth P and that of wine Q in
Ricardo’s numerical example, which is the ratio of 100(1 + R) and 80(1 + S). If we
use the discounting risk premium A, it is the ratio of 100A and 80. When A is equal
to 0.8, then, the terms of trade is equal to 1, as Ricardo considered in his explanation
of the gains from trade by the use of his famous four numbers (see Negishi 1982 and
Gandolfo 1986, pp. I, 28–32).
Finally, there exists the problem of the complete specialization. As is seen in
Fig. 4.1, England must be specialized completely in the production of cloth and give
up the domestic production of wine entirely, if we follow modern interpretation
of Ricardo’s theory. In other words, there are, then, no problems of the import-
competing industries. If we interpret that cloth represents manufactured goods in
general and wine, agricultural products, did Ricardo insist on such a radical program
as to abolish agriculture in England? Fortunately it was not so.5 Before to write
his Principles, Ricardo began his economics by joining in the bullion controversy
(1809–12) and in the controversy on the corn law (1813–15). In the latter, he was
against the law which restricts the import of corn and protects the agriculture.
Ricardo considered, however, that under the free trade England should undoubtedly
become a great manufacturing country, but she should remain a great agricultural
5I owe this point to Professor K. Mizuta of Higashi Nihon Kokusai University.

country also, as was assured by him later in his speech in the House of Common
(Ricardo 1952, p. 180). Furthermore, he also explained it in his pamphlet On
Protection to Agriculture (1822).
“[a]n objection which is frequently made against freedom of trade in corn
founded on the supposition that we should be importers of a considerable portion
of the quantity which we annually consume.—[I]t appears that no very great
quantity could be obtained from abroad, without causing a considerable increase
in the remunerating price of corn in foreign countries. In proportion as the quantity
required came from the interior of Poland and Germany, the cost would be greatly
increased by the expenses of land carriage. To raise a larger supply, too, those
countries would be obliged to have recourse to an inferior quality of land, and as
it is the cost of raising corn on the worst soils in cultivation requiring the heaviest
charges, which regulates the price of all the corn of a country, there could not
be a great additional quantity produced, without a rise in the price necessary to
remunerate the foreign grower. In proportion as the price rose abroad, it would
become advantageous to cultivate poorer lands at home; and therefore, there is every
probability that, under the freest state of demand, we should not be importer of any
very large quantity” (Ricardo 1951b, pp. 264–5).
Even in his Principles, furthermore, Ricardo argued “that a country possessing
very considerable advantages in machinery and skill, and which may therefore be
enabled to manufacture commodities with much less labour than her neighbours,
may, in return for such commodities, import a portion of the corn required for its
consumption, even if its land were more fertile, and corn could be grown with
less labour than in the country from which it was imported” (Ricardo 1951a,
p. 136). The country imports a portion, not the whole, of the corn, the complete
specialization in the manufacture is not considered. It is natural for Ricardo, who
considered the importance of the existence of lands with different fertility in
agriculture. Labor productivity changes with the fertility of lands on which labor
is applied. Ricardo’s four numbers should be interpreted as the labor productivities
on the marginal land, i.e., the least fertile land currently used. What is suggested is
to give up the domestic production only on such a land. The complete specialization
shown in Fig. 4.1 is, therefore, very un-Ricardian. It is entirely the result of the
modern interpretation which assumed away the existence of land and considered
that labor productivities are given constants.
4.1 Problems
4.1. By using a figure similar to Fig. 4.1, discuss the gains from trade for Portugal
in Ricardo’s numerical example.
4.2. Explain why, if you are a drunkard (a dandy), Portugal has a real per-capita
G.N.P. one half (one-ninth) greater than that of England in Ricardo’s numerical
example.
4.2 Appendix: Kojima on Ricardo 27
4.3. Assume that a units of cloth and b units of wine must be bought by the
subsistence wage, and denote the price of cloth by p, the price of wine by q, and
the rate of profit by r. Behind Ricardo’s numbers, then, the so-called price–cost
equations, p = (1 + r)100(ap + bq), q = (1 + r)120(ap + bq) must be satisfied on
the marginal land in England before trade. Write similar equations in the case of
Portugal and demonstrate that the rate of profit is lower in England than in Portugal
before trade (Negishi 1982; Gandolfo 1986, pp. 1–29).
4.4. Let us modify Ricardo’s numerical example that the labor of only 90 men is
necessary to produce a unit of wine on the intra-marginal land (with land rent) but
120 men are necessary, as Ricardo assumed, on the marginal land (without land
rent). Redraw Fig. 4.1 for this modified case.
4.2 Appendix: Kojima on Ricardo
Kojima6 has been insisting that Ricardo’s theory is not incomplete, for it is
possible to determine the terms of trade without introducing reciprocal demand,
i.e., the mutual demand between countries (Kojima 1951; Negishi 1996). Kojima’s
Ricardian model is, however, different from what we called the true Ricardian model
in the text of Chap. 4. Following what we called there the modern interpretation of
Ricardo, he also assumes that the labor is the only factor of production and that
the labor productivity is constant with respect to the level of production. What he
differs from the modern interpretation is that the gold as the international currency
is introduced and the specie-flow mechanism as well as the labor value theory of
gold is considered in his model.
As a matter of fact, Ricardo himself considered these problems. “Gold and
silver having been chosen for the general medium of circulation, they are, by the
competition of commerce, distributed in such proportions amongst the different
countries of the world, as to accommodate themselves to the natural traffic which
would take place if no such metals existed, and the trade between countries were
purely a trade of barter” (Ricardo 1951a, p. 137). “The metal gold, like all other
commodities, has its value in the market ultimately regulated by the comparative
facility or difficulty of producing it” (Ricardo 1951a, p. 193).
In addition to the famous numerical example of the pure barter trade, which we
discussed in Chap. 4, Ricardo also considered a numerical example of monetary
trade. “[S]uppose before the improvement in making wine in England, the price of
wine here were 50 l. per pipe, and the price of a certain quantity of cloth were 45 l.,
whilst in Portugal the price of the same quantity of wine was 45 l., and that of the
same quantity of cloth 50 l.; wine would be exported from Portugal with a profit of
6 K.Kojima is a leading theorist of international economics in Japan and now is professor emeritus
of Hitotsubashi University.
Table 4.1 Labor cost of

commodities in two countries England Portugal
Cloth 100 men 90 men
Wine 110 men 80 men
Gold 100 men 80 men
Table 4.2 Natural price

(money price) of England Portugal
commodities after trade Cloth 45 45
Wine 45 45
Gold 45 45
5 l., and cloth from England with a profit of the same amount. Suppose that, after the
improvement, wine falls to 45 l. in England, the cloth continuing at the same price”
(Ricardo 1951a, p. 138).
Kojima considered this example of monetary trade. By a series of numerical
examples, he described the process of the determination of the equilibrium terms
of trade between cloth and wine by the labor value of money and the distribution
of money between countries. Numerical examples, however, can describe the
equilibrium but cannot demonstrate its existence. This is why he cannot insist that
Ricardian theory can determine the terms of trade. To insist it, we have to construct
a formal model in which we can determine the equilibrium terms of trade and from
which Kojima’s numerical example can be generated.
Kojima started his series of numerical examples with Table 4.1 which gives the
amount of labor necessary to produce a unit of cloth, wine, and gold in England and
Portugal, corresponding to Ricardo’s numerical example of the case of trade with the
use of gold as money (before the improvement). The situation after the international
trade is, then, described by Table 4.2,7 since “it is the natural price of commodities
in the exporting country, which ultimately regulates the prices at which they shall be
sold,—, in the importing country” (Ricardo 1951a, p. 375). Having identical price,
gold is neither exported nor imported. The trade of cloth and wine is balanced at the
terms of trade of 1 : 1.
Table 4.1 is changed to Table 4.38 after the improvement in making wine in
England. Since no wine can be exported to England, Portugal has to pay gold for
her import of cloth. “But the diminution of money in one country, and its increase in
another, do not operate on the price of one commodity only, but on the prices of all,
and therefore the price of wine and cloth will be both raised in England, and both
lowered in Portugal” (Ricardo 1951a, pp. 139–140). After such adjustments, Kojima
considered, the final situation described by Table 4.4 will appear as an equilibrium
of international trade.9 By the comparison of Tables 4.3 and 4.4, it is clear that gold
7 Table 4.2 corresponds to Table 3 in Kojima (1951).

Table 4.3 Labor cost of

commodities in two countries England Portugal
Cloth 100 men 90 men
Wine 100 men 80 men
Gold 100 men 80 men
Table 4.4 Natural price

(money price) of England Portugal
commodities after trade Cloth 46.5 46.5
Wine 43.5 43.5
Gold 43.5 43.5
is now produced only in Portugal, since the labor cost of gold production is higher
than its natural price in England. England specializes in the production of cloth
while wine and gold are produced in Portugal. The terms of trade for England (the
price of cloth/the price of wine) is now 1.07 : 1 (i.e., 46.5/43.5 : 1).
Let us now construct our formal model from which Kojima’s numerical examples
can be generated. Consider international trade of cloth and wine between England
and Portugal. Let ace , awe , and age denote, respectively, the unit labor cost of cloth,
wine, and gold in England. Similarly, let acp , awp , and agp denote, respectively, those
in Portugal. Since England is specialized in the production of cloth and Portugal, in
the production of wine, the terms of trade P/Q must satisfy
acp P ace
> > (4.1)
awp Q awe
where P and Q are, respectively, the price of cloth and that of wine in terms of gold.
Let G and M signify the world stock of gold and its distribution to England.
International distribution of gold, which is used exclusively for money, can be
explained by the quantity theory of money,
PLe
= ve M (4.2)
ace
and
QL p
= v p (G − M) (4.3)
awp
where ve and v p denote the constant velocity of the circulation of the money and
Le and L p denote the given supply of labor, respectively, in England and Portugal.
If conditions (4.2) and (4.3) are satisfied, trade between England and Portugal is
balanced and there is no movements of gold between countries.
Suppose first that gold is produced only in Portugal.10 Then,
awp 11
Q= (4.4)
agp
and from (4.2), (4.3), and (4.4)
(agp v p G − L p )ve ace

P= . (4.5)
v p Le agp
We can determine, then, the equilibrium terms of trade P/Q from (4.4) and (4.5) as
P (agp v p G − L p)ve ace

= (4.6)
Q v p Le awp
without introducing reciprocal demands of cloth and wine.

To assure that there is no gold production in England,
ace
P> (4.7)
age
must be satisfied. In view of (4.5), then, this condition requires that
(agp v p G − L p)ve agp 12

> . (4.8)
v p Le age
If the right-hand side and left-hand side of (4.8) are equal, gold can be produced in
both countries.
Similarly, the terms of trade can be determined as
P ve L p ace
= (4.9)
Q (age ve G − Le )v p awp
if we suppose that gold can be produced only in England. If the right-hand side and
the left-hand side of (4.8) are equal,
Le Lp
G= + (4.10)
ve age v p agp
10 More exactly, gold can be produced, if there is a demand, as profitably as the other commodity.
11 The right-hand side of (4.3) denotes the nominal value of the national product of Portugal, which,
in view of (4.4), remains unchanged if she is also producing gold.
12 See Negishi (1996) for the further details, i.e., other conditions on parameters in the form of
inequalities.
and gold can be produced in both countries. Only in such a case, both P/Q in (4.6)
and (4.9) are equalized to ace agp /age awp . Otherwise, however, the terms of trade
when gold is produced in Portugal (4.6) is different from the terms of trade when
gold is produced in England (4.9), even if all the parameters are identical.
Now Kojima’s numerical examples in Tables 4.2 and 4.4 can be easily generated
from our model. Suppose that the velocity of the circulation of money is equal in
England and in Portugal. By taking the units properly, we can make ve = v p = 1. Let
us also assume that the supply of labor is equal in each country and we can make
Le = L p = 1, 000.
Assume first that the labor cost of cloth, wine, and gold in England and in
Portugal, i.e., ace , awe , age , acp , awp , and agp is given as figures in Table 4.1. If
we assume that gold is produced in Portugal, 1/45 units of gold is contained in a
unit of money, and the world stock of gold G is equal to 180/8, we can see that
P = 1, Q = 1, and P/Q = 1 from (4.4), (4.5), and (4.6), so that money prices of
cloth, wine, and gold are all 45 as is shown in Table 4.2. Since the right-hand side
and left-hand side of (4.8) are equal, furthermore, gold is also produced in England
and the terms of trade of 1 : 1 is determined uniquely in this case.
Secondly, assume that ace , awe , age , acp , awp , and agp are given as figures in
Table 4.3. Still assuming that gold is produced in Portugal, let us now suppose that
a unit of money contains 1/43.5 units of gold as in Table 4.4.13 Now the terms of
trade P/Q is no longer that of 1 : 1, but changed to that of 1.07 : 1 (46.5/43.5 : 1)
in Table 4.4. To explain this change, let us increase the world stock of gold G from
180/8 = 22.5 to 23.2 (= 1856/80) so as to satisfy (4.5). This is reasonable, since
in the disequilibria of transition from Tables 4.1 to 4.3, the diminution of money
reduces the cost of gold production in Portugal and it is profitable to produce gold
and to convert it into money. Since (4.8) is satisfied, there is no gold production in
England. If there is no increase in G, of course, the terms of trade of 1 : 1 remains
unchanged, though it coincides with the cost ratio in England as is seen in Table 4.3.
Then, cloth and wine are produced in England and wine and gold, in Portugal.
Now let us return to the situation of Table 4.1, but assume that a unit of money
contains 1/45 units of gold and G = 22.8 (= 1824/80). If we suppose that gold is
produced in Portugal, we have P/Q = 1.03 from (4.6). If we suppose, however, that
gold is produced in England, we have P/Q = 0.98 from (4.9). The terms of trade
can be determined uniquely, if we specify which country produce gold. Generally,
the terms of trade cannot be fully determinate in this Kojima model of Ricardo, since
different values are generated, depending on which country we specify to produce
gold. Furthermore, labor supplies are, in fact, not exogenous parameters in the strict
sense of the word, since they have to be implicitly assumed as adjusted instantly to
changes in the reciprocal demands (see Negishi 1996).
13 See Negishi (1996) for the case in which a unit of money contains still 1/45 units of gold.
Bibliography
Dornbusch, R. S., Fischer, S., & Samuelson, R. A. (1977). Comparative advantage, trade and
823–839.
Gandolfo, G. (1986). International economics. Berlin: Springer.
Kojima, K. (1951). Ricardo’s theory of international balance of payments equilibrium. The Annals
of the Hitotsubashi Academy, 2(1), 76–92.
Kojima, K. (1996). Trade, investment and pacific economic integration. Tokyo: Bunshindo.
Negishi, T. (1982). The labor theory of value in the Ricardian theory of international trade. History
of Political Economy, 14, 199–210.
Negishi, T. (1996). Japanese studies of Ricardo’s theory of foreign trade. The Japanese Economic
Review, 4784, 335–345.
Negishi, T. (2000). Economic thought from Smith to Keynes. Aldershot: Edward Elgar.
Ricardo, D. (1951a). On the principles of political economy and taxation. Cambridge: Cambridge
University Press.
Ricardo, D. (1951b). Pamphlets and papers, 1815–1823. Cambridge: Cambridge University Press.
Ricardo, D. (1952). Speeches and evidence. Cambridge: Cambridge University Press.
Samuelson, P. A. (1972). The collected scientific papers, vol. 3 (pp. 678–679). Cambridge, MA:
MIT Press.
Chapter 5
J. S. Mill and Reciprocal Demand
J. S. Mill (1806–1873)s Principles of Political Economy (1848) was written as

“a work similar in its object and general conception to that of Adam Smith, but
adapted to the more extended knowledge and improved ideas of the present age”
(Mill 1909, p. xxviii). It was highly successful as the last of the great books of the
classical economics founded by A. Smith. From the point of view of the history
of international trade theory, it is, in general, to be remembered by its extension of
Ricardo’s theory of comparative costs to take account of the effects of reciprocal
demand on the terms of trade. We must emphasize, however, that Mill seems to start
the so-called modern interpretation of Ricardo, which we criticized in Chap. 4.
Chapter 18 of Mill’s Principles, which contains his theory of international trade,
consists of two parts, the original first five sections and Sects. 6–9 which he added
in the third edition (1852). According to Edgeworth (1894),1 “[t]he splendid edifice
of theory constructed in the first five sections is not improved by the superstructure
of later date which forms the latter part of the chapter. This second story does not
carry us much higher.”
Mill opened the superstructure as follows. “Thus far had the theory of interna-
tional values been carried in the first and second editions of this work. But intelligent
criticisms (chiefly those of my friend Mr. William Thornton), and subsequent further
investigation, have shown that the doctrine stated in the preceding pages, though
correct as far as it goes, is not yet complete theory of the subject matter” (Mill 1909,
p. 596). The theory of international value discussed in the first sections is, of course,
that of reciprocal demand.
“This Law of International Value is but an extension of the more general law of
Value, which we called the Equation of Supply and Demand.—the supply brought
by the one constitute his demand for what is brought by the other. So that supply
and demand are but another expression for reciprocal demand: and to say that value
1 F.
Y. Edgeworth (1845–1926). Professor of Oxford University, who founded, with Walras, the
modern general equilibrium theory.

34 5 J. S. Mill and Reciprocal Demand
will adjust itself so as to equalize demand with supply, is in fact to say that it will
adjust itself so as to equalize the demand on one side with the demand on the other”
(Mill 1909, pp. 592–3).
Why, then, is this not complete? The reason is, Mill argues, “that several
different rates of international value may all equally fulfill the conditions of this
law” (Mill 1909, pp. 596–7). According to Mill, the existence of such a portion
of indeterminateness in the rate at which the international values would adjust
themselves indicates that not the whole of the influencing circumstances have yet
been taken into account. To supply such deficiency, Mill takes into considerations,
in addition to the quantities demanded in each country of the imported commodities,
“the extent of the means of supplying that demand which are set at liberty in each
country by the change in the direction of its industry” (Mill 1909, p. 597).
As for the former, Mill assumes a unit own-elasticity of demand with respect
to price, zero cross-elasticities of demand with respect to price and a unit income
elasticity of demand that “any given increase of cheapness produces an exactly
proportional increase of consumption; or, in other words, that the value expended
in the commodity, the cost incurred for the sake of obtaining it, is always the same,
whether that cost affords a greater or smaller quantity of the commodity” (Mill
1909, p. 598). In other words, the proportion in which the total income is to be spent
on each commodity is a given constant, irrespective of the level of income and the
prices of commodities.
For the two-country (England and Germany) two-good (cloth and linen) case,
then, Mill can demonstrate that the relative international value (the terms of trade)
is uniquely determined. Let us assume that England (Germany) has the comparative
advantage in the production of cloth (linen), and England (Germany) is specialized
in the production of cloth (linen) after trade. The terms of trade t (the price of cloth
in terms of linen after trade) is solved from
pm
n= (5.1)
t
where m is “the cloth previously (i.e., before trade) required by Germany, (at the
German cost of production),” n is “the quantity of cloth which England can make
with the labour and capital withdrawn from the production of linen (after trade)”
and p is “the cost value of cloth (as estimated in linen) in Germany” (Mill 1909,
pp. 600–1).2
See Fig. 5.1 which describes the situation of Germany. The quantity of linen is
measured vertically, and that of cloth, horizontally. The maximum quantity of linen
Germany can produce is OB1 and that of cloth, OA1 . The point G indicates the
production and consumption of cloth and linen in Germany before trade (at autarky).
Then, Mill’s m is equal to OC. The slope of the line A1 B1 is p, i.e., p = OB1 /OA1 .
2 As was pointed out and corrected by Chipman (1979), however, Mill made a slip and could not
derive (5.1) correctly.
5 J. S. Mill and Reciprocal Demand 35
Fig. 5.1 Germany
1
1
Fig. 5.2 England
Similarly, Fig. 5.2 describes the situation of England. The maximum quantity
of linen England can produce is OB2 and that of cloth, OA2 . Suppose the point
E indicates the production and consumption of cloth and linen in England before
trade (at autarky). If England is specialized in the production of cloth after trade,
then, Mill’s n is equal to DA2 .
Then (5.1) can be explained as follows. German expenditure on cloth before
trade is pm in terms of linen, since p is also the before trade price of cloth in
terms of linen there. Now German demand for English cloth after trade is pm/t,
from the assumption of the unit own-elasticity of demand with respect to price
which is now changed from p to t, while German income in terms of linen remains
unchanged before and after trade at OB1 in Fig. 5.1. This German demand should
be equal to the after trade supply of cloth from England, which is equal to n by
definition. Equation (5.1) expresses the equality of demand and supply of cloth in
the international market. We can solve (5.1) for the terms of trade which will prevail
after trade, i.e., t, from the data available to us before trade, i.e., p, m, and n, if we
assume that each country is specialized completely after trade.
Chipman (1965, 1979) evaluated Mill’s solution of t from (5.1) very high,
as the historically first demonstration of the existence and the uniqueness of the
equilibrium of demand and supply. In the original first five sections of Chap. 18 of
Mill’s Principles, the equilibrium of international trade is certainly well described
by the principle of reciprocal demands, but it is merely described by numerical
examples and its existence is not yet demonstrated until Mill added new sections
in which (5.1) is included. A question remains, however, whether Mill should have
introduced, in addition to the quantities demanded in each country of the imported
commodities, “the extent of the means of supplying that demand which are set
at liberty in each country by the change in the direction of its industry.” Already
Bastable (1990, p. 29)3 argued as follows.
“The attempt made by Mill to amend his theory by introducing the additional
element of the amount of capital set free for the production of exports is, as he
even admit, a failure; for, in the case of two countries and two commodities, the
amount of free capital, or, as I should prefer to say, “productive power,” is evidently
determined by reciprocal demands, so that nothing is gained by the laborious and
confusing discussion in Sects. 6, 7, 8 of Chap. xviii.”
As Chipman (1979) showed,
m = a 1 A1 , n = b 2 A2 (5.2)
where A1 (A2 ) is the maximum quantity of cloth Germany (England) can produce,
and a1 (b2 ) is the constant proportion in which expenditure is assumed to be devoted
to cloth (linen) in Germany (England). This is because the German national income
in terms of cloth is OA1 before trade, as is seen in Fig. 5.1, the English national
income in terms of cloth is OA2 before trade in Fig. 5.2, and b2 = 1 − a2 , where a2
is the constant proportion in which expenditure is assumed to be devoted to cloth in
England. Then, (5.1) can be written as
tb2 A2 = a1 B1 (5.3)
where B1 is the maximum quantity of linen Germany can produce, as is shown

in Fig. 5.1, since p = B1 /A1 . The right-hand side of (5.3) is the demand for cloth
of Germany and the left-hand side is the demand for linen of England, both in
3 C.F. Bastable (1855–1945). Professor of University of Dublin. See also Chap. 6 for the role he
played in the discussion of the infant industry protection.
5.1 Problems 37
terms of linen. The terms of trade t can be uniquely determined by the equation
of reciprocal demands (5.3) and there is no need of introducing “the quantity of
cloth which England can make with the labour and capital withdrawn from the
production of linen,” i.e., n. Even from the point of view of Chipman, therefore,
Mill’s superstructure is to be admitted as laborious and confusing.
What is more important for us is, however, to confirm that Mill considered
the model of an international trade, which we called the modern interpretation of
the Ricardian model in Chap. 4. Such a model is entirely different from the true
Ricardian model. This is because such entirely un-Ricardian assumptions are made
in the model that (1) there exists only one factor of production, labor, which requires
remuneration, (2) the total labor population of each country is given, and (3) the
productivity of labor remains unchanged irrespective of the level of production.
In other words, the existence of lands of different qualities which implies the
diminishing returns is assumed away so that each country is to be specialized
completely in the production of the exportables in a two-country two-good model
of international trade.
Although Mill used the expression “the labour and capital” to define his n in
the above (Mill 1909, p. 600), it is clear that by this labor and capital he meant a
single factor of production whose total quantity in each country and the productivity
are given and unchanged. This is because, firstly, he assumed the complete
specialization after trade. “Let us now suppose that England, previously to the
trade, required a million of yards of linen, which were worth, at the English cost
of production, a million yards of cloth. By turning all the labour and capital with
which that linen was produced to the production of cloth, she would produce for
exportation a million yards of cloth” (Mill 1909, p. 598). Secondly, as we already
explained, (5.1) presupposes that the national income of each country in terms of
her exportables remains unchanged between before and after trade.
Of course, Mill should not be blamed for this. These are simplifying assumptions
to derive (5.1). Any theorist is entitled to assume something to develop his theory.
Mill should be applauded because of these assumptions, by which he could
successfully solve (5.1) explicitly for the terms of trade. Besides, Mill did not
mention wrongly that it is the Ricardian model. We should conclude, then, that the
so-called Ricardian model of the modern interpretation of Ricardo should rather be
called Mill’s model of international trade.
5.1 Problems
5.1. Consider the maximization of utility function U = a log x + (1 − a) log y, being

subject to the budget constraint px + qy = Z, where a, x, y, p, q, and Z are,
respectively, positive constant less than 1, the quantity of good x, that of good y,
the given price of x, that of y, and the given income. Calculate the elasticity of x
with respect to p, q, and Z. See Problem 2.1.
5.2. In Fig. 5.2, England’s consumption of two goods is at the point E in autarky
(before trade). After trade, she is specialized in the production of cloth at the
point A2 . Draw the locus of the consumption points after trade, starting from E,
as the terms of trade is improved from OB2 /OA2 , assuming assumptions on the
demand elasticities which Mill made to derive (5.1) (See the explanation of offer
curves in Chap. 8 below).
5.2 Appendix: Mill and Thornton
The so-called Mill’s superstructure in Chap. 18 of his Principles is constructed so

as to avoid the non-uniqueness of the terms of trade, determined by the principle
of reciprocal demands. There seems to be, however, two different kinds of non-
uniqueness of the equilibrium terms of trade. The first case is that of multiple (or
even continuous) intersections of demand and supply curves, which Mill ruled out
by his assumptions on demand functions. The second case of non-uniqueness is
caused by the possible shifts of demand and supply curves so that the equilibrium
finally to be reached cannot be determined uniquely. Such shifts may be due to
exchanges at non-equilibrium prices.4
Mill stated that his superstructure is to reply to intelligent criticisms of his friend
William Thornton.5 As far as the publication is concerned, Thornton’s criticism of
the law of supply and demand started from Thornton (1866), which was reprinted
with revisions in Thornton (1869), though Chipman (1979) conjectured, perhaps
correctly, that Thornton had in earlier private conversation stimulated Mill to
reconsider his theory of international value.6 It is, then, very likely that Thornton’s
criticism was originally concerned with the second case of non-uniqueness but Mill
misunderstood it as that of the first case.
Thornton (1866) presented several counterexamples to the “equation theory”
that the equation of supply and demand determines price, among which the most
important is the example of the so-called Dutch auction for fish resorted by certain
fishermen, and its contrast with the usual English auction.
“When a herring or mackerel boat has discharged on the beach, at Hastings or
Dover, last night’s take of fish, the boatmen, in order to dispose of their cargo,
commonly resort to a process called “Dutch auction.” The fish are divided into lots,
each of which is set up at a higher price than the salesman expects to get for it, and
he then gradually lower his terms, until he comes to a price which some bystander
is willing to pay rather than not have the lot, and to which he accordingly agrees.
4 As for exchanges at non-equilibrium prices, see Hicks (1946, pp. 127–129). See also Hollander
(1985, pp. 276–277).
5 As for W. T. Thornton (1813–1880), see Ekelund (1997) and Picchio (1987).
6 Both Mill (since 1836) and Thornton (since 1847) were members of Political Economy Club.
Thornton (1866) seemed to be discussed in the Club, on 6, December, 1866. See Negishi (1988).
5.2 Appendix: Mill and Thornton 39
Suppose on one occasion the lot to have been a hundredweight, and the price agreed
to twenty shillings. If, on the same occasion, instead of the Dutch form of auction,
the ordinary English mode had been adopted, the result might have been different.
The operation would then have commenced by some bystander making a bid, which
others might have successfully exceeded, until a sum was arrived at beyond which
no one but the actual bidder could afford or was disposed to go. The sum would not
necessarily be twenty shillings: very possibly it might be only eighteen shillings.
The person who was prepared to pay the former price might very possibly be the
only person present prepared to pay even so much as the latter price; and if so,
he might get by English auction for eighteen shillings the fish for which at Dutch
auction he would have paid twenty. In the same market, with the same quantity
of fish for sale, and with customers in number and every other respect the same,
the same lot of fish might fetch two very different prices” (Thornton 1866, 1869,
pp. 47–48).
Mill thought that this is an example of the first case of the non-uniqueness of the
equilibrium, defined in the above, in the sense that there exists no unique intersection
of the given demand and supply curves. In his review of Thornton (1869), Mill
interpreted this example that “the demand and supply are equal at twenty shillings,
and equal also at eighteen shillings” (Mill 1967, p. 637) and that it is an exception
to the rule that demand increases with cheapness. Then Mill recanted the wages
fund doctrine7 in view of this particular case of indeterminacy due to demand that
is inelastic with respect to price. Supply being given constant, this is the case where
schedule of supply and demand are coincidental, at least within certain limits.
“When equation of demand and supply leaves the price in part indeterminate,
because there is more than one price which would fulfill the law (of the equation
of demand and supply)—the price, in this case, becomes simply a question whether
sellers or buyers hold out longest; and depends on their comparative patience, or
on the degree of inconvenience they are respectively put to by delay.—If it should
turn out that the price of labour falls within one of the excepted cases—the case
which the law of equality between demand and supply does not provide for, because
several prices all agree in satisfying that law; we are already able to see that the
question between one of those prices and another will be determined by causes
which operate strongly against the labourer, and in favour of the employer.—The
doctrine hitherto taught by all or most economists (including myself), which denied
it to be possible that trade combinations can raise wages, or which limited their
operation in that respect to the somewhat earlier attainment of a rise which the
competition of the market would have produced without them,—this doctrine is
deprived of its scientific foundation, and must be thrown aside” (Mill 1967, pp. 642–
43, 646).
7 Wages fund doctrine is the short-run wage theory of classical economics. The wage is determined
by the demand for labor (wages fund) and supply of labor (labor population). Given the size of
wages fund, therefore, the total wage income cannot be increased by trade (labor) unions.
Thus, Mill worried about the first case of the non-uniqueness, not only in his
superstructure of international value, but also in his discussion of wages fund
doctrine, on the basis of his interpretation of Thornton’s example of Dutch and
English auctions.
As was pointed out elsewhere for the case of wages fund doctrine, however, Mill
actually misunderstood the true implication of Thornton’s example of Dutch and
English auctions (Negishi 1986). Mill considered that it is an exceptional case to
the rule that demand increases with cheapness. Thornton then replied to Mill as
follows. “In this particular case it would not be possible for supply and demand to
be equal at two different prices. For the case is one in which demand would increase
with cheapness. A hawker who was ready to pay 8 s. for a hundred herrings would
want more than a hundred if he could get hundred for 6 s. There being then but a
given quantity in the market, if that quantity were just sufficient to satisfy all the
customers ready to buy at 8 s., it follows that it would not have sufficed to satisfy
them if the price had been 6. If supply and demand were equal at the former price,
they would be unequal at the latter” (Thornton 1870, pp. 57–58).
If the demand and supply are equalized at the price determined in Dutch auction,
then demand is larger than the supply at the lower price determined in English
auction. Since a single person is assumed to get all the supply, however, he will
not bid up the price further in English auction, since he knows that by so doing
he cannot satisfy his remaining demand. The lesson of this example of Thornton
(1866) is, therefore, that exchange is possible, and even inevitable, to take place at
such a price that demand is not equalized to supply there. Other examples given in
Thornton (1866), those of a glover and of a horse, can also be interpreted in the
same way.
Thornton (1866) continued his criticism against demand and supply equilibrium
theory. “Even if it were true that the price ultimately resulting from competition is
always one at which supply and demand are equalized, still only a small portion
of the goods offered for sale would actually be sold at any such price. Suppose
the glover to whom we have already once or twice referred, to have five hundred
pairs of gloves on hand, to begin by selling at three shillings a pair, and to be
tempted, by the rapid sale of two hundred pairs at that price to raise the price to
four shillings; suppose him to be subsequently tempted to raise it to five and six
shillings successively, but not to be able to sell at the last-named price, and therefore
to reduce it to five shillings, at which price the last hundred pairs are sold. The price
ultimately resulting from competition would then be five shillings, and this may, for
the sake of argument, be also assumed to be a price at which supply and demand
would be equalized. But at this price only one-fifth of the whole quantity would be
sold, the other four-fifth having been sold at price at which supply was in excess
of demand.—But when we speak of prices depending on certain causes, we surely
refer to prices at which all goods, or at least the great bulk of them, not that at which
merely a small remnant of them, will be sold. How can we say that the equation of
supply and demand determines price, if goods are almost always sold at prices at
which supply and demand are unequal?”
5.2 Appendix: Mill and Thornton 41
If Thornton (1866) is ever concerned with the non-uniqueness of equilibrium

price, it is now clear that it is the second case defined in the above which is due to
shifts of demand and supply curves caused by exchanges at non-equilibrium prices.
Mill misunderstood, however, Thornton’s criticism and consider the first case of the
non-uniqueness in his superstructure.8
If Mill’s superstructure aimed to reply to Thornton’s criticism of demand and
supply equilibrium theory, then, Mill should have dealt with the second case of the
non-uniqueness of the equilibrium. Mill’s assumption on demand made to avoid
the first case of the non-uniqueness is, fortunately, also helpful to deal with that of
the second case. In addition to Mill’s assumption, let us assume that two countries,
Germany and England, have the identical taste, so that the world demands for
both commodities, cloth and linen, are independent of the distribution of income
between countries. In other words, the changes in demand for any commodity of
any country caused by a redistribution of the world income is offset by those of the
other country completely.
In the model used in Chap. 5, Germany is specialized to the production of linen
B1 (see Fig. 5.1) and England, that of cloth A2 (see Fig. 5.2). The world income
is then B1 + tA2 , where t is the international price of cloth in terms of linen. The
condition for the demand and supply equilibrium for cloth in the world market is
a(B1 + tA2) = A2t (5.4)
where a is the constant proportion in which the expenditure is devoted to cloth

(identical for both countries). The equilibrium terms of trade t can be solved as
aB1
t= (5.5)
bA2
where b = 1 − a is the constant proportion in which the expenditure is devoted

to linen. It is independent of any redistribution of the world income, caused by
exchanges made at non-equilibrium prices.
Thus, Mill’s model in his superstructure can deal with the second case of the
non-uniqueness, i.e., Thornton’s criticism of demand and supply theory, if the
identical taste is assumed for two countries. This additional assumption seems to
be a neutral one. A unique rate of international value can be determined by the
principle of reciprocal demands equation, even if demand and supply curves of
individual commodities are shifted as a result of exchange transactions at other rates
of international value.
Still Thornton can wonder that the equilibrium theory is a truth of small
significance since it does not explain disequilibrium prices at which the bulk
of the goods offered for sale are actually sold. It is true that Mill can explain
8 In1871, however, Mill seemed to understand Thornton correctly. See Mill (1909, p. xxxi). See
also Negishi (1986).
the equilibrium terms of trade, but cannot explain the distribution of gains from
trade between countries, since the latter depends on how international trade at
disequilibrium terms of trade takes place. Mill the classical economist could not but
express his hope that only a small portion of goods may be sold at disequilibrium
(Mill 1967). After the marginal revolution, however, we can reply to Thornton that
the equilibrium theory is of great significance even if a small portion of goods are
sold at the equilibrium price which is finally established after the bulk of goods
are sold at disequilibrium. This is because the marginal rates of substitution (ratio
of marginal utilities between two commodities) are equalized among buyers and
sellers through such trades at equilibrium price so that a Pareto efficient allocation
of commodities is established.
5.3 Problems for Appendix
5.3. By using a demand curve, explain Thornton’s example of the different prices
in Dutch and English auctions (see Negishi 1986).
5.4. By using Edgeworth’s box diagram, explain the conclusion of this appendix.
Bibliography
Bastable, C. F. (1900). Theory of international trade. London: Macmillan.

Chipman, J. S. (1965). A survey of the theory of international trade: part 1: The classical theory.
Econometrica, 33, 477–519.
Chipman, J. S. (1979). Mill’s “superstructure”: how well does it stand up? History of Political
Economy, 11, 477–500.
Edgeworth, F. Y. (1894). The theory of international values, III. Economic Journal, 4, 424–443.
Ekelund, R. B. (1997). W. T. Thornton: savant, idiot, or idiot-savant? Journal of the History of
Hicks, J. R. (1946). Value and capital. Oxford: Oxford University Press.
Hollander, S. (1985). The economics of John Stuart Mill. Oxford: Blackwell.
Mill, J. S. (1909)[1848]. Principles of political economy. London: Longmans, Green and Co.
Mill, J. S. (1967)[1869]. Thornton on labour and its claims, in idem. Essays on economics and
society (pp. 631–68). Toronto: University of Toronto Press.
Negishi, T. (1986). Thornton’s criticism of equilibrium theory and Mill. History of Political
Economy, 18, 567–577.
Negishi, T. (1998). Mill’s superstructure, how it should have been. Aoyama Journal of International
Politics, Economics and Business, 42, 27–39.
Picchio, A. (1987). Thornton, William Thomas, 1813–1880. In J. Eatwell, M. Milgate, & P.
Newman (Eds.), The new palgrave, vol. 4 (p. 636). London: Macmillan.
Thornton, W. T. (1866). A new theory of supply and demand. Fortnightly Review, 6, 420–434.
Thornton, W. T. (1869). On labour: its wrongful claims and rightful dues, its actual present and
possible future. London: Macmillan.
possible future (2nd ed.). London: Macmillan.
Chapter 6
Mill and Infant Industry
Starting with Adam Smith’s criticism against mercantilism, economists of the

classical school generally advocated the free trade and were critical to the protection
of domestic industries. J. S. Mill admitted, however, the protection of the so-called
infant industry, though he imposes a condition which an industry must satisfy to
be protected. Then, it was Bastable who followed Mill to add another necessary
condition for protection. This Mill–Bastable infant industry dogma was discussed
critically by some modern economists from the point of view of the dynamic theory
of the gains from trade.
Let us first summarize the theory of consumers’ and producers’ surplus, which is
very convenient to discuss the problem of gains from trade. In Fig. 6.1, the quantity
of a commodity is measured horizontally, and its price and cost of production,
vertically. The curve DD∗ is the demand curve of domestic consumers, and the
curve SS∗ is the supply curve of the domestic industry. The equilibrium in autarky
is the point A where OC of the commodity is produced and sold at the price of
OB. The consumers’ surplus is indicated by the triangle DAB, while the producers’
surplus, by the triangle SAB. The reason why DAB is the consumers’ surplus is as
follows. The consumer who buys the first unit is willing to pay as much as OD,
but actually pays only OB, so he gains the surplus BD. The consumer who buys
the second unit gains similarly, though his gain is slightly smaller than BD. The
consumer who buys the last, the OC’th unit, gains no surplus, since he is willing to
pay as much as AC = BO and actually pays BO. The total surplus consumers obtain
is, therefore, the triangle DAB, since they are willing to pay as much as DACO, but
actually pay BACO only.1 The producers’ surplus SAB is, then, explained as the
profit of the domestic industry, i.e., difference between the revenue BACO and the
cost of production OSAC, since the supply curve SS∗ of a competitive industry is
the marginal cost curve.
1 Itis to be noted that, to measure the utility in terms of money, we are here implicitly assuming the
constancy of the marginal utility of money. In other words, the demand curve is considered as if it
is the marginal utility curve. See Problems 6.1 and 6.2.

44 6 Mill and Infant Industry
Fig. 6.1 Consumers’ and

producers’ surplus
Suppose the price of this commodity in international market is given as OS in

Fig. 6.1. In other words, the supply curve of foreign exporters is the straight line
SF. In the domestic market, after trade, consumers buy SF of this commodity at
the price OS and gain the consumers’ surplus DFS, which is increased from the
before trade one DAB by the amount SBAF. The before trade producers’ surplus
BAS disappears, however, since the domestic industry cannot supply at all at the
price of OS. Even then, the combined surplus of consumers’ surplus and producers’
surplus is increased from DAS before trade to DFS after trade by ASF. This shows
the gains from the trade, as far as this single commodity is concerned. Now any
attempt to protect this domestic industry can be shown to decrease these gains from
trade. For example, suppose the domestic industry is protected so as to be able
to supply OC, with the import of the commodity reduced to EF. Now, domestic
industry’s negative surplus (i.e., loss) ASE must be deducted from the gains from
free trade ASF, so that the gains from trade become only AEF. If the domestic
industry is protected by a prohibitive duty (tariff) BS on each unit of import, the
price in the domestic market returns to OB and the domestic industry enjoy the
producers’ surplus BAS again, but the gains from trade disappear.
Why is the domestic industry so inefficient that it cannot compete with the
foreign suppliers? It may simply be that this country happens to have no comparative
advantage in the production of this commodity. Then there is no problem at all of
the protection of the domestic industry. Even if this country has the comparative
advantage, however, the domestic industry may not be able to compete with foreign
matured industry simply because the domestic one is young and premature. Then,
there may be a case for the protection of the industry. J. S. Mill clearly admitted this
case and carefully argued as follows.
6 Mill and Infant Industry 45
“The only case in which, on mere principles of political economy, protecting

duties can be defensible, is when they are imposed temporarily (especially in
a young and rising nation) in hopes of naturalizing a foreign industry, in itself
perfectly suitable to the circumstances of the country. The superiority of one country
over another in a branch of production often arises only from having begun it sooner.
There may be no inherent advantage on one part, or disadvantage on the other, but
only a present superiority of acquired skill and experience. A country which has
this skill and experience yet to acquire, may in other respects be better adapted to
the production than those which were earlier in the field: and besides, it is a just
remark of Mr. Rae, that nothing has a greater tendency to promote improvements
in any branch of production than its trial under a new set of conditions. But it
cannot be expected that individuals should, at their own risk, or rather to their certain
loss, introduce a new manufacture, and bear the burden of carrying it on until the
producers have been educated up to the level of those with whom the processes
are traditional. A protecting duty, continued for a reasonable time might sometimes
be the least inconvenient mode in which the nation can tax itself for the support
of such an experiment. But it is essential that the protection should be confined to
cases in which there is good ground of assurance that the industry which it fosters
will after a time be able to dispense with it; nor should the domestic producers ever
be allowed to expect that it will be continued to them beyond the time necessary for
a fair trial of what they are capable of accomplishing” (Mill 1909, p. 922).
Thus Mill insisted that not all the domestic industries should be protected and that
the period of the protection should be limited. In other words, the cost of protection,
i.e., the loss of the gains from free trade, should be bounded and finite. Even if it is
not infinite, however, the country cannot bear the cost, unless she can expect some
benefit from it. The cost benefit analysis of the protection is necessary. This is the
point insisted by C. F. Bastable who followed J. S. Mill to consider the required
conditions for the possible employment of the policy of protection of domestic
industries. “The onus of proof rests with those who advocate their employment,
and they are bound to show (1) that the industry to be favoured will after a time be
self-supporting, and (2) that the ultimate advantage will exceed the losses incurred
during the process. A careful computation of the different elements involved—the
loss in each year of protection, with interest on the losses during earlier years, the
estimated amount of gain to accrue when the time for independence is reached—”
(Bastable 1892, p. 135).
It is M. C. Kemp who nicely sums up the classical Mill–Bastable dogma as
the proposition that if an industry passes both the Mill’s test and the Bastable’s
test it should be protected until it can stand on its own feet.2 Kemp distinguishes,
firstly, the learning process of infant industries into two cases, the dynamic internal
economies and the dynamic external economies. In the former, only the firm that
actually carries on production currently gains in experience and in the future the
accumulated knowledge becomes the exclusive property of that firm. In the latter,
2 See Kemp (1960, 1964, pp. 184–191). See, however, also Mundell (1957).
however, the knowledge accumulated is nonappropriable and can be tapped by any

newcomer in the future. “If the learning process is internal to the firm it is difficult
to sustain the conclusion of Mill and Bastable. For suppose that the Mill–Bastable
test is passed. Then while the unprotected firm may sustain losses during part of
its learning period, it will later enjoy profits which, by assumption, will more than
compensate it for the losses.—If, on the other hand, the process is external to the
firm, one is led to very different conclusions. In this case the pioneer must share his
harvest of knowledge with newcomers; hence, the private incentive to undertake
investments which pass the Mill–Bastable test may be insufficient, and a temporary
tariff or other means of protection may be justified” (Kemp 1964, p. 187).
In addition to Mill’s test and Bastable’s test, now we have the third test for the
protection of the infant industry, Kemp’s test, that there must be dynamic external
economy. The reason that Bastable’s test is not sufficient is, according to Kemp, that
it implies that profitability of the private investment. We must note, however, that
the future gains in Bastable’s test is here interpreted as the profit of the industry, i.e.,
the producers’ surplus only. If, in addition to the producers’ surplus, the changes
in the consumers’ surplus in the future are taken into consideration, there is a
possibility that Bastable’s test does not necessarily imply the profitability of the
project as the private investment. It is possible that the test is passed since the present
value of the sum of the producers’ surplus (profit) and the increase in consumers’
surplus in the future is larger than the present cost, but it is not profitable as the
private investment since the present value of the profit in the future is smaller than
the present cost.3
In Fig. 6.2, the demand and supply of a commodity in the domestic market are
measured horizontally to the right from the origin O, the supply from the foreign
exporters, to the left from the origin, and the price and cost, vertically. The supply
curve of the domestic industry ss∗ , the domestic demand curve DD∗ and the supply
curve of foreign exporters EE∗ are given for the present period. When the domestic
industry does not exist, at the price Od the domestic demand is bd which is entirely
supplied from the import ed (bd = ed). After the domestic industry is created, the
price falls to Oh and the demand is hf, of which hg is supplied from the domestic
industry, fg is imported from the foreign country (hi = fg). Since ss∗ is the marginal
cost curve and no fixed cost is assumed to exist, the loss of the domestic industry is
sagh. The increase of consumers’ surplus is bdhf, which is larger than the decrease
of the foreign surplus deih. Therefore, the present social cost of the domestic
industry is less than sagh.
In the future, the domestic supply curve is shifted downward, as is shown in
Fig. 6.3. This is made possible by the learning, acquisition of experiences and
knowledge from the production in the present. If the domestic industry had not been
developed (i.e., s were located above b as s is located above d in Fig. 6.2), ab would
have been demanded and bc would have been imported (ab = bc) at the price of Ob.
After the domestic industry grows up, df is demanded at the price of Of, fe being
3 See Negishi (1968), and also Corden (1974, pp. 257–8).

6 Mill and Infant Industry 47
Fig. 6.2 Domestic industry at present
supplied from the domestic industry and ed (= fg) being imported. The increase of
consumers’ surplus is abfd, which is larger than the decrease of the foreign surplus
bcgf. Domestic industry’s two period profit is sef in Fig. 6.3 minus sagh in Fig. 6.2:
sagh is something like a fixed cost for the future production of the domestic industry.
It should be noted that to pull down s in Fig. 6.3 below b, the level of production of
the domestic industry in the present should be non-negligible (or substantial), say,
Om, and sagh is also non-negligible, since s is assumed to be located non-negligibly
above d in Fig. 6.2.
If abfd minus bcgf (in Fig. 6.3) is greater than sagh (in Fig. 6.2) minus sef (in
Fig. 6.3), the domestic industry should be fostered, not only from the nationalistic
but also from the world point of view. However, if sef is smaller than sagh, there is
no private incentive for this. The subsidy is needed to compensate for the difference
between sagh and sef. In other words, the reason for the protection is the lumpiness
of the present cost of the infant industry which generates the dynamic (over period)
internal economy (increasing returns to scale or diminishing cost to scale due to the
existence of a fixed cost).
Fig. 6.3 Domestic industry in the future
6.1 Problems
6.1. Consider the maximization of utility being subject to the budget constraint
where the price of goods and the income of the consumer are given constants. Derive
the inverse demand function of a good and consider the condition which is necessary
to consider it the marginal utility function.
6.2. Discuss the consumers’ surplus from the point of view of the ordinal (not
cardinal) utility. Draw indifference map between money and a good and discuss
the condition necessary to define the consumers’ surplus uniquely (See Hicks 1946,
pp. 38–41).
6.3. While cost of the protection is finite in view of Mill’s test, the benefit from the
industry in the future can be expected indefinitely. Does Bastable’s test make sense,
since the total benefit from the industry is infinitely large? (Consider the discounted
present value.)
6.4. What can be considered as examples of the dynamic external economy
discussed by Kemp?
Bibliography
Bastable, C. F. (1892). The commerce of nations. London; Methuen.

Corden, W. M. (1974). Trade policy and economic welfare. London: Oxford University Press.
Hicks, J. R. (1946). Value and capital. London: Oxford University Press.
Bibliography 49
Kemp, M. C. (1960). The Mill–Bastable infant industry dogma. Journal of Political Economy,
LXVIII, 65–67.
Mill, J. S. (1909). Principles of political economy. London: Longmans, Green and Co.
Mundell, R. A. (1957). International trade and factor mobility. American Economic Review, 47,
321–335.
Negishi, T. (1968). Protection of infant industry and dynamic internal economies. Economic
Record, 44, 56–67.
Chapter 7
Marx and International Exploitation
In his Capital (1867–94) and Theories of Surplus Value (1905–10), K. Marx insisted
two different types of exploitation. The first one is, of course, the exploitation of
labor by capital in the case of equal labor quantity exchange, while the second
is the exploitation of poor countries by rich ones through unequal labor quantity
exchanges. The exploitation theory of interest, i.e., the exploitation of labor by
capital, was criticized by Boehm-Bawerk from the point of view of the compar-
ison of values differently dated.1 Marx’s own consideration of the international
exploitation was fragmentary, but it was succeeded and developed by modern
Marxian economists. The same criticism seems to be applicable to this Marxian
theory of international trade, however, as the one used by Boehm-Bawerk against
the exploitation theory of interest (profit).
The exploitation theory of interest is based on the fundamental assumption of
the capital theory of the classical economics to which Marx identified himself as the
sole orthodox successor. The assumption is that the variable capital (the wage goods
like food and necessaries) is advanced by capitalists to laborers, which is quite in
contrast to modern neo-classical assumption that wage is paid out of current, not
past, output. Capital must be advanced because there is a time lag between input
of labor and output of commodities and laborers cannot waste output since they are
stripped of any means of subsistence.
To see the essence of the exploitation theory of interest, let us consider a simple
case of an economy composed of labor power2 and wheat. Homogeneous land is
assumed to exist infinitely, and the existence of constant capital (materials, tools,
machines, etc.) is ignored, as was done by Marx himself (1954, pp. 206–7), so
as to make the story as simple as possible. The only capital to be advanced is
the variable capital, to be paid in exchange for the labor power to laborers who
1 E.V. Boehm-Bawerk (1851–1914) was a follower of C. Menger and the professor at Vienna
University. He contributed greatly to the theory of capital.
2 Marx distinguished labor power (capacity for labor) from labor. The former exercises the latter to
produce value (Marx 1954, p. 164).

52 7 Marx and International Exploitation
are stripped of any means of subsistence, which takes the form of wheat, the sole
product of the economy. Exchanges are carried out according to the embodied labor
values. Commodities for which the same amount of labor is necessary to produce are
exchanged each other. Variable capital to be advanced is, then, given by the amount
of wheat necessary to reproduce, in the household of laborers, the labor power to be
used up in the production of wheat. The period of production is naturally 1 year and
the harvest of new wheat is 1 year later than the first payment of wheat wage from
the capitalists’ stock of wheat accumulated from past harvests.
Now because of the peculiar property of labor power to be the source of the
surplus value, the amount of output of new wheat is larger than the amount of wheat
advanced as the variable capital. The difference is called surplus product and the
ratio of the surplus product to the variable capital is defined as the rate of surplus
value or the rate of exploitation. Since exchanges in the labor market are made
according to embodied labor values, this surplus and exploitation occur under the
equal labor quantity exchange. The whole product is a labor product, but only a
part of its embodied labor value corresponding to the advanced variable capital is
paid, while the rest corresponding to surplus product is unpaid and exploited by the
capitalists.
If the new wheat harvested at the end of the production period and the old one
advanced by capitalists to laborers are identical, not only physically, but also socially
and economically, the exploitation theory makes a sense and there is no objection
to it. The new wheat and old wheat are, however, dated differently and there is
no assurance that they are identical in their relations to capitalists and laborers.
When the old wheat is advanced, for example, it might be very scarce while the
new wheat is not yet available until it is redundant. Unless they are identical, there
is no guarantee to be able to compare the physical amount of new and old wheat
and to talk about the surplus and exploitation. It is of no use in this respect merely
to translate physical amount of wheat into expression in terms of labor. Since there
can be no labor movement between different time points, embodied labor cannot be
used to compare the value of commodities dated differently.
Boehm-Bawerk (1959, pp. 263–5) attacked the exploitation theory of interest
on the ground that the future and present goods are wrongly considered identical.
He considered the following example. Suppose a single worker spend 5 years to
complete independently a steam engine from the beginning, which commands, when
completed, a price of $5,500. There is no objection to give him the whole steam
engine or $5,500 as the wage for 5 years’ continuous labor. But when? Obviously
it must be at the expiration of 5 years. It is impossible for him to have the steam
engine before it is in existence. He cannot receive the steam engine valued at $5,500
and created by him alone, before he has created it. His compensation is the whole
future value at a future time.
But the worker having no means of subsistence cannot and will not wait until his
product has been fully completed. Suppose our worker wishes, after the expiration
of the first year, to receive a corresponding partial compensation. The worker should
get all that he has labored to produce up to this point, say, a pile of unfinished ore,
or of iron, or of steel material, or the full exchange value which this pile of material
7 Marx and International Exploitation 53
has now. The question is how large will that value be in relation to the price of
the finished engine, $5,500. Can it be $1,100, since the worker has up to this time
performed one-fifth of the work?
Boehm-Bawerk said “No.” One thousand one hundred dollars is one-fifth of the
price of a completed, present steam engine, which is different from what the worker
has produced in the first year, i.e., one-fifth of an engine which will not be finished
for another 4 years. The former fifth has a value different from that of the latter
fifth, in so far as a complete present machine has a different value from that of an
engine that will not be available for another 4 years. Our worker at the end of a
year’s work on the steam engine that will be finished in another 4 years has not yet
earned the entire value of one-fifth of a completed engine, but something smaller
than it. Assuming a prevailing interest rate of 5%, Boehm-Bawerk concluded that
our worker should get the product of the first year’s labor which is worth about
$1,000 at the end of the first year. Thus, Boehm-Bawerk criticized those who insist,
ignoring wrongly the difference between present and future goods, that there is
exploitation unless workers do receive the entire future (not the present) value of
his product now (not in the future) though it is available only in the future.
Let us now turn to Marx’s theory of international exploitation. Marx insists that
the richer country exploits the poorer one through international trade. Though we
can see this both in Capital and in Theories of Surplus Value, let us quote from the
latter. “And even according to Ricardo’s theory, three days of labour of one country
can be exchanged against one of another country—. Here the law of value undergoes
essential modification. The relationship between labour days of different countries
may be similar to that existing between skilled, complex labour and unskilled,
simple labour within a country. In this case, the richer country exploits the poorer
one, even where the latter gains by exchange, as John Stuart Mill explains in his
Some Unsettled Questions” (Marx 1971, pp. 105–6).3
Construct a simple model to show that the richer country with high labor
productivity can exchange a smaller quantity of its labor for a larger quantity of
the poorer country which has low labor productivity (Negishi 1999). Consider a
two-country, two-commodity model. Let us assume that the real subsistence wage
consists of one unit of the second commodity.4 Then, the price–cost relations in the
i-th country (i = 1, 2) in autarky (before trade) are
pi = (1 + ri )ai1 , (7.1)
1 = (1 + ri )ai2 (7.2)
where pi is the price of the first commodity in terms of the second one, ri is the
rate of profit, and ai j ( j = 1, 2) are the quantity of the input of labor necessary to
produce a unit of the j-th commodity.
3 See Essay 1 in Mill (1844), which was later developed into his Principles.
4 In other words, the value of labor power is assumed to be equal to that of the second commodity.
Without loss of generality, consider that the i-th country has comparative
advantage in the i-th commodity, so that
a11 a21
p1 = < = p2 . (7.3)
a12 a22
Furthermore, let us assume that the second country has absolute advantage in the
production of the first commodity, although the first country has a comparative
advantage in it, so that a11 > a21 and therefore a12 > a22 from (7.3). In other words,
the second country is the richer country, with a higher labor productivity, and the
first is the poorer one, with a lower labor productivity.
After trade, the i-th country specializes in the production of the i-th commodity,
i = 1, 2. Then, the price cost relations are
p = (1 + r1)a11 , (7.4)
1 = (1 + r2)a22 (7.5)
where p denotes the international price of the first commodity in terms of the second.
Since a11 /a12 < p < a21 /a22 and a11 > a21 ,
a11 > pa22 . (7.6)
Since a unit of the first commodity, which contains a11 units of the labor of the first
country, is exchanged with p units of the second commodity, which contains a22
units of the labor of the second country, (7.6) implies that a larger quantity of the
labor of the first country (the poorer country) is exchanged in international trade for
a smaller quantity of labor of the second country (the richer country). This unequal
exchange is due to the difference in the rate of profit, caused by the immobility of
capital between countries. Since we have, from (7.4) and (7.5),
(1 + r1 )a11
p= (7.7)
(1 + r2 )a22
then, from (7.6) and (7.7),
r2 > r1 . (7.8)
Thus, the rate of profit is higher in the richer country, where the productivity of labor
is higher, than in the poorer country, where the labor productivity is lower.
It is shown that the richer country of higher labor productivity can exchange a
smaller quantity of its labor for a larger quantity of the poorer country of a low labor
productivity. Even though labor is not mobile internationally, we can compare labors
located at different countries, since they are reproduced by the same consumption
of one unit of the second commodity which is freely traded internationally. In view
of the difference in the rate of profit, however, this unequal exchange of labor
quantities does not imply an unequal exchange of value whereby the richer country
is exploiting the poorer one.
In the above we followed Boehm-Bawerk and denied Marx’s theory of the

exploitation of labor by capital which states that the difference between the value
of the labor product and the value of labor power (i.e., wage) is exploited by
the capitalists. To compare the value of the product available in the future and the
value of labor power as paid in the present, we have to discount the former by the
rate of interest. Then, since the discounted present value of the product is equal to
the value of the wage, there is no exploitation of labor by capital. Similarly, the
value that was expended in the past should be augmented by the rate of interest to
be compared in the present. Commodities exchanged in international trade embody
labor quantities expended in the past, unless we assume that the production is
instantaneous. The value of such labor (not labor power) embodied, i.e., its quantity,
should be augmented by the rate of interest in the value comparison of commodities
exchanged internationally.
In our simplified model considered above, the role of interest should be played
by the rate of profit in the intertemporal comparison of values. A unit of the first
commodity, which contain a11 units of labor of the first country, is exchanged for
p units of the second commodity, which contains a22 units of labor of the second
country. Since the rate of profit is r1 in the first country and r2 in the second country,
then it is, not a11 but (1 + r1 )a11 that should be compared not with pa22 , but with
(1 + r2 )pa22 . In view of (7.7), then, the value of a unit of the first commodity (a11
units of labor of the first country) is identical to the value of p units of the second
commodity (a22 units of labor of the second country). The international exchange
of a unit of the first commodity for p units of the second one is an exchange of equal
values, and there exist no exploitation of the first country by the second one.
Provided that we can use the rate of profit as the discounting factor in comparing
values differently dated, there is no international unequal exchange of values, and
therefore no international exploitation between countries, even though there is
international unequal exchange of labor quantities. Thus, we have to deny Marx’s
theory of international exploitation, which is based on an international unequal
exchange of labor quantities.
In the history of the theory of international exploitation, Marx was followed
by Bauer who discussed international exploitation arising from differences in the
organic composition of capital.5
“The capital of the more highly developed country has the higher organic
composition of capital—Now Marx has made it possible for us to understand that,
thanks to the tendency to an equalization of profit rates, the workers of each country
do not produce value only for their own capitalists; rather the surplus value produced
by the workers of both countries is divided between the capitalists of both countries,
5 The organic composition is defined as the ratio of the constant capital to the variable capital.
Since Marx considered that only the variable capital can produce the surplus value, the higher
organic composition implies the lower rate of profit, if commodities are exchanged according to
embodied labor values. The competition among capitalists requires the exchanges according to
prices of production, so that the rate of profit is equalized. The price of production is higher than
the value, if the organic composition is higher.
not according to the quantity of labor performed in each of the two but according
to the quantity of capital active in each of the two countries. Since, however, in
the more highly developed country more capital goes with the same quantity of
labor, therefore the more highly developed country attracts to itself a larger share
of surplus value than corresponds to the quantity of labor performed in it. It is as
though the surplus value produced in both countries were first heaped up in a single
pile and then divided among the capitalists according to the size of their capitals. The
capitalists of the more developed country thus exploit not only their own workers,
but continually appropriate also a portion of the surplus value produced in the less
developed country (Bauer 1907, pp. 246–7, translated by Sweezy 1946, pp. 290–1).
Sweezy (1946, pp. 291–2) was critical of Bauer on two points. Firstly, he
emphasized that the assumption of the perfect international mobility of capital is
necessary in order to apply Marx’s theory of organic composition of capital not
only to the distribution of surplus value between domestic industries but also to
its distribution between two different countries. Secondly, Sweezy pointed out that
the perfect international mobility of labor must be assumed in order to apply the
result of an analysis of a closed economy to the world economy. For Sweezy, these
assumptions were far from realistic.
As for the second assumption, we can dispense with it, since we are assuming
that everywhere the subsistence wage consists of a given amount of the same wage
goods which are internationally freely traded. We may, however, accept the first
assumption, since realistically the perfect mobility of capital between countries
might now be better than the complete immobility of capital, as is suggested by
Emmanuel, a representative modern Marxian economist of international trade.
“The capital factor mobile, but the labor factor immobile—It is this—case that
seems to me to fit present-day reality the best, and for this reason it will furnish
the basic condition of the following thesis—Sufficient mobility of capital to ensure
that in essentials international equalization of profit takes place, so the proposition
regarding prices of production remains valid—(Emmanuel 1972, pp. xxxiii–xxxiv).
Let us consider a simple two-country two-commodity model. The first commod-
ity is assumed to be produced by the use of the first commodity itself and labor,
while the second commodity is produced by the use of labor only. The subsistence
wage is assumed to be equal to a unit of the second commodity. The organic
composition of capital is defined as the ratio of the value of non-labor input to the
value of the labor input and is higher in the production of the first commodity than
in the production of the second, since it is zero for the latter.
The price–cost relation is
p = (a11 p + a12)(1 + r) (7.9)
and
1 = a22 (1 + r) (7.10)
where p is the price of the first commodity in terms of the second one, r is the rate of
profit, a11 is the quantity of input of the first commodity necessary to produce a unit
of the first commodity (a constant), and a12 and a22 are, respectively, the quantity
of input of labor necessary to produce a unit of the first and second commodities
(constants). Equations (7.9) and (7.10) are solved as
a12
p= (7.11)
a22 − a11
and
1 − a22
r= . (7.12)
a22
To assure p > 0 and r > 0, we must assume that a22 > a11 and 1 > a22 . Since input
coefficients are different between country A and country B, the values of p and r
are different between two countries if there is no international trade and investment
(i.e., capital movement).
Suppose p in country A is lower than p in country B before international trade
and investment. After trade and investment, then, country A is specializing in the
production of the first commodity, country B is specializing in that of the second
commodity, and r is equalized between the two countries. The price–cost relation is
still (7.9) and (7.10), and p and r are given again by (7.11) and (7.12). However, a11
and a12 are now input coefficients in country A while a22 is that of country B.
The quantity of labor embodied in a unit of the second commodity is a22 , by
definition. Then, what about the quantity of the labor embodied in a unit of the first
commodity? The standard way of calculation in Marxian economics is as follows
(e.g., Morishima 1973, p. 11). If the quantity in question is x, then,
x = a11 x + a12 (7.13)
must be satisfied, since the embodied labor value of the non-labor input (constant
capital) is simply transferred to the value of the output. Then, the quantity of the
labor embodied in a unit of the first commodity is a12 /(1 − a11).
A unit of the first commodity produced in country A is exchanged for p units
of the second commodity produced in country B. If labor units embodied are
compared,
a12 a22 a12
< (7.14)
1 − a11 a22 − a11
in view of (7.11) and the assumption that a22 < 1. This implies that the country
specializes in a commodity with a higher organic composition of capital receives
more labor than it offers in international trade, as was insisted by Bauer.
But who is exploited by whom in this case? There is no change in real wages after
international trade and investment takes place. Nor is any change in the rate of profit
in the alleged exploited country B, since it is specializing in the production of the
second commodity, i.e., the wage good. Something must be wrong in this account
of exploitation. In view of (7.14), there is no exploitation if there is no non-labor
input, i.e., if a11 = 0. If there is something wrong in our calculation of the quantity
of embodied labor, then it must be related to the calculation of embodied labor of
the non-labor inputs. In (7.13) this is simply added to the quantity of labor input.
Unless production is instantaneous, however, labor embodied in non-labor inputs
is the labor in some past periods. Can we simply add the labor in the past period
to the labor in the present one? For comparison of the quantities of labor, i.e., the
embodied labor values, in the different periods, we have to follow Boehm-Bawerk
again in using the rate of profit.
The quantity of labor in the (t − 1)th period must be multiplied by (1 + r) to be
compared with the quantity of labor in the t-th period. If we replace (7.13) by
x = a11 x(1 + r) + a12, (7.15)
then
a22 a12
x= (7.16)
a22 − a11
since (1 + r) = 1/a22 in view of (7.12). The corrected quantity of labor embodied in

a unit of the first commodity, i.e., x, is now equal to the quantity of labor embodied
in p units of the second commodity, since a unit of the second commodity contains
the quantity of labor a22 and p = a12 /(a22 − a11 ). In terms of our corrected quantity
of embodied labor, there is no unequal international exchange of labor. In other
words, there exists no international exploitation, contrary to what Bauer insisted
upon.
7.1 Problems
Even though the Marxian theory of exploitation does not make sense, we cannot
deny the possibility of other explanations of the international exploitation (see
Negishi 1999).
7.1. Assume the perfect mobility of capital and the complete immobility of labor
internationally. How can we explain the international wage differentials? What
behavior of labor supply is necessary to produce the subsistence wage? What is
necessary to make the level of wage higher than the subsistence level?
7.2. Modify (7.1) and (7.2) so that the rate of profit is equalized internationally,
the rate of wage is at the subsistence level in one country and at the level higher
than it in another country. Derive the necessary condition on p (i.e., conditions on
reciprocal demands) for unequal exchange of labor. Who is exploiting whom in this
case?
Bibliography 59
Bibliography
Bauer, O. (1907). Die nationalitaetenfrage und die sozialdemokratie. Vienna: Wiener Volksbuch-
handlung.
Boehm-Bawerk, E. V. (1959). Capital and interest, the history and critique of interest theories
(G. D. Huncke, & H. F. Sennholz, Trans.). South Holland, Ill: Libertian Press.
Emmanuel, A. (1972). Unequal exchange (B. Pearce, Trans.). New York: Monthly Review Press.
Marx, K. (1954). Capital, vol. 1. Moscow: Progress Publishers.
Marx, K. (1971). Theories of surplus-value, vol. 3. Moscow: Progress Publishers.
Mill, J. S. (1844). Essays on some unsettled questions of political economy. London: Longmans,
Green, Reader and Dyer.
Morishima, M. (1973). Marx’s economics. Cambridge: Cambridge University Press.
Negishi, T. (1999). Unequal exchange and exploitation. Japanese Economic Review, 50, 113–121.
Sweezy, P. M. (1946). The theory of capitalist development. London: Dobson.
Chapter 8
Marshall and Offer Curve
The marginal revolution against the classical economics is often attributed to

W. S. Jevons, C. Menger, and L. Walras, who published their works in 1870s.
From the point of view of international trade theory, however, we may also add the
name of A. Marshall,1 though his Principles of Economics was published in 1890.
As a matter of fact, his early tracts on The pure theory of foreign trade and The
pure theory of domestic value are both dated 1879. Marshall reworked some of
the theories of J. S. Mill into rigorous diagrams, which includes, among others,
those of Marshallian offer curves to study the equilibrium terms of trade in a
two-commodity two-country model. Marshall introduced the graphic apparatus of
offer curves, though he did not show how they are derived from the underlying
demand and production. It was left for later day’s economists, for example, Meade
(1952), who skillfully derived offer curves by the use of trade indifference curves.
Marshall interpreted, in his tract on The pure theory of foreign trade, the
law of international demand into the language of diagrams. What he considered
is, essentially, our old familiar two-commodity two-country model. In Fig. 8.1,
the quantity of cloth x is measured horizontally, and that of linen y, vertically.
The curve OE is called England’s demand curve. England is exporting cloth and
importing linen. Any point on the curve OE indicates “the number of yards of
cloth, the expenses of producing and exporting which will be covered annually by
the proceeds of the sale in England of an amount of linen.” In other words, any
point on the curve shows “the amount of cloth which England will be willing to
give annually for an amount of linen” chosen. Similarly, the curve OG is called
Germany’s demand curve. The intersection of these two curves A gives, then,
the equilibrium of the international trade, the quantity of cloth and that of linen
1 A.Marshall (1842–1924) was professor of economics at the University of Cambridge. Against

Walras’s general equilibrium theory of a closed economy, generally he tried to develop partial
equilibrium theory of single industries and forged many analytical instruments still useful in
applied economics.

62 8 Marshall and Offer Curve
Fig. 8.1 Offer curves
exchanged between England and Germany. In other words, the slope of the line OA
indicates the equilibrium terms of trade for England, the ratio of the price of the
exportables to that of the importables (Marshall 1879, pp. 135–137).
Edgeworth2 reproduced curves which Marshall called demand curves but named
them supply-and-demand curves of international trade and pointed out that these
curves are quite different from the ordinary demand or supply curves used in
the partial equilibrium analysis or the analysis of a single industry. These curves
are derived from the general equilibrium analysis or the simultaneous analysis of
two industries in each country. “There is more than meets the eye in Professor
Marshall’s foreign trade curves” (Edgeworth 1925, v. 3, p. 143). “[A] movement
along a supply-and-demand curve of international trade should be considered
as attended with rearrangement of internal trade; as the movement of the hand
of a clock corresponds to considerable unseen movements of the machinery”
(Edgeworth 1925, v. 2, p. 32).
Marshall’s curves in Fig. 8.1 are now called offer curves.3 To derive offer curves
of a country, which relates her export and import, we have to consider how the
consumption and production of commodities are determined in the domestic market,
since the export (import) of a commodity is the excess of the supply of the domestic
producers over the demand from the domestic consumers (the demand from the
domestic consumers over the supply of the domestic producers).
A consumer determines the demand for commodities so as to maximize the utility
of consumption, being subject to the budget constraint, if prices of commodities
and the disposable income are given. This requires that the indifference curve
corresponding to the maximized utility level is tangent to the budget line whose
slope is given by the price ratio. If the income is distributed equally among all
2 Edgeworth (1845–1926) was professor of economics at Oxford and founded, with Walras, the
basis of modern general equilibrium theory.
3 See Newman (1965, p. 104), e.g., for who first called the curves offer curves.
8 Marshall and Offer Curve 63
Fig. 8.2 Indifference and transformation curves
the consumers and also they have identical taste, we can aggregate consumers’
behavior so that the aggregate demand can be derived from the maximization of
the aggregate utility, being subject to the aggregate budget constraint. Figure 8.2
shows an indifference curve corresponding to the aggregate utility. We measure
the quantity of commodity x horizontal y to the left from the origin O, and that
of commodity y vertically. The curve a is the indifference curve corresponding to
the maximized aggregate utility and the line α is the aggregate budget line whose
slope indicates the relative price of x and y. The curve a is tangent to α at C,
where the aggregate utility is maximized, being subject to the aggregate budget.
As for the domestic production, the transformation curve which gives possible
combinations of two commodities the country can produce is a straight line as
in Figs. 5.1 and 5.2, if the labor is the only factor of production. Let us suppose,
however, that both x and y industries have the given quantity of the non-labor
specific factor of production while labor is freely mobile between industries. Now
the transformation curve is a curve concave to the origin. In Fig. 8.2, it is the curve
SR which is concave to the origin T . The maximum quantity of x which can be
produced by using the total labor force is ST while that of y is RT. The reason for
such a shape of SR is that the marginal productivity of labor is diminishing in each
industry, because the quantity of the specific non-labor factor of production is given.
The ratio of the increase of y to the decrease of x will diminish, if we move the labor
gradually from the x industry to y industry.
Suppose, in Fig. 8.2, the curve a and SR are tangent to each other at C, and
the line α is the common tangent line. Looking from the origin O, the point
C indicates the domestic demand for two commodities, while the same point C
Fig. 8.3 Derivation of a trade indifference curve
shows the domestic supply of two commodities, if we look it from the origin T .
Since (domestic supply − domestic demand) = export, and (domestic demand −
domestic supply) = import, the position of T relative to O indicates the export of
x and the import of y. In other words, T gives a combination of the export of x, i.e.,
DE = FT, and the import of y, i.e., AB = FO, which assures the aggregate domestic
consumers the level of utility corresponding to the given aggregate consumption
indifference curve a , when the domestic production is carried out so as to maximize
the value of its products at the given relative price.4
Using Fig. 8.2, Meade derived the trade indifference curve a of a country from
the aggregate consumption indifference curve a . In Fig. 8.3, Ic is a consumption
indifference curve and the transformation curve PR is tangent to it at the point S. The
point Q indicate the corresponding export of x and import of y. The transformation
curve PR is slid so that Ic is always tangent to it and the PQ line remains parallel
to the x axis. Suppose PRQ is moved to P R Q and the point S to the point S . The
transformation curve P R is now tangent to the consumption indifference curve Ic
at the point T . If we move the transformation curve further in this way, we have
the locus of points Q, Q , and so on. Such a locus of Q is called trade indifference
curve corresponding to the consumption indifference curve Ic . To any consumption
indifference curve, we can construct a corresponding trade indifference curve.
The shape of a trade indifference curve looks very similar to that of the
consumption indifference curve from which it is derived. As a matter of fact, the
4 Figure 8.2 is taken from Takayama (1972, p. 17).

8 Marshall and Offer Curve 65
Fig. 8.4 Derivation of an

offer curve
slope of two curves are identical at the corresponding points, i.e., the point S and
the point Q, the point T and the point Q , etc. Consider the slope of QQ . The
vertical distance between Q and Q , which is the increase in the import of y or
(the increase in consumption of y) + (the decrease in production of y), is (a + a).
The horizontal distance between Q and Q , which is the increase in the export of x or
(the decrease in consumption of x) + (the increase in production of x), is (b + b).
The slope of QQ is then (a + a)/(b + b). Now let Q converge to Q. Then S
converges to S. The ratio a/b converges to the slope of the transformation curve at
S and the ratio a /b converges to the slope of the consumption indifference curve
at S. Since a/b = a /b in the limit, it is also equal to (a + a )/(b + b ) (Meade 1952,
p. 14).
Just as a consumer must obey the budget constraint in the determination of the
demand for commodities, a country must observe the balance of trade constraint in
the determination of international trade. When the terms of trade or relative price of
the exportables and the importables is given, it is given as the line OT in Fig. 8.4,
where the export (import) of x is measured horizontally to the right (left) from the
origin O, and the import (export) of y, vertically to the above (below) from the
origin O. The balance of trade requires the equality of the value of the export to
that of the import, so that the slope of OT is equal to the given terms of trade. The
export of x and the import of y are determined at the point A on the line OT , where a
trade indifference curve is tangent to the line OT . This is because, by construction,
the slope of trade indifference curve at A is identical to that of the corresponding
points on the corresponding consumption indifference curve and the transformation
curve.
If the given terms of trade is changed in Fig. 8.4, OT is rotated around O and
the position of point A is changed so that a new trade indifference curve is tangent
to the new OT line. The locus of such changing point A is exactly the curve which
Marshall considered in Fig. 8.1, i.e., the so-called offer curve. In other words, a
movement along Marshall’s curve in Fig. 8.1 should be considered as attended with
rearrangements of domestic consumption and production in Figs. 8.2, 8.3, and 8.4.
Figure 8.1 describes the movement of the hand of a clock, while the corresponding
unseen movements of the machinery is explained in Figs. 8.2, 8.3, and 8.4.
8.1 Problems
8.1. Consider a two-commodity model. Derive the offer curve by the use of the
general equilibrium domestic demand and supply curves which describe demand
and supply as functions of the relative price. See Gandolfo (1986, p. 46).
8.2. In Fig. 8.2, the transformation curve SR is tangent to the relative price line α
at C. In other words, the value of the total products is maximized. Why?
8.3. Derive offer curves in J. S. Mill’s model of international trade described in
Chap. 5 (Figs. 5.1 and 5.2) above.
8.4. If the equilibrium A is displaced in Fig. 8.1, England is assumed to act in such
a way as to change the amount of x in the horizontal direction of her offer curve.
Similarly, Germany is assumed to adjust y vertically in the direction of her offer
curve. Discuss the stability of the equilibrium A. See Samuelson (1955, pp. 266–7).
Bibliography
Edgeworth, F. Y. (1925). Papers relating to political economy. London: Macmillan.

Gandolfo, G. (1986). International economics, vol. 1. Berlin: Springer.
Marshall, A. (1879)[1975]. The pure theory of foreign trade. In J. K. Whitaker (Ed.), The early
economic writings of Alfred Marshall, 1867–90, vol. 2 (pp. 111–181). London: Macmillan.
Meade, J. E. (1952). A geometry of international trade. London: Allen & Unwin.
Newman, P. (1965). The theory of exchange. Englewood Cliffs, NJ: Prentice-Hall.
Samuelson, P. A. (1955). Foundations of economic analysis. Cambridge, MA: Harvard University
Press.
Takayama, A. (1972). International trade. New York: Holt, Rinehart and Winston.
Part II
Modern Theory and Recent Developments
of International Trade
Chapter 9
Theory of Production
To study the modern theory of the international trade, it is necessary to be familiar

with the elementary theory of production. The basic model of an economy to
be considered is a two-commodity two-factor one of a country. There are two
industries, each of which produces a single different commodity from the input of
two factors of production, say, labor and capital.1 The technology of an industry is
given as an aggregate production function which relates the inputs of two factors of
production to the output of a commodity.
Consider a production function
X = F(L, K) (9.1)
where X, L, and K are, respectively, the output, the input of labor, and the input of
capital. F is assumed to be increasing with respect to L, given K, and with respect
to K, given L. Suppose L is increased by ΔL, given K. The resulted increase in X is
ΔF, and its rate of change is ΔF/ΔL, i.e.,
ΔF F(L + ΔL, K) − F(L, K)

= . (9.2)
ΔL ΔL
If ΔL is infinitesimal, then ΔF/ΔL is denoted by ∂ F/∂ L and is called the partial

derivative of F with respect to L.2 It is the marginal product of labor and itself is
also a function of L and K. It is assumed to be decreasing with respect to L, since K
is kept unchanged. Similarly, we can define the marginal product of capital, ∂ F/∂ K.
In Fig. 9.1, L is measured horizontally, and K, vertically. The curve AA is called
an isoquant and shows the relation between L and K in the production function (9.1),
1 Itmight be better to use land instead of capital (Kemp 1964, pp. 9–10). We simply follow,
however, the more conventional way so as to be able to use such familiar technical terms as capital–
labor ratio, wage–rental ratio, etc.
2 “Partial” here implies only L is changed while K is kept unchanged.

70 9 Theory of Production
Fig. 9.1 Expansion line
when the output X is given. Its slope is given by
∂F
= − ∂L
dK
(9.3)
dL ∂F
∂K
since
∂F ∂F
dL + dK = 0 (9.4)
∂L ∂K
must be satisfied for infinitesimal changes in L and K, dL and dK, so as to keep
the level of X unchanged. The line BB, on the other hand, is called a cost line,
corresponding to
wL + rK = C (9.5)
where w, r, and C denote, respectively, the rate of wage, the rent for capital, and the
total cost of production. Its slope is
dK w
=− . (9.6)
dL r
Since AA and BB are tangent to each other at E, the cost of production for the given
output X corresponding to the isoquant AA is minimized at the point E. It is easily
seen that the higher is the wage–rental ratio w/r, the higher is the capital–labor ratio
K/L.
9 Theory of Production 71
In the modern theory of international trade, it is often assumed that the production
function (9.1) is a linear homogeneous function. This assumption implies that the
output X is changed proportionally if all the inputs, i.e., L and K, are changed
proportionally. More exactly,
F[(λ L), (λ K)] = λ F(L, K) for any λ > 0. (9.7)
In other words, economies or diseconomies of scale are ruled out and the constant
returns to scale is assumed. There exists no specific factor of production which is
used only in a certain industry.
If a production function is a linear homogeneous function, which is also called
a homogeneous function of the degree 1, the corresponding marginal products of L
is a homogeneous function of the degree 0, which implies the unchanged marginal
product of labor for the proportional changes in L and K. In other words, (9.7)
implies for any λ > 0,
FL [(λ L), (λ K)] = FL (L, K) (9.8)
where FL = ∂ F/∂ L is the marginal product of labor. This can be seen by considering
an infinitesimal change of L in (9.7). Since the rate of change of both sides must be
equal,
d(λ L)
FL [(λ L), (λ K)] = λ FL (L, K) (9.9)
dL
since a change in L first induces the change in (λ L), which in turn induces the
change in F in the left-hand side. Then, (9.8) follows from (9.9). Similarly,
FK [(λ L), (λ K)] = FK (L, K) for any λ > 0 (9.10)
where FK = ∂ F/∂ K is the marginal product of capital, can be derived from (9.7) so
that the marginal product of capital is also a homogeneous function of the degree 0.
In Fig. 9.1, the isoquant AA is shifted but its slope remains unchanged if we
change labor input L and capital input K proportionally. If wage w and rent r remain
unchanged, then, the point A of the minimized cost moves on the line OD which is
called the expansion line. In other words, if factor prices remain unchanged, output
and all the inputs are changed proportionally.
In a competitive equilibrium, the marginal product of labor should be equalized
to the rate of wage w expressed in terms of the product, and the marginal product of
capital, to the rate of rent r for capital. Suppose, for example, the marginal product
of labor is higher (lower) than the rate of wage. Then, it is profitable to increase
(decrease) the labor input. If the production function F is linear homogeneous with
respect to labor L and capital K, then, the total product X is exhaustively distributed
to the factors of production, i.e., labor and capital, so that
X = F(L, K) = wL + rK (9.11)
where w and r are expressed in terms of x. This can be demonstrated by the

differentiation of (9.7) with respect to λ , or the consideration of the effect of an
infinitesimal change in λ . Because the rate of change induced by it must be equal in
both sides of (9.7), we have
d(λ L) d(λ K)
FL [(λ L), (λ K)] + FK [(λ L), (λ K)] = F(L, K) (9.12)
dL dL
since in the left-hand side, (λ L) and (λ K) are changed first by a change in λ and
then F is changed by changes in (λ L) and (λ K). Since w = FL [(λ L), (λ K)] and
r = FK [(λ L), (λ K)], we have (9.11) by substitution of λ = 1 into (9.12).
Now let us consider a two-commodity two-factor model of a country. The factor
endowments, the total quantity of L and K, are assumed to be given and factors
are freely mobile between two industries, x industry and y industry. Production
functions of two industries are
X = F(Lx , Kx ), (9.13)
Y = G(Ly , Ky ) (9.14)
where X, Lx , Kx , Y , Ly , Ky are, respectively, the output of commodity x, labor and

capital inputs in x industry, the output of commodity y, labor and capital inputs
in y industry. Production functions F and G are linear homogeneous. If factor
endowments are given,
Lx + Ly = L∗ , Kx + Ky = K ∗ (9.15)
must be satisfied, where L∗ and K ∗ are the total available quantity of labor and
capital.
The transformation curve between x and y is a straight line in Figs. 5.1 and 5.2,
where labor is the only factor of production. It is a curve concave to the origin in
Figs. 8.2 and 8.3, where we assume the existence of the given specific factor of
production in each industry, in addition to labor which is freely mobile between
industries. Having two freely mobile factors of production now, we can have the
transformation curve concave to the origin, without assuming the existence of the
industry specific factors. In Fig. 9.2, the quantity of x is measured horizontally, and
that of y, vertically. The transformation curve is AB, where OA is the maximum
quantity of x to be obtained if all the labor and capital are used to produce x, and
OB is that of y similarly obtained. The curve AB is not convex to the origin O,
i.e., concave to the origin or a straight line, since the mid-point between A and B,
the point E, can be producible by dividing both labor and capital equally between
x and y industries, since OA = 2aO and OB = 2bO. The reason is, of course, that
9.1 Problems 73
Fig. 9.2 Transformation y

curve
b E
x
O a
production functions are linear homogeneous, so that proportional changes in inputs

result in the same proportional change in output.
It can be shown, furthermore, that the transformation curve AB is not a straight
line AB and E is located below, not on, the curve AB. Consider the so-called box
diagram of Fig. 9.3. The labor is measured horizontally, and the capital, vertically.
The horizontal distance between Ox and L∗ (or that of Oy and K ∗ ) is the total
quantity of labor, and the vertical distance between Ox and K ∗ (or that of Oy and
L∗ ) is the total quantity of capital. The inputs of labor and capital in x industry
are measured from the origin Ox , horizontally to the right and vertically upward,
and those in y industry, from the origin Oy , horizontally to the left and vertically
downward. The curve Ix (Iy ) is the isoquant of x (y) corresponding to the level of
output obtained from the inputs of L∗ /2 and K ∗ /2. In other words, the point E in
Fig. 9.3 corresponds to E in Fig. 9.2. Unless the form of two production functions,
F and G, are accidentally identical, Ix and Iy cannot be tangent to each other. If they
cross each other, as in Fig. 9.3, however, we can increase the outputs of both x and
y, by moving from E downward to the right, between Ix and Iy . This implies, in
Fig. 9.2, E is located below the curve AB and therefore the transformation curve AB
is concave to the origin.
9.1 Problems
9.1. Consider the so-called Cobb–Douglas production function, X = ALα K (1−α ) ,

where X, L, K, A, and α are, respectively, the output, labor input, capital input,
a positive constant, and a constant such that 0 < α < 1. Confirm that it is linear
homogeneous and that X is exhaustively distributed if the wage is equalized to the
marginal product of labor and the rent, to the marginal product of capital.
Fig. 9.3 Box diagram *
9.2. In Fig. 9.2, consider any two points on the transformation curve AB. Show that
the mid-point between such points is located below the curve AB, so that the curve
is concave to the origin.
Bibliography
Chapter 10
Heckscher–Ohlin Theory (1)
In the classical theory of international trade, the comparative advantage in the

sense of the comparative costs is simply given exogenously. In other words, it
is presupposed that different countries have different technology of production,
which includes the difference in natural conditions for the production like the
climate. In the modern theory of international trade, however, it is assumed that
different countries have the identical technology which is given in the form of the
identical production function. The comparative advantage of the different countries
is explained, then, not by the difference in technology, but by the difference in the
factor endowments. Such a modern theory is generally known as Heckscher–Ohlin
theory, because the groundwork for substantial developments in the theory is laid
by Eli Heckscher (1919) and Bertil Ohlin (1933).
The assumptions of the free trade and the identical production functions leads
to the international equalization of factor prices, the wage and the rent, since the
domestic factor prices are obtained from the common international commodity
prices through the marginal productivity theory of factor prices. The capital–labor
ratio, the ratio of the capital input to that of labor, must be identical for the same
industry located in the different countries. For a country to which the capital (labor)
is richly endowed relative to the labor (capital), then, the scale of the industry with
the higher (lower) capital–labor ratio should be large so that the capital (labor)
can be fully utilized (employed). In other words, the capital (labor) rich country
is specialized, though not necessarily completely, in the production of commodity
for which capital (labor) is more intensively used relative to labor (capital). If
there is no international difference in the taste of consumers, therefore, we can
conclude that the capital rich country exports the capital intensive commodity and
imports the labor intensive one, and vise versa. This is the so-called Heckscher–
Ohlin theorem.
Consider a two-commodity two-factor model of a country. The factor endowment
given to this country is L∗ of the labor population and K ∗ of the capital accumulated.
These factors must be allocated to two industries so that

76 10 Heckscher–Ohlin Theory (1)
L1 + L2 = L∗ , (10.1)
K1 + K2 = K ∗ (10.2)
where L1 , L2 , K1 , and K2 are, respectively, the labor input in the first and second
industries, and those of capital. The production functions of the two industries are
X1 = F1 (L1 , K1 ), (10.3)
X2 = F2 (L2 , K2 ) (10.4)
where X1 and X2 are, respectively, the output of the first commodity produced in the
first industry and that of the second commodity in the second industry.
Let us assume the constant returns to scale so that production functions are linear
homogeneous. Suppose both L1 and K1 are multiplied by (1/L1 ) in (10.3). Since
this proportional change in inputs produces the change in the output of the same
proportion, we have from (10.3)
X1 L K K
1 1 1
= F1 , = F1 1, = f1 (k1 ) (10.5)
L1 L1 L1 L1
where k1 = K1 /L1 . In other words, the per-capita output of the first commodity is a
function of the capital–labor ratio of the first industry only. Similarly, from (10.4),
we have
X2 L K K
2 2 2
= F2 , = F1 1, = f2 (k2 ). (10.6)
L2 L2 L2 L2
The per-capita output of the second commodity is a function of the capital–labor

ratio of the second industry only.
Because of the constant returns to scale, the scale of the production does not
matter, so that we have only to consider the case of a single laborer, i.e., the per-
capita production functions, f1 (k1 ) and f2 (k2 ). The marginal product of capital in
the first industry, which should be equalized to capital rent r, is
d f1 (k1 )
r= = f1 (k1 ) (10.7)
dk1
where r is the rent in terms of the first commodity. Similarly, we have
d f2 (k2 )
r=p = p f2 (k2 ) (10.8)
dk2
where p is the price of the second commodity in terms of the first.

10 Heckscher–Ohlin Theory (1) 77
As for the wage w, we can use the fact that the total product is distributed
exhaustively by the marginal product factor pricing,1 so that
f1 (k1 ) = w + rk1 = w + f1 (k1 )k1 (10.9)
where w is the wage in terms of the first commodity. Similarly,
p f2 (k2 ) = w + p f2 (k2 ) (10.10)
where p is the price of the second commodity.

Now we have four unknowns, w, r, k1 , and k2 , which are to be determined
by four equations (10.7), (10.8), (10.9), and (10.10), when p is given. If we
assume the unique solution,2 then, the factor prices, w and r, are equalized between
any two countries which have the identical production functions and trade two
commodities at the same international price. This is the so-called international
factor price equalization theorem. Factor prices are equalized internationally, if
factors are completely mobile between countries. Even if factors of production
are completely immobile, however, factor prices are equalized internationally, if
the commodities are freely mobile internationally, so that there is no import tariff,
quantity restrictions of imports, etc.
If the factor prices, w and r, and the capital–labor ratio in two industries, i.e.,
k1 and k2 , are given, then, the scale of two industries are determined by (10.1) and
(10.2), which can be rewritten as (10.1) and
k1 L1 + k2 L2 = K ∗ . (10.11)
Since capital–labor ratio, k1 and k2 , are already determined, we can solve (10.1) and
(10.11) to obtain L1 and L2 . If we divide both sides of (10.11) by L∗ ,
K∗
a 1 k1 + a 2 k2 = (10.12)
L∗
where a1 = L1 /L∗ and a2 = L2 /L∗ are relative scale of the first and the second
industries such that a1 + a2 = 1.
Consider now an international economy of two countries having the identical
production functions but different factor endowment ratio K ∗ /L∗ . If the free trade
of commodities is assumed, factor prices and capital–labor ratios in two industries,
k1 and k2 , are equalized between two countries. Suppose the first commodity is
more capital intensive than the second so that k1 > k2 and that the first country has
the higher factor endowment ratio K ∗ /L∗ than the second. Then, from (10.12), we
can see that the relative scale of the first industry a1 is larger in the first country
1 See Chap. 9, pp. 69–70 for the exhaustive distribution.

2 See Chap. 12, p. 90 for the case where the solution is not unique.
Fig. 10.1 Segment of * *

equalization
than in the second country. In other words, in a country to which the capital (labor)
is abundantly endowed relative to the labor (capital), the scale of the industry with
the higher (lower) capital–labor ratio should be large. If there is no difference in
consumer taste between countries, the capital (labor) rich country exports the capital
(labor) intensive commodity and imports the labor (capital) intensive commodity.
While this Heckscher–Ohlin theorem itself sounds plausible, we must note that
it is based on a lemma, the factor price equalization theorem, which we must admit
is a somewhat counter-intuitive one. It must be emphasized that the latter theorem
is based on an implicit supposition of the incomplete specialization. Although each
country is specialized in the industry in which the factor of production abundantly
endowed is intensively used, the specialization is not complete in the sense that
the production in the other industry is not yet terminated completely. To prove the
factor price equalization, it is necessary to assume that each country is producing
both commodities, however small is the output of one commodity. Let us consider
how plausible is the situation of the incomplete specialization and the factor price
equalization.
In Fig. 10.1, we measure the wage–rental ratio w/r horizontally, and the ratio of
the endowed capital to the endowed labor K ∗ /L∗ of the two countries as well as
the capital–labor ratio of the two industries, vertically. The upward sloping curves
k1 and k2 show that the capital–labor ratio is an increasing function of the wage–
rental ratio w/r in each industry. It is assumed that the first industry is always more
capital intensive than the second, i.e., k1 > k2 for any level of w/r. Suppose the first
country is capital rich country and K ∗ /L∗ is OB, while the second country is capital
poor with K ∗ /L∗ = OC. From the point of view of the first country, the possible
range of the wage–rental ratio w/r is be, since she is not completely specialized
Fig. 10.2 No segment of * *

equalization
in an industry if Ob < w/r < Oe. Similarly, the possible range of the wage–rental
ratio for the second country is ad. In the case of Fig. 10.1, the two ranges overlap
and admit of a common part bd, which is called the segment of equalization (See
Gandolfo 1994, p. 87).
If this segment of equalization exists, it is possible to have the equalization
of factor prices and incomplete specialization. Suppose the equalized w/r is Oc.
The capital–labor ratio in the first industry, k1 , is OA, while that in the second
industry, k2 , is OD in both of the two countries trading each other. In the first
capital rich country, OB is a weighted average of OA and OD, as can be seen in
(10.12). The weight for the first, capital intensive industry, a1 is relatively large, so
that the country is incompletely specialized in the first industry, since the weight
for the second industry, a2 , is, though small, but still positive. Similarly, in the
second, capital poor country, OC is a weighted average of OA and OD, and a2 is
relatively large, so that she is incompletely specialized in the labor intensive second
industry, while a1 is small but still positive.
The reason that there exists a segment of equalization bd in Fig. 10.1 is that
the points of the endowed capital–labor ratio of two countries, i.e., B and C,
are, though different, closely located each other. In other words, two countries
are similar, though not identical, countries from the point of view of the factor
endowments. Figure 10.2, on the other hand, describes the situation in which two
countries involved are very much different from the point of view of the factor
endowments. The wage–rental ratio w/r is measured horizontally, and the capital–
labor ratio, vertically. The implications of upward sloping curves k1 and k2 are
identical to those in the case of Fig. 10.1. The difference from Fig. 10.1 is, however,
there exists no segment of equalization. The endowed capital–labor ratio K ∗ /L∗ of
the first, capital rich country is OB, while that of the second, capital poor country
is OC. The possible range of the wage–rental ratio w/r for the first country is be,
while that of the second country is ad. The two ranges do not overlap and admit
of no common part. The incomplete specialization and the factor price equalization
is impossible in this case. The reason is clearly that two countries are too much
different from the point of view of the factor endowment.
10.1 Problem
10.1. Suppose both of the two industries have Cobb–Douglas production functions.
What is the condition for the first industry to have higher capital–labor ratio than the
second industry?
Bibliography
Gandolfo, G. (1994). International economics. Berlin: Springer.

Heckscher, E. F. (1919). Utrikshandelns verkan pa inkomstfoerdelningen. Ekonomist Tradskrift,
21(Del 2), 1–32.
Ohlin, B. (1933). Interregional and international trade. Cambridge, MA: Harvard University Press.
Chapter 11
Heckscher–Ohlin Theory (2)
The basic economic model of the so-called Heckscher–Ohlin Theory is the two-
country two-commodity two-factor model where production functions are linear-
homogeneous and the endowment of two factors of production is exogenously given
for each country. Besides the factor price equalization theorem and Heckscher–Ohlin
theorem discussed in Chap. 10, we can have also two additional well-known the-
orems derived from this Heckscher–Ohlin model. The first one is the so-called
Stolper–Samuelson Theorem (Stolper and Samuelson (1941)) which considers the
effects of an import tariff on the factor prices. In other words, it concerns with the
changes in factor prices caused by a change in the relative price of two commodities.
The second one is called Rybczynski Theorem (Rybczynski 1955) which deals with
the changes in the scale of production of two industries caused by a change in the
factor endowment in the case of a small country. This is the problem of the effects
of the economic growth on the relative scale of the two domestic industries, when
the relative price of two commodities is assumed to be constant.
Consider the unit cost of production of the i-th commodity Ci
Ci = wLi + rKi (i = 1, 2) (11.1)
where w, r, Li , and Ki are, respectively, the rate of wage, capital rent, labor input,
and capital input. Since Ci is the unit cost, Li and Ki must satisfy
1 = Fi (Li , Ki ) (i = 1, 2) (11.2)
where Fi is the production function of the i-th industry to produce the i-th
commodity. Let us consider the effect of infinitesimal changes in w and r, dw and
dr, on Ci . The unit cost Ci is changed not only directly by changes in w and r, dw
and dr, but also indirectly through the changes in Li and Ki , dLi and dKi , which are
caused by changes in w and r. In view of (11.1), then,
dCi = wdLi + rdKi + Li dw + Ki dr (i = 1, 2) (11.3)

where dCi is the change in the unit cost of the i-th commodity caused directly and
indirectly by changes in the wage and the rent. Since the labor and capital inputs, Li
and Ki must satisfy (11.2), however, we have
∂ Fi ∂ Fi
dLi + dKi = 0 (i = 1, 2) (11.4)
∂ Li ∂ Ki
where (∂ Fi /∂ Li ) and (∂ Fi /∂ Ki ) are the marginal product of labor and that of capital.
Since the marginal product of labor (capital) is equal to wage (rent),
∂ Fi ∂ Fi
= w, =r (i = 1, 2) (11.5)
∂ Li ∂ Ki
we have from (11.3), by substituting (11.4) and (11.5),
dCi = Li dw + Ki dr (i = 1, 2). (11.6)
In other words, we can estimate the changes in the unit cost, assuming as if the
inputs of labor and capital do not change.
Let us now assume that the first commodity is the capital intensive commodity
while the second commodity is labor intensive commodity, so that we always have
K1 K2
k1 = > k2 = . (11.7)
L1 L2
If both w and r rises but the wage–rental ratio w/r rises too, then, the unit cost of
the labor intensive second commodity rises more than the unit cost of the capital
intensive first commodity. Since the unit cost is equal to the price of the commodity,
this implies that the relative price of the second commodity in terms of the first
one, which is now defined as p, must rise. In Fig. 11.1, we measure the price of the
second commodity in terms of the first, p, horizontally and the wage–rental ratio,
w/r, vertically. The relation between p and w/r is shown by the upward sloping
curve.
Suppose the country under consideration is a capital rich country so that she
is exporting the capital intensive first commodity and in return importing the
labor intensive second commodity. If the government imposes tariff on import,
the domestic price p of the second commodity in terms of the first must rise, if
the international price remains unchanged.1 The wage–rental ratio w/r must rise
and therefore capital–labor ratio in both industry, k1 and k2 , must rise, as is seen
in Fig. 11.1.
In Chap. 10, we derived the capital rent r as
d f1 (k1 )
r= = f1 (k1 ) (11.8)
dk1
1 See Metzler (1949) for the case where this assumption does not hold.
Fig. 11.1 Price and

wage–rental ratio
where f1 (k1 ) is defined as the per-capita production function of the first commodity
and f1 is the marginal product of capital, which is decreasing with respect to k1 . The
capital rent r in terms of the first commodity falls, then, with the rise in k1 which
is caused by the rise in p. It also falls in terms of the second commodity, since we
also have
pd f2 (k2 )
r= = p f2 (k2 ) (11.9)
dk2
in Chap. 10, where f2 (k2 ) is defined as the per-capita production function of the
second commodity and f2 is the marginal product of capital, which is decreasing
with respect to k2 . The capital rent in terms of the second commodity, r/p, falls by
the rise in k2 .
As for the rate of wage w, we can consider
dg1 (1/k1 )
w= = g1 (1/k1 ) (11.10)
d(1/k1 )
where g1 (1/k1 ) is defined as the per-unit of the capital production function of the
first commodity, F1 (L1 /K1 , 1), an increasing function of labor capital ratio, L1 /K1 =
(1/k1 ), and g1 is the marginal product of labor, which is decreasing with respect
to (1/k1 ). The wage w in terms of the first commodity rises, then, with the fall
in (1/k1 ), which is caused by the rise in p. Similarly, we can also show the rise in
w/p, the wage in terms of the second commodity.
Now we can conclude that the imposition of an import tariff is favorable to the
owners of the scarcely endowed factor of production. If labor is the scarce factor of
the United States, it is a wise policy of labor unions to demand the government to
impose import tariffs.
Fig. 11.2 Allocation of labor

and capital
In Fig. 11.2, labor is measured horizontally, and capital, vertically. Originally

BO1 of labor and BO2 of capital are endowed to a country. If the relative price of
two commodities is given, factor prices, wage w and capital rent r, capital–labor
ratio in two industries, k1 and k2 , and the output of two commodities, X1 and X2 ,
are determined as is explained in Chap. 10. In Fig. 11.2, the input of labor and
capital in the first industry is measured from the origin O1 while those in the second
industry, from the origin O2 . The straight line O1 E is the expansion line of the first
industry. It is the locus of the points which minimize the cost of production for the
given isoquants. Similarly, the straight line O2 E is the expansion line of the second
industry. The scale of the first industry, i.e., the output of the first commodity, is
given by the length of O1 E and that of the second industry, but that of O2 E. The
first industry is labor intensive and the second, capital intensive.2
Suppose that the original labor endowment BO1 is increased by BC while the
capital endowment remains at the original level BO2 . Assuming that this country
is a small country so that the given relative price of two commodities remains
unchanged, what will be the changes in the size of two domestic industries? Since
factor endowments in general is increased, that is, the endowment of one factor
is definitely increased, while the endowment of no other factors is decreased, one
might suppose that the scale of both industries will be expanded. In other words, the
possible set of output of two commodities is enlarged or the transformation curve
of two commodities is shifted outwardedly. It is difficult to suppose that the scale of
any industry is diminished.
In Fig. 11.2, now the original box O1 AO2 B is enlarged to that of O1 AO2C. The
origin from which inputs in the second industry are measured is shifted from the
point O2 to the point O2 . Since the relative price of two commodities remains
unchanged, there are no changes in factor prices and capital–labor ratios. The slope
of the expansion line of the first industry is unchanged, but the line itself can be
extended to the point E by the increase BC in the endowed labor. The expansion
2 For the expansion line, see Chap. 9, p. 71.

11.1 Problems 85
Fig. 11.3 Rybczynski

theorem
line of the second industry, on the other hand, now has to start from the new origin
O2 but its slope remains unchanged, so that it is O2 E , which is parallel to the old
expansion line O2 E. It is clear that the output of the first commodity is increased,
but that of the second is decreased, since the length of O1 E is longer than that of
O1 E, but that of O2 E is shorter than that of O2 E.
In Fig. 11.3, the output of the first commodity is measured horizontally, and that
of the second, vertically. The original transformation curve is AB. The set of points
OAB is the possible set of output of two commodities. The point E is the original
equilibrium, where the curve AB is tangent to the price line p, the slope of which
shows the given relative price of two commodities. When the labor endowment
is increased, the transformation curve is shifted outwardly to A B . The possible
set of output is enlarged to OA B . But the change is asymmetrical or skewed so
that the new equilibrium point E is located below the point E. The output of the
first commodity is increased, but that of the second is decreased.
When the endowment of only one factor is increased, the industry which uses
this factor intensively should be expanded. But this requires also the increase in
the input of other factor in this expanding industry, the supply of which must come
from the other industry. The industry which uses intensively the factor of which the
endowment is not increased, therefore, should be contracted.
11.1 Problems
11.1. By using (10.9), demonstrate that the wage w rises with the rise in the capital–
labor ratio, which is caused by the rise in p.
11.2. Discuss the following argument of Ricardo (1951, p. 46). “Adam Smith, and
all the writers who have followed him, have, without one exception that I know of,
maintained that a rise in the price of labour would be uniformly followed by a rise
in the price of all commodities. I hope I have succeeded in showing, that there are
no grounds for such an opinion, and that only those commodities would rise which
had less fixed capital employed upon them than the medium in which price was
estimated, and that all those which had more, would positively fall in price when
wages rose.”
Bibliography
Metzler, L. A. (1949). Tariffs, the terms of trade, and the distribution of national income. Journal
of Political Economy, 57, 1–29.
Ricardo, D. (1951). On the principles of political economy and taxation. Cambridge: Cambridge
University Press.
Rybczynski, T. M. (1955). Factor endowment and relative commodity prices. Economica, 22,
336–341.
Stolper, W. F., & Samuelson, P. A. (1941). Protection and real wages. Review of Economic Studies,
9, 50–73.
Chapter 12
Leontief Paradox
There exist two possible methods for the investigation, the inductive inference
and deductive inference. The inductive inference collects empirical observations
and infers the general conclusion from them. Although it is very useful for the
practical purposes, the inductive inference cannot arrive to the definite conclusion.
You observe as many white swans as you like in the northern hemisphere, but cannot
conclude that swan is white, since you cannot exclude the possibility to see black
swans in the southern hemisphere. You can collect all the cases in the past, but
cannot conclude definitely, since you cannot know the possible cases in the future.
To arrive at the definite conclusion, therefore, we have to rely on the deductive
inference. It starts with some assumptions and derive logical conclusions from
them. Some assumptions may be derived from inductive inference founded on the
empirical observations while other assumptions are made merely as the simplifying
assumptions. As far as the assumptions are admitted, and the logical operations are
correct, the derived conclusion must be accepted.
Since there is no assurance that assumptions are empirically true, however, the
conclusion of the deductive inference cannot be assured to be empirically right.
Such a conclusion must be subject to the empirical tests, since to test the empirical
validity of assumptions themselves is, in general, more difficult. If the conclusion
is empirically refuted, something must be wrong with respect to assumptions. We
have to discard at least some of them and replace them by new assumptions. With
the new set of assumptions, then, the deductive inference and the empirical test
of its conclusion must be repeated. If the conclusion is not refuted, however, it
does not imply that it is empirically true. It merely means that it is not refuted
temporarily, since we cannot exclude the possibility that it will be refuted by
some other empirical tests in the future. When carefully planned experiments are
possible, as in the case of some natural sciences, it is easy to refute empirically
the conclusion of the deductive inferences. When it is not, as in the case of social
sciences, the empirical test must rely on the observation of empirical data which we
cannot control and it is very difficult to see whether the data given is relevant and
appropriate for the refutation of the conclusion.

88 12 Leontief Paradox
Our Heckscher–Ohlin theorem, discussed in Chap. 10, is a typical example of the

conclusion of the deductive inference logically derived from assumptions. From the
assumptions on the production functions, consumer taste, the endowments of factors
of production, etc., the theorem is derived logically that a country exports those
commodities which use intensively such factors of productions that are endowed
abundantly, and imports those commodities which use intensively such factors of
production that are endowed scarcely. It was Leontief (1954) who tried to check this
theorem empirically.
Leontief (1954) considered a Heckscher–Ohlin international trade model of two
countries, the USA and the rest of the world. Two commodities considered are
a composite commodity “US exports” and another composite commodity “US
competitive import replacements.” Competitive import is defined as the import of
commodities which can be and are, at least in part, actually produced by domestic
industries. In other words, two domestic industries are the export industry and the
import competing industry. The question to be studied is, then, whether it is true that
the United States exports commodities the domestic production of which absorbs
relatively large amounts of capital and little labor and imports foreign goods which,
if produced at home, would employ a great quantity of indigenous labor but a small
amount of domestic capital. By applying his famous input–output analysis to the
1947 input–output table of the US economy, Leontief computed the total (direct
and indirect) input requirements of capital and labor per unit of two composite
commodities. The result is that the capital–labor ratio is approximately 14 for US
Exports and 18 for US Import replacements. In other words, USA exports labor-
intensive commodities and imports capital-intensive commodities. This is called
Leontief paradox, since USA was considered generally to be a capital abundant
country relative to the rest of the world. Heckscher–Ohlin theorem seems to be
empirically refuted.
One might wonder, of course, whether Leontief’s result was relevant or
appropriate to the empirical test of Heckscher–Ohlin theorem. As a matter of fact,
Leontief himself was the first who wondered. If the labor is measured by using
the man-years units and the capital, by dollars in 1947 prices, as was done in his
study, certainly USA is a relatively capital abundant country. Since labor is more
efficient in USA than in the rest of the world, however, USA should be considered
as a labor abundant country, if the labor is measured by using the efficiency units.
“[W]ith a given quantity of capital, one man-year of American labor is equivalent
to, say, three man-years of foreign labor. Then, in comparing the relative amount
of capital and labor possessed by the United States and the rest of the world—the
total number of American workers must be multiplied by three—. Spread thrice as
thinly as the unadjusted figures suggest, the American capital supply per “equivalent
worker” turns out to be comparatively smaller, rather than larger, than that of many
other countries.” Leontief’s empirical data, then, does not refute Heckscher–Ohlin
theorem.
Another possibility is that the choice of labor and capital as two factors of
production, as was done by Leontief, might not be appropriate for the case of
trade between USA and the rest of the world. In view of the fact, for example,
12 Leontief Paradox 89
that US oil-fields are less rich than those in Venezuela or in the Arabian countries,
it is important to consider natural resources as factors of production. Then, the
United States might import goods intensive in natural resources, which is the
relatively less abundant factors there, and export goods intensive in capital and
labor relative to natural resources. If so, again, Leontief’s empirical data does not
refute Heckscher–Ohlin theorem. For the details of literature which insists on this
possibility, see Gandolfo (1994, p. 96).
If one accept Leontief’s empirical result as the refutation of Heckscher–Ohlin
theorem, on the other hand, some of assumptions of the theorem must be rejected.
Among the assumptions of Heckscher–Ohlin theorem, one of the assumptions
which can be most easily doubted is that of the identical consumption pattern
between different countries. This may be particularly so in the case of Leontief
paradox since the per-capita income level is much different between USA and the
rest of the world. Suppose, relatively speaking, USA is abundantly endowed with
capital and labor, scarcely. If Heckscher–Ohlin theorem is right as far as the pattern
of incomplete specializations is concerned, USA is specialized in the production of
those commodities which are relatively capital intensive, though the specialization is
incomplete. Suppose, however, the domestic consumption pattern in USA is biased
to capital intensive commodities. Even though the domestic supply of such capital
intensive commodities is large, the domestic demand for them may be still larger.
Then, the domestic excess demand for such capital intensive commodities should
be supplied by the imports from the rest of the world, a country where capital is,
relatively speaking, scarcely endowed.
Since the information on technology is mobile internationally, we may assume
the identical production functions between different countries. As people are not
freely mobile between countries, however, we may not assume the identical level
of income, which include not only wage income but also that from capital,1 so that
the consumption pattern can be different in different countries. It is well known
that the ratio of the food consumption in the total expenditure is less for the high
income families than for the low income ones (Engel’s law). Can we suppose
high income USA families prefer the capital intensive commodities rather than
the labor intensive commodities? Unfortunately, this is not certain. High income
families may prefer such labor intensive commodities as expensive homemade
goods like homespun cloth rather than to such capital intensive commodities as
machine-made cheap goods produced in large-scale production factories.
Another assumption of Heckscher–Ohlin theorem, which may be doubted, is that
of no factor intensity reversal. It is assumed in the theorem that the commodity 1,
for example, is always more capital intensive (capital–labor ratio is higher) than
the commodity 2 for any value of the wage–rental ratio, i.e., k1 > k2 for any
given w/r. Suppose, however, that the commodity 1 is more capital intensive than
1 If
capital is abundantly endowed relative to labor, in comparison with the rest of the world, per-
capita income level is higher in such a country than in the rest of the world, since per-capita income
from capital is larger.
Fig. 12.1 Factor intensity

reversal
1
Fig. 12.2 Wage–rental ratio

not equalized
the commodity 2 when the wage–rental ratio is high, w/r > (w/r)1 ; but is less
capital intensive when the wage–rental ratio is low, w/r < (w/r)1 , as is shown in
Fig. 12.1. Then in Fig. 12.2, we cannot have the unique solution of w/r for the given
value of p, i.e., the price of the second commodity in terms of the first. For the same
p in the international market, it is possible that w/r for the country 1 is Oa while it is
Ob for the second country. The commodity 1 is less capital intensive (k1 < k2 ) in the
first country while it is more capital intensive (k1 > k2 ) in the second country. Then,
Heckscher–Ohlin theorem cannot be valid generally, since the exportables have the
same kind of factor intensity in both countries. It remains valid for one country only.
The case like Fig. 12.1 occurs if the production functions are of CES (constant
elasticity of substitution) type,
Y = [aL−b + (1 − a)K −b]−1/b (0 < a < 1, b > −1) (12.1)

Bibliography 91
where Y , L, and K denote, respectively, the output, the labor input, and the capital
input, while a and b are given constants.2 Using production functions of this type,
Minhas (1962) found that factor intensity reversals were quite frequent in the real
world. This suggests that, theoretically speaking, Leontief paradox is very likely to
occur.
What is most important with respect to Leontief paradox is, however, that the
presence of the paradox could be by no means systematically confirmed by the
subsequent studies carried out with respect to both USA and other countries. How,
then, should we interpret the significance of this almost single empirical evidence
against Heckscher–Ohlin theorem? It depends on the nature of the theorem. If it is
the exact theorem which insists that each country always exports the commodity
which uses her more abundant factor more intensively, it should be refuted by a
single empirical evidence against it. In view of the long-run and the aggregate nature
of Heckscher–Ohlin two-country two-commodity two-factor model, however, we
should rather interpret the theorem that most of the countries generally export
the commodities which use, on the average, their abundantly endowed factors
more intensively. Then, we have to retain such a theorem on the long-run average
tendency, unless it is repeatedly and systematically refuted empirically. This is the
reason why, in spite of Leontief’s paradox, Heckscher–Ohlin theorem is still in
the textbooks of international trade theory as one of the basic theorems.
12.1 Problems
12.1. Demonstrate that there is no factor intensity reversal if the production

functions are of Cobb–Douglas type.
12.2. a. Demonstrate that w/r = [a/(1 − a)]kb+1 , where w/r is the wage–rental
ratio and k is the capital–labor ratio, if the production function is of the constant
elasticity of substitution type, i.e., (12.1). Consider the elasticity of k with respect
to (w/r) and show it is constant.
b. Explain, then, that we have a case of Fig. 12.1, if two commodities have
the identical production function, i.e., the constant elasticity of substitution
production function but with different values of b.
Bibliography
Gandolfo, G. (1994). International economics, I. Berlin: Springer.

Kemp, M. C. (1964). The pure theory of international trade. Englewood Cliff, NJ: Prentice Hall.
2 Forthe constant elasticity of substitution production functions, see Kemp (1964, pp. 22 and 57).
See also Problem 12.2.
Leontief, W. (1954). Domestic production and foreign trade; the American capital position
re-examined. Economia Internazionale, 7, 3–32.
Minhas, B. S. (1962). The homohypallagic production functions, factor intensity reversals, and the
Heckscher–Ohlin Theorem. Journal of Political Economy, 70, 140–168.
Chapter 13
Domestic Distortions
Both countries can gain in welfare from international trade between them. The level
of welfare in each country is higher than in autarky. The gains from trade in this
sense have often been discussed so far. Both England and Portugal gain in Ricardo’s
famous numerical example of cloth and wine trade in Chap. 4. Even Marx admitted
that the poorer country, which he believed exploited by the richer country, gains
from international trade in Chap. 7. Finally, Meade demonstrated clearly the gains
from trade by using his trade indifference curves in Chap. 8. In Fig. 8.3, a trade
indifference curve of a country is tangent to her budget line (the balance of trade
line) at the point A, which indicates the export of x and the import of y. It is clear
that the point A is located on a higher trade indifference than the point O which
is the point of the autarky (no trade). To each trade indifference curve we have a
corresponding consumption indifference curve. The level of the utility of aggregate
consumers is, therefore, higher at point A than at O.
This demonstration of gains from trade is, however, based on an implicit
assumption that factors of production are fully and efficiently utilized in the
domestic economy. This is clear from the construction of trade indifference map
from the consumption indifference map in Fig. 8.2 where transformation curve is
tangent to a consumption indifference curve. The transformation curve shows the
maximum combinations of outputs of two commodities which can be produced
from the inputs of factors of production endowed. In other words, any points on
the transformation curve made possible by the full and efficient utilization of such
factors of production. If there exists unemployment of labor and/or underutilization
of capital, however, the combination of outputs of commodities cannot be located
on the transformation curve. If there exist such distortions in the domestic economy,
not only the domestic resources are fully and efficiently utilized in the autarky, but
also the introduction of international trade does not necessarily increase the welfare
of the country. In other words, there is a possibility of the negative gains from trade.
Let us consider the case of the labor unemployment caused by the rigidity of the
rate of wage in the labor market. Suppose a country produces two commodities so
that the given labor population must be allocated in the labor market between two

94 13 Domestic Distortions
Fig. 13.1 Unemployment 1 1
industries. Figure 13.1 shows it. The labor is measured horizontally, and the labor
productivity, vertically. The given total labor population is shown by the horizontal
distance between the two origins, O1 and O2 . The employment of the labor in the
first industry, which produces the first commodity, is measured from the origin O1
horizontally to the right, while the employment of the labor in the second industry
to produce the second commodity, from the origin O2 horizontally to the left. The
marginal productivity of labor in the second industry, which is expressed in terms of
the second commodity, MPL2 , is measured vertically from the origin O2 . Similarly,
the marginal productivity of labor in the first industry, expressed in terms of the first
commodity is MPL1 . Suppose the price of the first commodity is P1 and that of the
second commodity, P2 . Then, the marginal productivity of labor in the first industry,
expressed in terms of the second commodity, (P1 /P2)MPL1 is measured vertically
from the origin O1 .
The curve A2 A2 is the marginal productivity curve of the second industry. If O2 a
of labor is employed, the marginal productivity is ab in terms of the second
commodity. Since the employment is determined by the equality of the wage and the
marginal productivity, O2 a of labor is employed if the wage is ab. The dotted curve
A1 A1 is the marginal productivity curve of the first industry, measured in terms of
the second commodity. If the wage in terms of the second commodity is ab, O1 a of
labor is employed in the first industry. When the labor market functions well with the
flexible wage, we have a full employment equilibrium where the equilibrium wage is
ab, since O1 a + O2a = O1 O2 . Now suppose the given relative price (P1 /P2) declines
and the curve A1 A1 is now shifted downward to the curve B1 B1 . The equilibrium in
the labor market is shifted from the point b to the point d. The labor is still kept
fully employed, though the equilibrium wage is reduced. Such is the labor market
assumed to function behind the trade indifference curves in Fig. 8.3 (Chap. 8, p. 64),
where we are assured for the gains from trade.
13 Domestic Distortions 95
Fig. 13.2 Domestic

distortion
If the rate of wage, in terms of the first commodity, is rigid at O1W0 , however,
we cannot expect the full employment. Still O2 a is employed in the second industry,
but only O1 c is now employed in the first industry, when its marginal productivity
curve (in terms of the first commodity) is shifted down to B1 B1 . Out of total labor
population O1 O2 , ac is now left unemployed. The reduction in the relative price
(P1 /P2 ) does not expand the size of the second industry and merely diminished the
scale of the first industry. It is clear that the country’s production is now located not
on the transformation curve but below the curve, or inside the area bounded by the
curve. This is a typical example of the domestic distortion.
The transformation curve of the country is the curve AB in Fig. 13.2, where the
output of the first commodity, x1 , is measured horizontally, and that of the second
commodity, x2 , vertically. The point E in Fig. 13.2 corresponds to the point b in
Fig. 13.1. The point E is still on the transformation curve AB. At point E, the labor
is still fully employed and it is necessary to decrease the output of the second
commodity, if the first commodity is to be increased. The point G in Fig. 13.2 is,
however, like the point d in Fig. 13.1. It is located inside of the transformation curve
AB. It is possible to increase the output of the first commodity without reducing that
of the second commodity, since the labor is not fully employed there.
The downwardly sloping line which passes the point G in Fig. 13.2 is the price
line whose slope indicates the given price ratio (P1 /P2 ). It becomes steeper as
(P1 /P2 ) is higher. The point G is that of the autarky, which indicates the production
as well as the consumption of two commodities. In other words, at this point an
aggregate consumption indifference curve is tangent to the price line. If the price
ratio (P1 /P2 ) declines, the curve B1 B1 in Fig. 13.1 is shifted further down, so that
the employment as well as the output in the first industry is decreased, but those
in the second industry remain unchanged. In Fig. 13.2, the point of the domestic
production is shifted from G to H.
Suppose the given international price ratio (P1 /P2 ) is lower than that of the
autarky. After international trade, the country’s production moved from G to H in
Fig. 13.2. The new price line HI is the country’s budget line (trade balance line)
which is tangent to an aggregate consumption indifference curve at I. In other words,
this country imports the first commodity by exporting the second commodity. It is
clear, however, the utility level of the aggregate consumer at I (after trade) is lower
than that at G (autarky). The country’s gains from trade are negative. There is no
wonder, since the international trade intensified the domestic distortion, i.e., the
unemployment of labor due to the wage rigidity.1
One might wonder why there is no wage reduction in the face of unemployment.
This is a difficult question, not because there can be no answer, but because
there exist too many different answers. Perhaps, one might consider, for example,
efficiency wages (Blanchard–Fischer 1989, pp. 455–463). The productivity of labor
may be affected by the wage the firm pays. When laborers’ efficiency is affected by
the wage, a reduction in the wage may in the end increase rather than decrease cost.
The wage may accordingly be sticky because it is costly for firms to cut it, even
though there are unemployed laborers who might accept the wage lower than the
current wage.
Let us suppose that the second industry is the manufacture which is located in the
urban area and manufacturing firms consider efficiency wages so that the wage there
is fixed in terms of their own product, i.e., the second commodity. The employment
and output of the second industry is fixed and remain unchanged for any changes in
the relative price of two commodities. Even so, however, there is no unemployment,
if the first industry is the agriculture in the rural area where wages are flexible
and laborers are completely mobile between two areas. In Fig. 13.2, the domestic
production, if arrived at E, remains at E for further reduction of the relative price
(P1 /P2 ). There exist, then, gains from trade.
To explain the existence of unemployment, particularly in the high wage urban
area, Harris–Todaro (1970) argued that those unemployed do not move to the rural
area where they can find employment, at lower wages though. This is because the
expected income in the urban area is not lower than that in the rural area. As in
Fig. 13.1, we measure in Fig. 13.3 the employment in the first industry from the
origin O to the right and that in the second industry, from O∗ to the left. The marginal
productivity of labor in the second industry in terms of the second commodity, i.e.,
its own product, is measured from O∗ vertically, while that in the first industry, also
in terms of the second commodity, from O vertically. The curve MM shows the
marginal productivity in the second industry and the wage is fixed at O∗W there,
so that the employment is O∗ N. The marginal productivity in the first industry is
shown by LL and the employment there is OG, with the wage O∗V . The urban
unemployment is GN (Corden 1974, p. 145).
Why? Figure 13.3 is drawn in such a way that the area O∗ WJN is equal to the
area O∗ VRG. Out of the total labor population in the urban area, O∗ G, only O∗ N can
1 Figures 13.1 and 13.2 are taken from Itoh and Negishi (1987, pp. 4–5).
13.1 Problems 97
Fig. 13.3 Urban unemployment
be employed with the wage O∗W , while NG are unemployed (i.e., with zero wage).
Assume that those who are employed and those who are unemployed are chosen at
random with an equal probability of employment for each individual laborer and
with no serial correlation in employment. In other words, the urban labor force
steadily turns over and jobs are not tenured. Then, the expected (average) urban
wage is O∗ WJN/O∗ G, which is equal to the wage in rural area GR = O∗V , since
O∗ WJN is equal to O∗ VRG.
Now in Fig. 13.2, the domestic production, if arrived at E, cannot remain at E
for further reductions of the relative price (P1 /P2 ). It moves on EF from E toward
H. There can be, again, a case of the negative gains from trade due to the domestic
distortion.
13.1 Problems
13.1. In Fig. 13.2, suppose the terms of trade (relative price of the exportables and
the importables) is given by the slope of the line HI, where H is the point of the
domestic production and I, that of the consumption. If the terms of trade improves,
how does the welfare of the country change?
13.2. In Fig. 13.3, a laborer is assumed to be indifferent between the certain wage
income in rural area and the uncertain wage income in urban area if the expected
value of them is equal. Is he rational? Consider a game to toss a coin until it shows
heads. If the first head appears at the n’th toss, a price of $2n is paid. No rational
person would be willing to pay an arbitrarily large amount for the right to participate
in this game, though the expected value of the gain in the game is infinitely large.
See Borch (1990, pp. 4–5) for this St. Petersburg Paradox.
13.3. Can we decrease unemployment by giving subsidy to the second industry?
See Corden (1974, pp. 146–148).
Bibliography
Blanchard, O. J., & Fischer, S. (1989). Lectures on macroeconomics. Cambridge, MA: MIT Press.
Borch, K. (1990). Economics of insurance. Amsterdam: North-Holland.
Corden, W. M. (1974). Trade policy and economic welfare. Oxford: Oxford University Press.
Harris, J. R., & Todaro, M. P. (1970). Migration, unemployment and development. American
Economic Review, 60, 126–142.
Itoh, M., & Negishi, T. (1987). Disequilibrium trade theories. Chur: Harwood.
Chapter 14
Export Promotion and Welfare
In view of the acknowledged success of the post-war Japan, where export promotion
policies have played an important role, the role of subsidies in export industries
should be studied from the point of view of the welfare of a country in an
international economy. Except for the case of domestic distortions (see Prob-
lem 13.3), however, it is difficult to justify export subsidy by the use of the standard
two-country two-commodity model of the international trade theory. In fact, it
seems to be a well-established fact the export subsidies always reduce a country’s
economic welfare in a competitive economy with full price flexibility.
Because of the condition of the balance of trade, the import of a country must be
financed by her export. If the terms of trade, the relative price of the exportables to
the importables, is improved, therefore, the level of welfare of the country can be
increased.1 To improve the terms of trade, firstly, we can consider the imposition
of an import tariff. It reduces the domestic demand for the importables, and its
price is reduced, since the demand decreased against the unchanged supply in the
world market. Secondly, we can rely on a tax on the export, which also improves the
terms of trade through the higher price of our exportables, since our supply of it is
reduced against the unchanged demand in the world market (Lerner 1936s symmetry
theorem). In other words, export subsidies and the export taxes have opposite effects
on the economic welfare of the country.
Two Japanese economists, Itoh and Kiyono (1987), solved this difficulty by the
consideration of the model of more than two commodities. By so doing, they can
distinguish between nonmarginal goods and the marginal goods. Marginal goods
are defined as those that would be exported by only a small amount or not at all
under free trade but whose export can be promoted considerably by export subsidies.
1 See Fig. 8.3, p. 64, where the volume of the export is measured horizontally, and that of the import,
vertically. The slope of the line OT signifies the terms of trade. If the slope of the line OT is steeper,
the point A, where OT is tangent to a trade indifference curve, is on the curve corresponding to a
higher level of the welfare.

100 14 Export Promotion and Welfare
Export subsidies on marginal goods and those on nonmarginal goods have opposite
welfare effects for the country imposing these subsidies. The former enhances the
imposing country’s welfare, while the latter worsen it.
Consider, at first, a two-country, three-good, one-factor model, i.e., a variant of
the so-called Ricardian model of the modern interpretation, which we discussed in
Chap. 4. Suppose that there are two countries, the home and the foreign countries,
and three goods, goods 1, 2, and 3. The production technology of this Ricardian
economy can be represented by labor requirement coefficients of the three goods in
the two countries. Let ai and a∗i , i = 1, 2, 3, represent, respectively, the amount of
labor required to produce one unit of good i in the home country and the amount
required in the foreign country. The goods are numbered so that the smaller the
subscript number the greater is the home country’s comparative advantage in the
production of the good concerned, i.e.,
a∗1 a∗ a∗
> 2 > 3. (14.1)
a1 a2 a3
It is assumed that each country has an inelastic given supply of labor.

Suppose the pattern of specialization and trade are such that the home country
produces goods 1 and 2 and the foreign country produces goods 2 and 3. The home
country exports, i.e., supplies good 1 monopolistically in the world market, and
the foreign country exports, i.e., supplies good 3 similarly in the world market,
while there is no international trade of good 2. Assume that the home government
places an ad valorem export subsidy on good 2.2 Then, the home country exports
not only good 1 but also good 2 and imports good 3, while the foreign country
exports only good 3 and imports not only good 1 but also good 2. Since the home
production of good 2 must be increased for export, the labor has to be shifted from
the production of good 1 to that of good 2. The export, i.e., the supply of good
1 in the world market must be reduced so that its price has to be raised. In the
foreign country, the import of good 2 reduces the domestic production and the labor
is shifted to the production of good 3. The export, i.e., the supply of good 3 in the
world market must be increased so that its price has to be reduced.
Even if the subsidization of the export of good 2 reduces the welfare of the home
country, therefore, there is an indirect possibility of the welfare increase through
its effect on the relative price of good 1 to good 3. Export subsidies on a marginal
good, good 2, can increase the welfare through the improved terms of trade between
nonmarginal goods, goods 1 and 3. As a matter of fact, such a policy can increase
the welfare by expanding the range of the exportables, from good 1 to goods 1 and 2.
Itoh and Kiyono (1987) demonstrated this generally and rigorously by using
the so-called Ricardian model of Dornbusch, Fischer, and Samuelson (1977) where
2 If
the home government imposes only an export subsidy on good 2, the home consumers will
purchase from the foreign producers. To avoid this two-way trade problem, let us assume that the
home government imposes a prohibitively high import tariff on good 2.
14 Export Promotion and Welfare 101
Fig. 14.1 Marginal industry *
number of goods are infinitely large. The demonstration is highly mathematical, but
fortunately, Itoh and Kiyono also produced figures in which we can see their point
rather intuitively.
Goods are now indexed by a real number n on the closed interval [0, 1] on the real
line, such as the horizontal line O1 in Fig. 14.1. The labor requirement coefficients
of good n in the home and foreign countries are denoted, respectively, by a(n)
and a∗ (n). The index is ordered so that a∗ (n)/a(n) is a decreasing function of n.
In other words, the goods are indexed so that the home country has a comparative
advantage in the production of goods with smaller n.
Then the location of the marginal industry N in the home country can be seen
from
w a∗ (N)
∗
= (0 < N < 1) (14.2)
w a(N)
where w and w∗ are, respectively, the rate of wage in home and foreign countries.
Since the right-hand side of (14.2) is decreasing with respect to N, the relation (14.2)
can be shown by a downward sloping curve B in Fig. 14.1, where w/w∗ is measured
vertically, and n and N, horizontally. If n < N, a∗ (n)/a(n) > w/w∗ so that the n-th
industry is exporting industry, while if n > N, a∗ (n)/a(n) < w/w∗ so that the good
n is imported.
In addition to (14.2), we need the demand side condition too, so as to determine N.
Suppose the given labor population of home and foreign countries are, respectively,
L and L∗ . If we assume that consumers demand functions are such that the
expenditure share of good n in the total expenditure is given and unchanged,
irrespective to its price,3 the expenditure on imports is an increasing function of
3 For this assumption, see Problems 2.1 and 5.1.

102 14 Export Promotion and Welfare
the number of goods to be imported. Then, the average propensity to import, i.e.,
the ratio of the expenditure on imports to GNP, of the home and foreign countries,
respectively, m(N) and m∗ (N), are such that m is decreasing and m∗ is increasing
with respect to N. Then, the balance of trade requires
m(N)wL = m∗ (N)w∗ L∗ (14.3)
so that
w m∗ (N) L∗
= · (14.4)
w∗ m(N) L
which is increasing with respect to N. It can be shown by an upward sloping curve

A in Fig. 14.1.
From (14.2) and (14.4), or by the intersection E of curves A and B in Fig. 14.1,
we can determine N, the range of the exportables [0, N] and that of the importables
[N, 1] of the home country. Since we have from (14.4)
wL m∗ (N)
∗ ∗
= (14.5)
w L m(N)
the relative GNP of the home country to the foreign country is increasing with
respect to N. The home country’s welfare level rises with the relative GNP since
a higher relative income allows home consumers to purchase larger amounts of
foreign goods.
If export subsidies are introduced by the government of the home country, curves
A and B are shifted in Fig. 14.1 and the location of N must be changed. Suppose a
uniform rate ad valorem export subsidy is given to all the goods n < N, i.e., on all
the goods currently exported. The curve B remains unchanged for n ≥ N, but for
n < N it is turned clockwise around E in Fig. 14.1. This is because the effect of
such export subsidies is similar to those of the reduction in a(n) for n < N. The
curve A is uniformly shifted upward, since the relative GNP of the home country is
now smaller than the relative wage rate by the cost to finance export subsidies. The
intersection of the modified curves A and B is located to the left of the original N.
In other words, the range of the exportables of the home country is shrunk. With
unchanged labor population, this expands the export volume of the goods that would
be exported even under free trade and worsens the terms of trade and decreases the
welfare of the home country.
To improve the welfare of the country, therefore, export subsidies should not be
given to goods which can be exported even under free trade. It should rather be
limited to such goods n as n > N in Fig. 14.1. The curve B is shifted upwardly for
n > N and the new intersection with the curve A, also shifted upwardly, is shifted to
the right of N. The range of the exportables of the home country is expanded. With
unchanged labor population, then, this decreases the export volume of the goods
n < N that would be exported even under free trade and improves the terms of trade
and increases the welfare of the home country.
Bibliography 103
The lesson from this problem is clear. It is dangerous to consider the problem of
international trade always by the use of the standard two-country two-good model.
In such a model, one good is called an export good and the other an import good, and
the role played by each good is not affected by export subsidies at all. As Itoh and
Kiyono emphasized, however, what is important is the fact that export subsidies can
increase the national welfare only by expanding the set of the exportables, i.e., by
turning the importables into the exportables.
14.1 Problems
14.1. Discuss what policies and institutions were effective to increase exports in
the post-war Japan.
14.2. By taking costs of transportation into consideration, demonstrate the possi-
bility of the non-traded goods through the necessary modification of Fig. 14.1. See
Dornbusch et al. (1977) for the details.
Bibliography
Dornbusch, R., Fischer, S., & Samuelson, P. A. (1977). Comparative advantage, trade, and
823–839.
Itoh, M., & Kiyono, K. (1987). Welfare-enhancing export subsidies. Journal of Political Economy,
95, 115–137.
Lerner, A. P. (1936). The symmetry between import and export taxes. Economica, 3, 306–313.
Chapter 15
Oligopoly
Traditionally, the perfect competition has been assumed in the theory of international
trade. The theory of perfect competition presupposes that the market is so large that
no single suppliers, by itself, can affect the market price. In other words, the number
of suppliers is very large and they take or accept the market price to decide the plan
of their supplies. Though the world markets are generally large enough to permit
this assumption, however, we have to admit that some markets are dominated by a
few large firms which can manipulate the market price by themselves. The behavior
of such price making firms is considered in the theory of oligopoly.
The theory of oligopoly was first developed by A. A. Cournot, a French
mathematician, in 1838. He is a forerunner of the marginal revolution in economics
and his theory is still a basic theory of the modern theory of oligopoly. He started
with the theory of monopoly (the case where a single firm dominates the market)
and then proceed to the duopoly (the case where two firms dominate the market).
The theory of duopoly is then generalized to the theory of oligopoly. Finally,
Cournot demonstrated that the case of perfect competition can be considered as
a limiting case where the number of firms is infinitely large (see Cournot 1897).
This study of Cournot was the first essential application of the differential calculus
in economics. In the below, however, we try to consider the simplified cases without
using the differential calculus.
Let us start with the case of monopoly. The inverse market demand function is
p = a − bq (15.1)
where p, q, a, and b are, respectively, the price, the quantity demanded (to be
supplied) of a commodity, and positive constants. The unit cost of production c
is simply a given positive constant. The monopolist maximizes the profit, pq − cq.
In view of (15.1), then, we have
pq − cq = (a − c − bq)q
a − c 2 (a − c)2
= −b q − + (15.2)
2b 4b

106 15 Oligopoly
so that the monopolist should supply q = (a − c)2 b.1 Then the maximized profit
is (a − c)2 /4b. In other words, the monopolist should charge such price as p =
(a + c)/2, in view of (15.1).
In the case of duopoly, a firm, say, the first firm, must share the market demand
with the other firm, say, the second firm. The first firm faces the inverse demand
function
p = a − b(q + x) (15.3)
where q is its own supply while x is the supply of the second firm. The first firm
maximizes the profit pq − cq, which is, in view of (15.3)
pq − cq = (a − c − bq − bx)q
a − c − bx 2 (a − c − bx)2
= −b q − + (15.4)
2b 4b
so that the first firm should supply q = (a − c − bx)/2b, if it conjectures that x

remains unchanged. This is Cournot’s basic assumption.
The second firm behaves similarly. Being in the same conditions as the first (i.e.,
a, b, and c are identical), it is reasonable to suppose that q = x in the duopoly
equilibrium. Each firm supplies, then, q = x = (a − c)/3b so that the total supply is
2(a − c)/3b. The duopoly price is, in view of (15.3), p = (a + c)/3, which is lower
than the monopoly price, (a + c)/2. This is the Cournot’s solution of the duopoly
problem.2
In Fig. 15.1, the price and cost are measured vertically, and the quantity,
horizontally. The market demand curve is AB and the unit cost of production is C.
The monopoly price is Pm and the duopoly price is Pd . In the case of monopoly, the
consumers’ surplus is ADPm and the monopolist’s profit is CEDPm . In the case of
duopoly, consumers’ surplus is increased to AFPd while duopolists’ joint profit is
CGFPd . The welfare indicated by the sum of the consumers’ surplus and profit is
clearly larger in the case of duopoly than in the case of monopoly. If we increase
the number of firms beyond 2 and continue to solve the problem of oligopoly with
larger and larger number of firms, using Cournot’s method shown in the above
case of duopoly, the equilibrium price p is continuously reduced and approaches
to the unit cost of production C. The equilibrium of the perfect competition is, of
course, established at p = C, where the welfare is maximized.
While Cournot’s problem was the effect of increasing the number of firms
in a given oligopoly market, our interest is slightly different in our studies of
the internationalization or the globalization of the economy. The problem to be
1 Since a signifies the highest price obtainable (when q is infinitesimal), we can assume a > c to
make the problem meaningful.

2 See Shubik (1987) for the significance of this solution of Cournot’s in the light of the modern
theory of games.
15 Oligopoly 107
Fig. 15.1 Monopoly and

duopoly
considered is the effect of simultaneous increase of the number of firms and of

the size of the market. In the most simplest case, it is not the comparison of the
monopoly and the duopoly in the same market, but the comparison of the monopoly
in the given market and the duopoly in the market twice enlarged.
Suppose the size of the market of a certain good in home country and that in
the foreign country are identical, with the inverse demand function (15.1), where
p, q, a, and b are, respectively, the price, the quantity demanded (to be supplied),
and positive constants. There exists only one firm in each market, having the
identical unit cost of production c. If the international trade, i.e., export and import,
is prohibited between two countries, each market is monopolized by the single
domestic firm. The resulted price is p = (a + c)/2 as was the case of Cournot’s
monopoly discussed in the above.
If the trade is liberated between two countries and the cost of transportation can
be ignored, two domestic markets are fused into a single large international market
with the inverse demand function
b
p = a − q. (15.5)
2
This is because the new market is twice as large as each old one, having demand
twice larger, i.e., q = 2(a − p)/b in (15.5) while q = (a − p)/b in (15.1). Now there
are two firms in this market, the home firm and the foreign firm. If we follow the
duopoly theory of Cournot, the price is now p = (a + 2c)/3. Since a > c,
a + 2c a + c
< (15.6)
3 2
so that the price is reduced in each country as a result of trade liberation. The welfare
in each country is increased, as is seen in Fig. 15.1.
108 15 Oligopoly
These gains in welfare cannot, however, be called gains from trade. Even
though in the same single market, still home consumers can be supplied solely
by the home firm and foreign consumers, solely by the foreign firm. There is
no need of export and import between countries. These gains are different from
the gains from international trade based on the comparative advantages between
countries differently conditioned (e.g., in taste, technology, and factor endowments,
etc.). They are gains from the intensification of competition among firms caused
by the market unification between countries similarly conditioned (e.g., in taste,
technology, factor endowments, etc.). What is important here is not the actual trade
but the possibility of free trade itself.
15.1 Problems
15.1. By the use of the differential calculus, solve the problem of monopoly, when
the market demand function is given as (15.1) in the text and the unit cost of
production is given constant c.
15.2. By the use of the differential calculus, solve the problem of duopoly, when the
market demand function is given as (15.1) in the text and the unit cost of production
is given constant c for both firms, following Cournot’s method.
15.3. By the use of the differential calculus, solve the problem of oligopoly of n
firms, when the market demand function is given as (15.1) in the text and the unit
cost of production is given constant c for all the firms, following Cournot’s method.
Show that the equilibrium price p approaches to c, as the number of firms n is
increased infinitely.
15.2 Appendix: Tariffs Versus Quotas
To protect domestic industries, for example, it is often proposed to impose tariffs

on import or to set, more directly, import quotas. In the world of the perfect com-
petition, the equivalence between tariffs and quotas has been widely recognized.3
In the world of oligopolies, however, we have to recognize the non-equivalence of
effects of tariffs and quotas on the price level in the domestic market. Following the
pioneering contribution made by Bhagwati (1965), the problem was studied by two
papers of Japanese economists, i.e., Shibata (1968) and Itoh and Ono (1982).
Particularly interesting is the non-equivalence in the case where the behavior
of domestic and foreign firms are not symmetric. As is explained in Chap. 15,
the behavior of duopoly firms is considered to be symmetric in Cournot’s theory.
3 See, however, Kreuger (1974) for the problem of rent-seeking in the case of quotas.
15.2 Appendix: Tariffs Versus Quotas 109
Each firm assumes that the unchanged supply of the other and tries to maximize
own profit. In the history of the theory of duopoly, Stackelberg (1934) called such
behavior of firms followership. If a firm acts as a follower, however, the other firm
can make a larger profit by taking advantage of it. Stackelberg called such active
behavior of a firm leadership. While Cournot’s duopoly equilibrium is symmetric
follower–follower equilibrium, also an asymmetric leader–follower equilibrium is
certainly a possible alternative.4
Let us begin by restating Cournot’s theory of duopoly, unlike in Chap. 15, by
using the differential calculus. It is convenient to consider the inverse demand
function p = f (x1 + x2 ), where x1 and x2 are, respectively, quantities to be supplied
by the firm 1 and the firm 2, and p signifies the market price. The profit of the first
firm is
f (x1 + x2 )x1 − g(x1 ) (15.7)
where g(x1 ) is the total cost of production. Similarly, the profit of the second firm is
f (x1 + x2 )x2 − g(x2). (15.8)
Consider that each firm will independently maximize its profit, assuming that
the supply of the other is unchanged. The conditions for it are obtained by the
differentiation of (15.7) and (15.8), respectively, with respect to x1 and x2 ,
f (x1 + x2 ) + x1 f (x1 + x2 ) − g (x1 ) = 0 (15.9)
and
f (x1 + x2 ) + x2 f (x1 + x2 ) − g (x2 ) = 0 (15.10)
where f and g denote the derivatives of f and g. Since two firms are enjoying
entirely identical conditions, we should have x1 = x2 at equilibrium. By adding
(15.9) and (15.10) together, therefore, we have
x
2 f (x) + x f (x) − 2g =0 (15.11)
2
where x = x1 + x2 . By solving (15.11), we obtain the equilibrium price p = f (x).

The basic assumption of Cournot’s theory of duopoly is that each firm changes
its supply assuming that the supply of the other firm is unchanged. In other words,
each firm adjusts its supply to the given supply of the other. Stackelberg called such
a firm a follower. If the first firm is known to act as a follower, however, the second
firm need not act also as a follower. It can make its profit larger, by taking advantage
4 For the leader–follower problem, see Fellner (1965, pp. 71–72 and 98–119).
110 15 Oligopoly
of the fact that the first firm is a follower. If the first firm chooses x1 in accordance
with (15.9) when x2 is given, the second firm can make a conjecture on the behavior
of the first firm that
dx1 f + x1 f
= (15.12)
dx2 g − x1 f − 2 f
from the differentiation of (15.9) with respect to x2 , where f and g are the second
derivatives of f and g. Then the condition on the second firm’s supply to maximize
its profit is obtained by the substitution of (15.12) into
dx1
f (x1 + x2 ) + x2 f (x1 + x2 ) + x2 f (x1 + x2 ) − g (x2 ) = 0 (15.13)
dx2
which is obtained by the differentiation of (15.8) with respect to x2 . This is the

behavior of a firm Stackelberg called a leader. The conditions of Stackelberg
leader–follower equilibrium are given by (15.9) and (15.13).
As in the below, Stackelberg’s concept of leader and follower can be used in a
slightly different sense. Firms are called followers, if they take the price as given.
If they take the market price as given, they are perfectly competitive firms. The firm
is the follower in the duopoly, if it takes the price changed by the leader firm as
given. The leader in the duopoly can set the price freely, taking advantage of the
fact the other firm behaves as a follower.
Let us now consider a domestic market dominated by two firms, a home country
producer and a foreign country exporter. The demand function of the domestic
consumers is given as p = f (x + y) in the inverse form, where p, x, and y denote,
respectively, the price in the domestic market, the supply from the home producer,
and the supply from the foreign exporter (i.e., the import of the home country). The
home firm’s total cost function is g(x), while that of the foreign firm is h(y), which
includes import tariff charged by the government of the home country.
If both firms are followers and take p as given to decide on their supply, as in the
case of the perfect competition, the conditions
p = g (x), (15.14)
p = h (y) (15.15)
and
p = f (x + y) (15.16)
determine the equilibrium p, x, and y, where g and h are the derivatives of g and h
(i.e., marginal costs). If import quota y∗ , instead of import tariff, is imposed so that
y = y∗ , the equilibrium conditions are (15.14) and
p = f (x + y∗ ). (15.17)
The equilibrium price p remains unchanged, however, if the quota y∗ is set equal to
the equilibrium quantity y determined by the tariff equilibrium conditions (15.14),
(15.15), and (15.16). This is the equivalence of tariff and quota in the case of the
perfect competition.
How about if the foreign exporter is the leader while the home producer is the
follower? To make computations simpler, let us linearize the demand and marginal
cost functions as
p = f (x + y) = A − B(x + y), (15.18)

g (x) = a + bx (15.19)
and
h (y) = c + dy (15.20)
where A, B, a, b, c, and d are all positive constants.

When the foreign firm’s p is given, the home firm decides its supply x as
p = a + bx (15.21)
i.e., x = (p − a)/b. By substitution of this into (15.18), then, the foreign firm faces
the demand function
Ab + Ba − Bby
p= (15.22)
b+B
and its revenue is
y(Ab + Ba − Bby)
py = . (15.23)
b+B
The condition for the maximization of its profit (i.e., the marginal revenue = the
marginal cost) is
Ab + Ba − 2Bby
= c + dy (15.24)
b+B
from which the equilibrium supply is obtained as
Ab + Ba − Bc − bc
y= . (15.25)
Bd + bd + 2Bd
The equilibrium price is obtained by the substitution of (15.25) into (15.22).

Suppose now tariff is replaced by quota y∗ which is equal to the equilibrium y in
the case of tariff, i.e., (15.25). Then the foreign firm faces the demand curve (15.22)
for y ≤ y∗ . Since its cost is reduced by the abolition of tariff, the marginal revenue is
112 15 Oligopoly
larger than the marginal cost at y = y∗ . In other words, the profit of the foreign
firm reaches at the corner-maximum at y = y∗ , since y > y∗ is impossible. The
equilibrium supply of the foreign firm is unchanged and therefore the equilibrium
price in the domestic market remains unchanged. Provided that the home firm is the
follower, therefore, we still have the equivalence of tariff and quota even in the case
of a leader–follower equilibrium in the duopoly.
Finally, let us consider the case where the foreign exporter is a follower and
the home producer is the leader. We still assume linear demand and marginal cost
functions, (15.18), (15.19), and (15.20).
When the home firm’s p is given, the foreign firm decides its supply y as
p = c + dy (15.26)
i.e., y = (p − c)/d. By substitution of this into (15.18), then, the home firm faces the
demand function
Ad + Bc − Bdx
p= (15.27)
d +B
and its revenue is
x(Ad + Bc − Bdx)
px = . (15.28)
d +B
marginal cost) is
Ad + Bc − 2Bdx
= a + bx (15.29)
d+B
Ad + Bc − Ba − ad
x= . (15.30)
Bb + bd + 2Bd
The equilibrium price is obtained by the substitution of (15.30) into (15.27).

Suppose now tariff is replaced by quota y∗ which is equal to the equilibrium y in
the case of tariff. Then the home firm faces the demand function
p = A − B(x + y∗) (15.31)
and its revenue is
px = Ax − B(x + y∗)x. (15.32)

marginal cost) is
A − 2Bx − By∗ = a + bx (15.33)
A − By∗ − a
x= . (15.34)
2B + b
Even without solving y∗ explicitly, we can see that x in (15.30) is larger than x in
(15.34), if we suppose that a < c and b = 0.5 Since the supply of the foreign firm is
inelastic in the case of quota, the home firm can raise the price by the reduction of
its supply from its level in the case of tariff. In other words, the equilibrium price in
the domestic market is higher in the case of quota than in the case of tariff. Tariff and
quota are not equivalent if the home producer is the leader in the leader–follower
equilibrium of duopoly.
Now let us return to the case of the foreign leader considered above. At the
equilibrium reached by the foreign leadership, (p∗ , x∗ , y∗ ),
p∗ = A − Bx∗ − By∗ = a + bx∗ (15.35)
in the case of tariff as well as in the case of quota. Suppose that the role of leader
is changed from the foreign firm to the home firm. If the import is restricted by the
quota equal to y∗ , the home firm faces the demand function
p = A − B(x + y∗) (15.36)
so that its marginal revenue is MR = A − 2Bx − By∗. In view of (15.35), then,
A − 2Bx∗ − By∗ < a + bx∗ (15.37)
so that the marginal revenue is smaller than the marginal cost for the home firm
at the equilibrium reached by the foreign leadership. As the new leader, then, the
home firm raises the price p by the reduction of its supply x, so as to increase its
profit. At the new equilibrium reached by the home leadership, the price p is higher
than p∗ . Not only the profit of the home firm but also the profit of the foreign firm
are increased, since the supply of the latter firm is unchanged at y∗ .
Thus quotas make it profitable for the home producer to be a price leader and
for the foreign supplier to be a price follower. If two firms are rational, Itoh and
Ono (1982) consider, the Stackelberg duopoly equilibrium with the foreign firm’s
5 Remember that the tariff is included in h(y). See Bhagwati (1969, pp. 248–265), for the more
general considerations.
114 15 Oligopoly
leadership will never be chosen under an import quota.6 In other words, we have
the general non-equivalence of tariff and quota that the price in the domestic market
is always higher under the quota than under the tariff in the price leader–follower
duopoly of the home and the foreign firms.
15.3 Problems for Appendix
15.4. Demonstrate geometrically, i.e., by the use of the figure, that the tariff and the
quota are equivalent if the home firm is the follower. See Itoh and Ono (1982).
15.5. Demonstrate geometrically, i.e., by the use of the figure, that the tariff and the
quota are not equivalent if the home firm is the leader. See Itoh and Ono (1982).
Bibliography
Bhagwati, J. N. (1965). On the equivalence of tariffs and quotas. Baldwin, R. E., et al. (Eds.),
Trade, growth and the balance of payments. Chicago: Rand Nally.
Bhagwati, J. N. (1969). Trade, tariffs and growth. London: Weidenfeld and Nicolson.
Cournot, A. A. (1897). Researches into the mathematical principles of the theory of wealth (N. T.
Bacon, Trans.). London: Macmillan.
Fellner, W. (1965). Competition among the few. New York: Kelley.
Itoh, M., & Ono, Y. (1982). Tariffs, quotas, and market structure. Quarterly Journal of Economics,
295–305.
Kreuger, A. (1974). The political economy of rent-seeking society. American Economic Review,
64, 291–303.
Ono, Y. (1978). The equilibrium of duopoly in a market of homogeneous goods. Economica, New
Series XLV, 287–295.
Shibata, H. (1968). A note on the equivalence of tariffs and quotas. American Economic Review,
LVIII, 137–142.
Shubik, M. (1987). Cournot, Antoine Augustin. In J. Eatwell, M. Milgate, & P. Newman (Eds.),
The new palgrave, vol. 1 (pp. 708–712). London: Macmillan.
Stackelberg, H. V. (1934). Marktform und gleichgewicht. Wien and Berlin: Springer.
6 Ono (1978) first insisted the importance of this kind of the endogenous determination of the
leader–follower relation.
Chapter 16
Immiserizing Growth
Economic growth is generally to be welcomed from the point of view of the welfare
of a country. It is particularly so when the growth is due to the capital accumulation
rather than the increase in the population. Similarly, international trade is generally
considered to raise the level of welfare of a country, i.e., the gains from trade are
expected. One plus one may not, however, always make two. It is possible that the
combined effect of the economic growth and international trade is to decrease, rather
than to increase, the welfare of a country. This possibility was first pointed out by
Edgeworth (1894) and then taken up again by Bhagwati (1958) who called such an
economic growth the immiserizing growth.
The gains from trade for a country may be lost as a result of her growth. The
reason for this immiserizing growth is the deterioration of her terms of trade (i.e.,
relative price of the exportables and the importables) caused by the growth of a
country, which increases her export supplies and import demands in the world
market. As Bhagwati (1987) pointed out, the possibility of immiserizing growth is
due to the fact that the free trade is not the best, optimal, trade policy for such a large
country which can change the terms of trade through the changes in the volumes of
her exports and imports. In this sense, the problem of the immiserizing growth is
a problem of the so-called second best problems (see Sect. 15.2 Appendix to this
chapter).
The possibility of the immiserizing growth may be shown geometrically as in
Fig. 16.1. The production and consumption of the importables of a country are
measured horizontally, and those of the exportables, vertically, from the origin O.
The point A and the point B are, respectively, the point of production and that
of the consumption, before the economic growth takes place. The slope of the
line AB indicates the terms of trade, the relative price of the exportables and the
importables. The transformation curve is tangential to AB at A from the below and a
social indifference curve, at B from the above, though they are not drawn in to
avoid cluttering up the figure. The vertical distance between A and B indicates the
amount of the export, and the horizontal distance, that of the import. Similarly,
the point C and the point D are, respectively, the point of production and that of
the consumption, after the economic growth took place. The slope of the line CD

116 16 Immiserizing Growth
Fig. 16.1 Immiserizing .

growth
indicates the new terms of trade, which is deteriorated as the result of the increase
in the export of this country in the world market. The level of the welfare, indicated
by the consumption points B and D, is clearly reduced as a result of the economic
growth, since we can see that the consumption of both commodities is decreased
from the point B to the point D.
To see the possibility of the immiserizing growth analytically, let us consider
a drastically simplified two-commodity model of a country. Let us assume that the
country can produce only one commodity, i.e., her exportables, the volume of which
is given exogenously as Y . Out of Y , the country exports y, so that her consumers
can have Y − y of this commodity. Let us denote by x the volume of the import of
another commodity, the importables of this country. The condition of the balance of
trade requires
py = x (16.1)
where p denotes the terms of trade, i.e., the relative price of the exportables in terms
of the importables. The inverse demand function for the exportables in the world
market is
p = A − By (16.2)
where A and B are given positive constants. Finally, let us suppose that the
consumers of this country choose the consumption of two commodities so that
x =Y −y (16.3)
is satisfied. In other words, they have identical indifference map in which indiffer-
ence curves are L shaped with their corners on the positively sloped 45◦ line.
We can solve (16.1)–(16.3) for x, y, and p, if A, B, and Y are given. Suppose that
initially A = 2, B = 3, and Y = 1/2. Then, we have
16 Immiserizing Growth 117
√ √
3 3− 3
x= , y=
6 6
in view of y < Y . If, as the result of the economic growth of this country, Y is
increased from Y = 1/2 to Y = 3/4, we have
1 1
x= , y= .
4 2
√
Since the level of consumption is decreased from x = Y − y = 3/6, which is
approximately equal to 0.29, to x = Y − y = 1/4 = 0.25, we clearly have a case
of the immiserizing growth.
If this country is a small country so that the terms of trade is given and unchanged,
i.e., B = 0, we can solve (16.1)–(16.3) as
Y AY
x= , y=
1+A 1+A
so that
Y
Y −y = .
1+A
Since both x and (Y − y) increase as Y increases, there is no possibility for the

immiserizing growth for a small country. The reason for the immiserizing growth
is the deterioration of the terms of trade p when Y is increased as the result of the
economic growth.
It is well known that the free trade is the best policy for a small country to
which the terms of trade is given and unchanged. For example, see Fig. 16.2, where
the domestic demand for the importables is measured horizontally, and its price,
vertically. The curve DD is the domestic demand curve for the commodity, while it
is assumed that there exists no domestic industry which can supply this commodity.
The world price is simply given as OB, since the country is assumed to be a small
country. In other words, this country can import the commodity as much as she
demands at this price. The free trade equilibrium is A, where AB of the commodity is
imported. The domestic consumers’ surplus which indicates the level of the welfare
is given as DAB. Now suppose the government of this country imposes an import
tariff EB. The equilibrium point is shifted to C, where only EC of the commodity is
imported. The level of welfare is reduced, since the sum of the consumers’ surplus
DCE and the revenue of tariff ECFB is smaller than DAB. Likewise any other
conceivable interference to the free trade can be shown to reduce the level of the
welfare of a small country.
For a country facing the world markets, the conditions for the optimal orga-
nization of the economy consist of, firstly, the optimal condition concerning the
international trade, and secondly, the optimal condition concerning to the domestic
production. For the case of a small country, they are, firstly, the free trade, and
Fig. 16.2 Free trade for a small country
secondly, the maximization of the value of the domestic production. It is no wonder,

then, the welfare is increased by an increase in the value of the domestic production
(Y in the above analytical model), if the free trade is maintained. With the free trade,
therefore, there is no possibility of the immiserizing growth for a small country.1
If the country is large, however, the free trade is not the optimal condition for
the international trade. For example, see Fig. 16.3, where the domestic demand for
the importables is measured horizontally from the origin O to the right, the import,
i.e., the export of the foreign exporter, to the left, and the price of the commodity,
vertically. The curve DD is the domestic demand curve for the commodity, while it
is assumed that there exists no domestic industry which can supply this commodity.
The curve FF is the supply curve of the foreign exporter. The free trade equilibrium
is that the world price is OA, at which domestic demand AB is equal to the supply
from the foreign exporter AC. The consumers’ surplus is DBA. If the government
imposes import tariff CH, the domestic price is changed to OC, the price in the world
market is to OH, so that the domestic demand CE is equalized to the foreign export
GH. The level of welfare is raised, since the sum of the consumers’ surplus DCE
and the revenue of tariff CEIH is larger than the free trade consumers’ surplus DBA.
When the optimal condition for the international trade is not satisfied, as is shown
in the case of the free trade for a large country, the optimal condition for the domestic
production need not be the maximization of the value of the domestic production. In
our numerical example of the immiserizing growth, given in the above, the welfare
of the country is reduced, not raised, by the increase in the value of the domestic
production, since the value of x is reduced from 0.29 to 0.25 by the increase in the
1 See Johnson (1967), however, for the possibility of the immiserizing growth for a small country,
when she has a distortionary tariff in place.

16.2 Appendix: Second Best Problems 119
Fig. 16.3 Free trade for a large country
value of Y from 1/2 to 3/4. If some of the conditions for the optimal organization of
the economy are not satisfied, the realization of the rest of such optimal conditions
is no longer desirable. This is the so-called second best problem.
16.1 Problems
16.1. In Fig. 16.2, introduce the supply curve of the domestic industry and show
that the free trade is optimal for a small country. Similarly, discuss the case of the
exportables.
16.2. Using a figure similar to Fig. 16.3, demonstrate that the free trade is not
optimal for the case of the exportables for a large country.
16.2 Appendix: Second Best Problems
Let us imagine a problem of the optimization whose degree of freedom is more

than one, i.e., the number of variables exceeds that of constraints by at least two.
You have as many optimum conditions as the degree of freedom. Suppose it is
impossible to realize some of the optimum conditions. Then, does a change which
brings about the realization of the rest of the optimum conditions make things
better? A naive man guided by common sense might answer in the affirmative.
However, the correct answer is that it may make things better or worse, depending
on situations. This is the problem of the second best and second best theories try
to solve such questions as: in what situations the realization of a certain optimum
condition is still desirable and how a certain optimum conditions should be modified
in a certain situation, even if it does not lead to a maximum, when other optimum
conditions are not realized (see Lipsey and Lancaster 1956–7). In the below, two
of such problems are discussed for a country facing the international market (see
Negishi 1972, pp. 170–177).
Kemp (1964) argued that the foreign investment tends to be excessive from the
point of view of the investing country. The simplest model used by him runs as
follows (Kemp 1964, pp. 198–200). Let the amount of capital accumulated in the
investing country and the invested country be constant and, respectively, denoted
by k and k∗ . Assuming that each country which competes in the world market is
producing one and the same commodity, let us denote the production functions of
the investing and invested countries, respectively, by f and f ∗ . Since labor is
assumed to be immobile between countries, the amount of labor is assumed to be
constant and also the full employment is assumed, each country’s output is given by
f (k − z) and f ∗ (k∗ + z), where z is the amount of capital movement.
Under laissez faire, z is determined by the condition that the rate of return
to capital is equal in two countries, i.e., f (k − z) = f ∗ (k∗ + z). On the other
hand, the optimal amount of the foreign investment z from the point of view
of the investing country is determined by maximizing her national income, i.e.,
f (k − z) + z f ∗ (k∗ + z), the sum of the domestic output and the returns on the
foreign investment. The condition for the maximization of the national income is
f (k − z) = f ∗ (k∗ + z) + z f ∗ (k∗ + z). This implies that f (k − z) < f ∗ (k∗ + z),
since f ∗ < 0, i.e., the marginal productivity of capital is decreasing. The laissez
faire z is larger than the optimal z, since in the former case f = f ∗ . It is necessary
to restrict the amount of the foreign investment by imposing a tax on the earnings
of the capital invested in the foreign country. The rate of tax t is given from the
condition f = (1 − t) f ∗ , i.e., t = −z f ∗ / f ∗ .
In the above analysis, output and the factors of production are regarded as
distinct items and the amount of factors of production is simply assumed to be
unchanged. This implies either that the capital does not depreciate and only the
consumers’ goods are produced or that the analysis is that of the short-run or
temporary equilibrium in the sense that the capital goods produced are not to
be utilized in production within the period under consideration. As the result of
these assumptions, in such a static or timeless model, new investment abroad,
i.e., the changes in the amount of capital invested abroad are considered as the
movement of capital stock. In other words, the capital stock already constructed and
in use is assumed to be mobile between countries. “It is rare, however, that capital
instruments wander about” (Ohlin 1933, p. 404). Being simply an export surplus,
new investment abroad is more appropriately considered as the accumulation abroad
of the flow of the newly produced capital goods or the acquisition of the capital stock
already constructed and in use in the foreign country. We have to consider, therefore,
the optimal foreign investment in the long-run analysis of the dynamic process of
the production and the accumulation of capital goods.
For the sake of simplicity, let us confine ourselves to optimal foreign investment
in terms of long-run stationary equilibrium or balanced growth equilibrium. The
point is that the change in the amount of foreign investment has an effect on the
level of national income at the long-run stationary equilibrium or on the balanced
growth path through changes in the capital accumulation (investment and saving)
caused by changes in national income and its distribution of the investing and
invested countries. Suppose each country owns, respectively, k and k∗ amount of
capital stocks and the amount of the foreign owned capital in the invested country
is z. Under the assumption of full employment of the constant labor force, two
countries produce, respectively, f (k − z) and f ∗ (k∗ + z) amount of one and the
same commodity which may be either consumed or invested as the capital. There
is, therefore, no movement of commodities between countries except for the new
investment abroad and the interest payments on the capital stock invested in the
foreign country. Let a constant g such that 0 < g < 1 denote the uniform rate of
depreciation of the capital in two countries. Then the long-run level of the net
income of the investing country, which is to be maximized, is
f (k − z) + z f ∗ (k∗ + z) − gk (16.4)
i.e., domestic output plus interest earning from abroad less depreciation. Let us
assume that proportions s and s∗ of the capital earnings of the investing and invested
countries are saved, while all the wage incomes in both countries are consumed.
Then, the condition for the stationary state, i.e., of zero net investment, will be
s(k − z) f + sz f ∗ = gk (16.5)
for the investing country and
s∗ k∗ f ∗ = gk∗ (16.6)
for the invested country. Our problem is to choose z which maximizes (16.4) under
the restrictions of (16.5) and (16.6). Let us note that k and k∗ are not constant now,
but variables to be determined by (16.5) and (16.6) when z is given. By changing
the amount of the capital stock to be kept invested abroad, we are interested in
the maximization of, not the instantaneous net income (as in Kemp 1964), but the
long-run maintained level of net income, just as in the theory of the optimal saving
we are interested in maximizing, not instantaneous consumption, but the long-run
maintained level of consumption.
From (16.6) above, the marginal productivity of capital in the invested country is
constant, i.e.,
g
f ∗ = (16.7)
s∗
unless k∗ = 0. Assuming nonzero k∗ , we substitute (16.7) into (16.4) and (16.5).

If the national income in the investing country is constant, the marginal rate of
substitution between k and z in (16.4) is
g
dk 1 f −
∗
= s . (16.8)
dz f −g
On the other hand, from (16.5), the marginal rate of k and z is given as
sg
dk 2 s f + s(k − z) f −
= s∗ . (16.9)
dz s f + (k − z) f − g
If there exists an optimal solution with positive k, k∗ , and z, these two marginal rates
of substitution must be equal each other.2 Therefore, we have
dk 1 dk 2 s(k − z) f
= = . (16.10)
dz dz s(k − z) f − g + sg
From the usual assumption that f < 0 and s < 1, it is clear that 0 < dk/dz < 1.
Applying this result to (16.8), we see that f > g and f > f ∗ = g/s∗ at the
optimum. Since the private motive for foreign investment is satisfied when f = f ∗ ,
the amount of foreign investment tends to be too small under laissez faire. Then,
a subsidy rather than a tax on the foreign investment is needed from the long-run
point of view of the investing country.
The optimal rate of subsidy on foreign investment is calculated from (16.8)
and (16.10) as t = s(k − z)(s∗ − 1) f /(g − sg). The rationale of this result, which
seems rather paradoxical from the short-run point of view, may be that, if capital
accumulation is heavily dependent on capital earnings, it is wise for a country
to encourage capital export so as to keep home capital earnings and, therefore,
capital accumulation and the long-run level of consumption high. Another reason
may be that larger instead of diminishing returns can be expected by increasing the
foreign investment, since it will check the growth of domestic capital in the invested
country.3
Different results from the static or short-run analysis and the dynamic or long-
run analysis can be reconciled from the point of view of the second best theory. The
implications of the assumption that saving is equal to 100 % of capital earning is
that there is no room for the policy which directly aims to interfere with the amount
of saving and therefore of capital accumulation. What we can do is to change the
amount of saving indirectly through the changes in capital earnings by controlling
2 This excludes the possibility that s = s∗ .

3 Inthe case of the balanced growth, we have only to replace g by g+ h in the above analysis, where
h is the uniform rate of growth of the labor force in two countries.
the amount of foreign investment. Suppose now we can directly control the amount
of saving in each country and therefore k and k∗ are free from (16.5) and (16.6).
Then we can consider that k∗ is determined by the policy of the invested country.
With k∗ given, the investing country maximizes her national income (16.4) with
respect to k and z. Conditions for the maximization are
f − g = 0, (16.11)
∗ ∗
−f + f +zf = 0. (16.12)
Condition (16.12) is identical to the one obtained in the static analysis, while
condition (16.11) is that of the optimal saving (see, e.g., Phelps 1966, pp. 6–11). The
conclusion from the static or short-run model remains unchanged in the dynamic or
long-run situation, if the amount of foreign saving and that of capital are given and
domestic saving is optimal in the sense that the national income is maximized. Since
f > g in the case of (16.10) above, the condition for optimal saving (16.11) is not
satisfied. Therefore, the policy suggested (a subsidy on foreign investment) is an
example of the second best policy when the direct control of saving is impossible
and the optimal condition for saving is not satisfied.
According to the well-known theory of the optimal tariff developed in the case
of static or short-run analysis of international trade, the free trade (i.e., no tariff) is
optimal for a small country for which international prices of commodities are given
and unchanged. Now let us consider the question whether this remains true also in
the dynamic or long-run analysis, in which the adjustment of capital to the changes
in the rate of tariff is fully taken care of. We are again concerned with the case of
the stationary state or the balanced growth.
Assuming that the first commodity is consumers’ goods and the second, capital
goods in a two-commodity two-factor model of a small country, conditions for an
incomplete specialization equilibrium are as follows.
E1 + pE2 = 0, (16.13)
gk = X2 + E2, (16.14)
X1 = F(X2 , k), (16.15)
−q = F1 (X2 , k), (16.16)
qgk = s[r(q)k + w(q) + (q − p)E2] (16.17)
where unknowns to be determined are E1 , E2 , k, X1 , X2 , respectively, signifying

the import of the first and second commodities, the amount of the capital stock,
outputs of the first and second commodities, while the international and domestic
prices of the second commodity in terms of the first, p and q, and the labor force,
taken as unity, are given. Condition (16.13) is that of the balance of trade. Condition
(16.14) signifies the stationary state,4 in which g is the given rate of depreciation.
4 Please refer to footnote 3.

The transformation curve between X1 and X2 and its tangency with domestic price
line are represented by (16.15) and (16.16). The equality of saving and investment
is (16.17) under the assumption of a simple Keynesian saving function, where 0 <
s < 1 is the given marginal propensity to save and the rate of interest r and the wage
w (in terms of the first commodity) are assumed to be uniquely determined when q
is given (Kemp 1964, pp. 45–53).
Starting from the initial free trade situation (p = q), and assuming that the
second commodity is imported, we ask a question whether the imposition of a
tariff increases the net national income Y = rk + w + (q − p)E2 − qgk, i.e., whether
dY /dq > 0.5
Let us first consider
dk sr k + sw + sE2 − gk

= (16.18)
dq qg − sr
which is obtained from the differentiation of (16.13)–(16.17). It is easily seen from

(16.17) that qg − sr > 0 under the assumption that p = q. Consider Walras’ law
r(q)k + w(q) + (q − p)E2 = X1 + qX2 + (q − p)E2. (16.19)
The differentiation with respect to q yields r k +w −X2 = X1 +qX2 , where the prime
denotes the differentiation with respect to q under the assumption that k is constant,
i.e., X1 = dX1 /dq and the like. Since X1 + qX2 = 0 (see Kemp 1964, p. 21), and
therefore r k + w − X2 = 0, we see that sr k + sw + sE2 − gk < 0, using (16.14) and
s < 1. Therefore dk/dq < 0. Again using the fact that r k + w − X2 = 0, and (16.14),
the total differentiation of Y with respect to q is reduced to
dY dk
= (r − qg) . (16.20)
dq dq
Therefore, dY /dq > (<) 0 as r − qg < (>) 0.

On the other hand, if we differentiate Y with respect to k, assuming q is constant
(equal to p), we have
dY
= r − qg. (16.21)
dk
Perhaps we may say that the saving and capital accumulation relative to labor force
is more (less) than optimal when r − qg < 0 (> 0). We must, therefore, restrict
(encourage) the import of capital goods by a tariff (subsidy) when the saving is
more (less) than optimal.
5 Ifthe second commodity is exported, the question is whether an export subsidy increases the
national income.
Since r − qg is independent of k, however, the sign of dY /dq is unchanged when

p = q, provided that the economy is incompletely specialized. If the optimization
of saving is considered, and k is increased (decreased) when r − qg > 0 (< 0),
therefore, the output of the industry which is more capital (labor) intensive is
increased and that of the other industry is decreased, until the economy is completely
specialized (i.e., Rybczynski theorem, see Chap. 11).
When the economy is completely specialized in the first industry, we must
replace (16.13)–(16.17) by
gk = E2 , (16.22)
X1 = G1 (k), (16.23)
qgk = sX1 + s(q − p)E2 (16.24)
where (16.22) and (16.24) correspond to (16.14), and (16.17), and (16.23) is the
production function of the first commodity since X2 = 0. Instead of (16.18) we now
have from (16.22) to (16.24),
dk (1 − s)gk
= (16.25)
dq sG1 − qg
where sG1 − qg < 0, since
dk
= sX1 + s(q − p)E2 − qgk (t = time) (16.26)
dt
must be stable when p = q. Therefore, dY /dq = (r − qg)dk/dq > (<) 0, where

r = G1 and Y = X1 + (q − p)E2 − qgk, as (r − qg) < (>) 0, i.e., the import of the
second commodity must be tariffed or subsidized according to the level of the saving
relative to its optimal level. Since r = G1 < 0, we can reach the optimal saving by
further increasing (decreasing) k when r − qg > 0 (< 0).
If, on the other hand, the economy is specialized in the second commodity, we
have to replace (16.14)–(16.17) by
X2 = G2 (k), (16.27)
qgk = sqX2 + s(q − p)E2 (16.28)
where (16.28) corresponds to (16.17), and (16.27) is the production function of the
second commodity, since X1 = 0. Instead of (16.18) we have from (16.15), (16.27),
and (16.28),
dk (1 − s)gk
= (16.29)
dq sqG2 − qg
where the denominator of the right-hand side must be negative from the stability
condition of
dk
= sqX2 + s(q − p)E2 − qgk (t = time, p = q). (16.30)
dt
Therefore, dY /dq = (r − qg)dk/dq > (<) 0, where Y = qX2 − (q − p)E2 − qgk and
r = qG2 , as (r − qg) < (>) 0, i.e., the export of the second commodity must be
tariffed or subsidized as the saving is less or more than the optimal saving.
Summarizing the above arguments, we can now conclude as follows. At free
trade equilibrium, import of capital goods should be restricted (encouraged) by a
tariff (subsidy) if saving in proportion to the national income is more (less) than
the optimal. If the capital goods are exported, we have to have a tariff (subsidy) on
export as the saving is less (more) than the optimal. While here we are concerned
with the second best trade policies when the optimal condition on saving is not
satisfied, Smith (1977) considers the converse second best problems that the usual
rule for optimal saving will not continue to be valid if trade policies are not optimal.
Bibliography
Bhagwati, J. (1958). Immiserizing growth: A geometrical note. Review of Economic Studies, 25,
201–205.
Bhagwati, J. (1987). Immiserizing growth. In J. Eatwell, M. Milgate, & P. Newman (Eds.), The
new palgrave, vol. 2 (pp. 718–720) London: Macmillan.
Edgeworth, F. Y. (1894). The theory of international values, I. Economic Journal, 4, 35–50.
Johnson, H. G. (1967). The possibility of income losses from increased efficiency or factor
accumulation in the presence of tariffs. Economic Journal, 77, 141–144.
Kemp, M. C. (1964). The pure theory of international trade. Englewood Cliff, NJ: Prentice-Hall.
Lipsey, R. G., & Lancaster, K. J. (1956–1957). The general theory of the second best. Review of
Economic Studies, 24, 11–32.
Negishi, T. (1972). General equilibrium theory and international trade. Amsterdam: North
Holland.
Ohlin, B. (1933). Interregional and international trade. Cambridge, MA: Harvard University Press.
Phelps, E. S. (1966). Golden rules of economic growth. New York: Norton.
Smith, M. A. M. (1977). Capital accumulation in the open two-sector economy. Economic Journal,
87, 273–282.
Chapter 17
External Economies
As we saw in Chap. 3, already in 1776 Adam Smith pointed out the importance of
increasing returns to scale for international trade in his The Wealth of Nations. It is
rather recent, however, that the role of increasing returns has begun to be discussed
seriously in the modern theory of international trade. Since the assumption of perfect
competition, which has been widely made, is not consistent with the so-called
internal economies, Marshallian external economies have been considered by, e.g.,
Mathews (1949), Melvin (1969), and Kemp & Negishi (1970).1 While these early
Marshallian approaches focus on the relation of commodity prices and commodity
trade, Ethier (1979) suggests a drastically different one which focuses on prices of
factors of production (see Krugman 1987).
Let us begin with the recapitulation of the elementary demonstration of the gains
from trade under the assumptions of the perfect competition and decreasing returns.
In Fig. 17.1, the quantity of the importables (or the exportables, as the case may
be) is measured horizontally, and its price and cost, vertically. The curve AB is
the domestic demand curve and CD, the domestic supply curve, i.e., the increasing
marginal cost curve which implies the assumption of decreasing returns. In autarky,
the equilibrium is the point E, and the combined surplus of the consumers’ and
producers’ is AEC. If the world price is OH, which is lower than the autarky price
OL, the domestic demand is HF, the domestic supply is HG, and the difference FG
should be imported at the trade equilibrium. The combined surplus is now AFGC.
Since AFGC is larger than AEC, there exist gains from trade. Similarly, if the world
price is OK, which is higher than the autarky price OL, the domestic demand, supply,
and the export are, respectively, KJ, KI, and IJ. There exist gains from trade, since
AJIC is larger than AEC.
1 According to Marshall, economies of scale are divided into external economies, which depend on
the general development of the industry, and internal economies, which depend on the resources
of the individual firms engaged in it (Marshall 1961, Book IV, Chap. IX). The latter is not
consistent with the equilibrium of perfect competition for which the marginal cost of a firm must
be increasing. For internal economies, see Sect. 17.2 Appendix.

128 17 External Economies
Fig. 17.1 Gains from trade
In the traditional Walrasian general equilibrium theory, the marginal cost of

a price taking individual firm is assumed to be increasing with respect to the
level of output, as a necessary condition for its profit maximizing equilibrium.
In Marshallian terminology, the internal economies of scale is assumed away.
We can, however, still introduce economies of scale as the Marshallian external
economies. It implies the effect of the scale of the industry on the cost of firms. If the
scale of the industry is expanded, the marginal cost curve of single firms belonging
to the industry are shifted downwardly. Since the marginal cost curve of the price
taking firms are still assumed to be increasing, the introduction of these economies
of scale which are external to such firms is consistent with the assumption of the
perfect competition in the sense of Walras.2
Figure 17.2 is similar to Fig. 17.1, where the quantity of the exportables are
measured horizontally, and its price, vertically, and curves AB and CD are domestic
demand and supply curves. The latter curve is obtained by the aggregation of the
marginal cost curves of firms which is now dependent on the scale of the industry.
Let us assume that CD is obtained under the assumption that the scale of the
domestic industry is JE. The autarky equilibrium is at E and the combined surplus
is AEC. Suppose initially the price in the world market is OJ so that there exists
no comparative advantage for the country. If, however, the country can succeed to
export, the world price will be reduced below OJ. The scale of the domestic industry
is now expanded by the additional production for the export. Because of the external
2 See Chap. 3, p. 19 for the Walrasian and Sraffian views of the perfect competition. The latter view
is discussed in Appendix.
17 External Economies 129
Fig. 17.2 Gains from export
economies, then, the domestic supply curve3 is shifted to HF. Now the equilibrium
with international trade is that the domestic demand is GI, the domestic production
is GF, and the export is IF, with the price OG. The combined surplus is then AIFH,
which is larger than AEC at the autarky.
Now we must emphasize that the gains from trade in this case are not due to
the comparative advantage which is given exogenously and exists before trade.
They are due to the comparative advantage which did not exist before trade but
is created endogenously with trade. Traditional theories of trade theory, based on
the assumption of the diminishing returns, can explain gains from trade between
countries different in technology, endowment, taste, etc. With the assumption of
increasing returns, however, we may explain the gains from trade between similar
or identical countries. It is a well-known fact, of course, that nowadays most of the
international trade is carried on between large industrial countries which are very
similar each other.
With the importables, however, things are not so simple. Figure 17.3 shows the
case of the importables, the quantity of which is measured horizontally, and the
price and cost, vertically. The domestic demand curve is AB and supply curve is CD
before trade. The autarky equilibrium is at E with the combined surplus AEC. Now
after trade, suppose the country has to import this commodity (you cannot export all
the commodities!). At the world price OJ, the domestic demand is JI. Through the
competition with the import, the scale of the domestic industry may become small
so that external economies are lost with the result that the cost of production is
pushed up. If the domestic supply curve is pushed up to GH, the domestic industry
must disappear. With zero producers’ surplus, the combined surplus is now AIJ,
which may be smaller than the autarky one, AEC, if EIK < KCJ.
3 Tobe more loyal to Marshall, we have to name the curve, not the supply curve but the particular
expenses curve, since it is based on the assumption that the scale of the industry is given (Marshall
1961, p. 811).
Fig. 17.3 Negative gains from import
With this possibility of the negative gains from trade of the import competing
industry, the welfare gains for the country cannot be assured if external economies
exist in all the industries. What we can safely say is that there exist gains from trade
if as a result of trade industries enjoying increasing returns to scale are expanded
and only the industries subject to non-increasing returns are contracted.
Against the above rather inconclusive result, an alternative approach suggested
by Ethier (1979) and Krugman (1987) may shed some new lights on the problem of
the international trade under external economies. The point is to take factor prices
and their international equalization into consideration.
Consider the most simple case that the world consists of two countries, each with
only one factor of production called labor. Two countries are assumed to possess
identical technology to produce two commodities. Let us further assume, for the
sake of simplicity and also to consider the case of similar countries, that the given
labor population is equal in two countries. As for commodities, let us assume that
the commodity A is subject to Marshallian external economies while the commodity
B is produced at constant returns to scale.
In autarky, both countries have the identical per-capita GNP, since the rate
of wage in terms of the commodity B is equal. Before the consideration of the
international trade of commodities, suppose the labor is perfectly mobile between
countries, that is to say, consider the case of the completely integrated world
economy. It is clear that the per-capita GNP of this economy is higher than that
of each autarky economy, since the size of the increasing returns A industry in the
integrated economy is larger than each of those in the autarky economies. Now
the question is whether we can reproduce this efficient allocation in the integrated
economy through the free international trade of commodities when the labor is
immobile between countries.
If the size of the increasing returns industry A in the integrated economy is
such that the number of laborers employed there is smaller than the half of the
17.2 Appendix: Internal Economies 131
world population, we can reproduce the allocation of the integrated economy by

the international trade of commodities. Suppose the first country is specialized in
the production of the commodity B, while the second country is producing both
commodity A and commodity B. Since two countries are producing commodity B,
the rate of wage is equalized between two countries. In the second country, the
labor productivity is higher than in the case of autarky, since the scale of increasing
returns industry A is now larger than in the autarky. The rate of wage is then higher
than in the case of autarky and there exist gains from trade in terms of per-capita
GNP for both countries. We must note that even the first country can enjoy the gains
from trade, although she is importing the commodity A which is subject to external
economies.
Suppose, however, that the size of the increasing returns industry A in the
integrated economy is such that the number of laborers employed there is larger than
the half of total world population. Let us suppose further that the second country
is specialized in the increasing returns industry A, while the first country, in the
constant returns industry B. International equalization of wage is not assured and it
depends on the pattern of the world demand. The second country can still enjoy the
gains from trade, since the scale of her increasing return industry A is larger than
in the case of autarky, and therefore the rate of wage is increased from the autarky
level. For the first country importing the increasing returns commodity A, however,
the rate of wage may be lower than in the autarky, if the world demand pattern
is unfortunate for her product, i.e., the constant returns commodity B. Unless the
international factor price equalization is realized, gains from trade are not assured
for all the countries.
17.1 Problem
17.1. Consider the two-country, three-commodity, and two-factor model, where

two commodities are subject to the constant returns, while the third commodity
can enjoy increasing returns through Marshallian external economies. See Krugman
(1987).
17.2 Appendix: Internal Economies
If we follow the traditional Walrasian view of competition, internal economies in

the sense of Marshall (decreasing cost with respect to the scale of a firm) cannot be
consistent with the competitive equilibrium. The equilibrium size of a price-taking
firm is to be determined by the equality of the given market price and the marginal
cost of the firm, so that the latter should be increasing with respect to the scale of the
firm. From a new Sraffian view of competition, however, we can reconcile internal
economies with the competition, as will be discussed in this appendix.
One of the reasons why we are interested in this new view of the competition
is that we can solve a problem much discussed in 1970s’ Japan, i.e., the so-called
problem of the foreign exchange gains by using a model based on this view of the
competition. In 1970s the Japanese yen was much appreciated, but the consumers’
price of the importables was not much reduced so that it was much discussed who
gained from the appreciation of the currency and why. An answer to this problem
may be given below by the use of a model of an economy based on Sraffian view of
the competition (Negishi 1979).
Sraffa’s article “The laws of returns under competitive conditions” (Sraffa 1926)
is famous for his criticism against Marshall, but here we are interested particularly
for his view on the competition that the size of a competitive firm is determined not
by the cost of production but by the demand for its output.
“It is not easy, in times of normal activity, to find an undertaking which
systematically restricts its own production to an amount less than that which it could
sell at the current price, and which at the same time is prevented by competition
from exceeding that price. Business men, who regard themselves as being subject to
competitive conditions, would consider absurd the assertion that the limit to their
production is to be found in the internal conditions of production in their firm,
which do not permit of the production of a greater quantity without an increase
in cost. The chief obstacle against which they have to contend when they want
gradually to increase their production does not lie in the cost of production—but
in the difficulty of selling larger quantity of goods without reducing the price, or
without having to face increasing marketing expenses. This necessity of reducing
prices in order to sell a larger quantity of one’s own product is only an aspect of
the usual descending demand curve, with the difference that instead of concerning
the whole of a commodity, whatever its origin, it relates only to the goods produced
by a particular firm: and the marketing expenses necessary for the extension of its
market are merely costly efforts (in the form of advertising, commercial travellers,
facilities to customers, etc.) to increase the willingness of the market to buy from
it—that is, to raise that demand curve artificially.”
This is a view of competitive markets quite different from that of the current
mainstream neo-Walrasian economics. In Walrasian view of competitive markets, a
firm has no difficulty in selling whatever amount of its product at the given market
price, since the price is so adjusted that the total supply is always equalized to
the total demand in the market. The only obstacle against which a Walrasian firm
has to contend for its increasing production lies in the increasing marginal cost of
production. A Walrasian firm perceives an infinitely elastic demand curve for its
product and equalizes its marginal cost with the price to maximize its profit, i.e., the
so-called first postulate of the classical economics.
In the Sraffa’s view of a competitive market, a firm must face a kinked demand
curve for its product, since the price has to be reduced to sell a larger quantity of
its product while it is prevented by competition from exceeding the current price.
It cannot raise the price by the reduction of the supply quantity of its product. The
point of the kink in the demand curve for a firm is given by the level of demand
Fig. 17.4 Kinked demand

curve
for its product. The maximization of a firm’s profit requires that the marginal cost
is not higher than the price of the product, but higher than the marginal revenue
corresponding to the descending part of the demand curve.4
See Fig. 17.4, where we measure the price and the cost of a product vertically
and the level of output of a firm, horizontally. The market price is given as p, and
the firm can sell as much as pA at this price. If the firm wishes to sell a larger amount
than this, it must reduce the price below p, so that it faces a demand curve AD. In
other words, the firm faces a kinked demand curve pAD. The marginal revenue curve
for the firm is, then, pA up to the volume of output OB and EF beyond the output
OB. In other words, it jumps from A to E at the output OB. If the marginal cost
curve passes between points A and E as in the figure, the firm’s profit is maximized
at the output OB. Since the marginal cost curve need not to be increasing, the
existence of Marshallian internal economies can be consistent with the competition
in the sense of Sraffa. It can also be seen that this equilibrium remains unchanged
even if the marginal cost curve is shifted upwardly or downwardly, provided that it
is still located between A and E at the output OB. In other words, the price tends to
be rigid.5
Under the post-war regime of the fixed exchange rates, the conditions for the
favorable effects of the devaluation on the balance of payments were studied exten-
sively. Such a stability of the foreign exchange requires fairly large price elasticities
of foreign and domestic import demands. The reason why such elasticities are
relevant is that changes in foreign exchange rates are assumed to reflect soon on
4 Sraffa’smodel of a competitive firm is, therefore, formally similar to that of an oligopoly firm
later developed by Sweezy (1939). See Negishi (1998) for the implications of this model to the
microfoundations of Keynesian economics.
5 The price remains also unchanged against the changes in the aggregate level of the demand, since
the point A shifts horizontally. See Sweezy (1939).

changes in prices of the importables to be charged to the domestic consumers of the

importing countries. Being based on some empirical evidence (see, e.g., Harberger
1957), economists were almost unanimous to dismiss the so-called elasticity
pessimism and to be convinced of the stability of foreign exchanges when many
countries simultaneously adopted floating exchange rates. Experiences in 1970s
Japan showed, however, that changes in the domestic prices of the importables to
be charged to consumers are very slow and foreign exchange gains do not accrue to
consumers when Japanese yen was appreciated.
To explain such sticky or rigid domestic price of the importables in the face
of floating foreign exchange rates, let us discard a usual implicit assumption of
international trade theory, i.e., the assumption that domestic consumers can purchase
directly from foreign producers in the foreign market. This assumption may be
plausible in international trade among countries, like European ones, closely located
and socially and culturally very similar each other. In international trade between
countries far away and dissimilar, say USA and Japan, however, the role of
consumers is very limited and international trade is carried out almost exclusively
by the hands of specialized firms, i.e., exporters and importers. Domestic consumers
buy the importables in the domestic market from the foreign exporters or domestic
importers, because consumers have neither enough information nor suitable credit to
buy directly from foreign producers, cannot finance transportation costs individually
and are not well accustomed to do foreign exchange business.
Foreign exporters and domestic importers, which buy from foreign producers in
the foreign markets and sell to the domestic consumers in the domestic market, have
their marginal cost curves which are shifted as the foreign exchange rate changes
and face their kinked demand curves which can be explained in the same way as
in the case of the domestic producers. The situation is, therefore, exactly identical
to the case described in Fig. 17.4. The price of the importables to be charged to
the domestic consumers is very likely rigid in terms of the domestic currency even
if the marginal cost of foreign exporters and domestic importers is shifted in the
face of floating rate of foreign exchange. Domestic consumers face the prices of the
importables as well as those of the home produced goods which are rigid in terms
of domestic currency. Only flexible price, i.e., floating rate of foreign exchange,
can affect the economy merely through income effects which are due to foreign
exchange gains accrued to the domestic exporters and importers.
Such an absence of changes in consumers’ prices is, of course, considered to be a
short-run phenomenon. In the long-run foreign exchange gains induce entry of new
firms into profitable sectors. Newly entering firms have to undersell old existing
firms so that prices charged to consumers are reduced in such sectors. In the short
run, however, income effects are dominant. To see whether such income effects
stabilize or destabilize foreign exchange market, we have to develop more formal
models of the economy.
Suppose that international trade between two countries is exclusively carried
out by importers so that foreign exchange gains or losses, if any, accrue entirely
to importers of the importing country. Then a simple two-country model of
international trade is described by following equations.
The level of the output of the home country is denoted by X while the level of the
national income of the same country is denoted by x. The relation between X and x
is given by
x = X + (1 − R)M(x) (17.1)
where R is the rate of exchange defined as the price of the foreign currency in
terms of the home currency and M is the home demand for imports. In words, the
national income in terms of home currency is defined as the sum of the output, which
includes services of importers, and the foreign exchange gains (negative in the case
of losses) and the home demand for imports is an increasing function of the national
income, since the domestic price of the importables is constant in terms of the home
currency. Similarly, we have for the foreign country
1 ∗ ∗
x∗ = X ∗ + 1 − M (x ) (17.2)
R
where x∗ , X ∗ , and M ∗ are, respectively, the level of the national income, output, and
demand for imports of the foreign country.
The level of output is determined in a Keynesian way by the level of effective
demand, i.e.,
X = D(x) + M ∗ (x∗ ) (17.3)

and
X ∗ = D∗ (x∗ ) + M(x) (17.4)
where D and D∗ are, respectively, the domestic demand for domestic goods of
home and foreign countries, which are independent of prices since domestic prices
are constant in terms of the domestic currency in each country. Then, taking into
consideration the fact that there is exogenous demand represented by constant terms
in D, M, D∗ , and M ∗ , we note that
D + M < 1 (17.5)
and
D∗ + M ∗ < 1 (17.6)
where D is the derivative of D with respect to x and the like. In other words,
marginal propensities to consume are less than one.
By substituting (17.1) and (17.2) for X and X ∗ in (17.3) and (17.4), and
differentiating (17.3) and (17.4) with respect to R, we have, at R = 1,
dx −M(1 − D∗) + M ∗ M ∗
= (17.7)
dR A
and
dx∗ M ∗ (1 − D) − MM
= (17.8)
dR A
where
A = (1 − D )(1 − D∗) − M M ∗ (17.9)
which is positive in view of (17.5) and (17.6).

The numerator of (17.7) is negative if trade between two countries balances, i.e.,
B = M ∗ /R − M = 0. The appreciation of home currency, i.e., the reduction of R,
increases x in this case. The numerator of (17.8) is positive if the home country
is not in deficit, i.e., B is not negative. The appreciation of home currency then
reduces x∗ .
The stability of foreign exchange market is considered as follows. The change in
the balance of payments due to a change in foreign exchange rate is
dB dx∗ dx C
= M ∗ − M − M∗ = (17.10)
dR dR dR A
where, in view of (17.7) and (17.8)
C = (1 − D∗ − M ∗ )(MM − M ∗ ) − M ∗ M ∗ D . (17.11)
If, for example, the home country is in surplus, i.e., B = M ∗ /R − M > 0, and its
currency is appreciated, i.e., R is reduced, then B is increased, i.e., the surplus is
increased, since A is positive and C is negative in (17.10), in view of (17.9) and
(17.11). The foreign exchange market is unstable if trade balances and foreign
exchange gains (losses) accrue to the importing countries. The stabilizing effects
of changes in export and import are overtaken by the destabilizing effect of changes
in exchange rate, given export.6
Bibliography
Ethier, W. (1979). Internationally decreasing costs and world trade. Journal of International
Economics, 9, 1–24.
Harberger, A. C. (1957). Some evidence on the international price mechanism. Journal of Political
Economy, 65, 506–521.
Kemp, M. C., & Negishi, T. (1970). Variable returns to scale, commodity taxes, factor market
distortions, and their implications for trade gains. Swedish Journal of Economics, 72, 1–11.
6 SeeNegishi (1979) for the case where international trade is carried out only by exporters of
two countries and the case where international trade is carried out exclusively by exporters and
importers of one country.
Bibliography 137
Krugman, P. R. (1987). Increasing returns and the theory of international trade. In T. F. Bewley
(Ed.), Advances in economic theory. Cambridge: Cambridge University Press.
Marshall, A. (1961). Principles of economics. London: Macmillan.
Mathews, R. C. O. (1949). Reciprocal demand and increasing returns. Review of Economic Studies,
37, 149–158.
Melvin, J. (1969). Increasing returns to scale as a determinant of trade. Canadian Journal of
Economics, 2, 389–402.
Negishi, T. (1979). Foreign exchange gains in a Keynesian model of international trade. Economie
appliquee, 32, 623–633.
Negishi, T. (1998). Sraffa and the microfoundations of Keynes. European Journal of the History of
Sraffa, P. (1926). The laws of returns under competitive conditions. Economic Journal, 36,
535–550.
Sweezy, P. M. (1939). Demand under conditions of oligopoly. Journal of Political Economy, 47,
568–573.
Part III
Historical Appendix
Chapter 18
Adam Smith and Disequilibrium
Economic Theory
18.1 Introduction
A great classic often has many different aspects that permit many different and
mutually inconsistent interpretations by later scholars. The Wealth of Nations
(WN) of Adam Smith is a good example of such a classic. Smith’s theory of
natural prices has been interpreted and developed as an equilibrium theory by
modern economic theorists.1 We shall try, however, to interpret Smith’s economic
theory as disequilibrium theory. Of course, there already exist some disequilibrium
approaches to Smith on the dynamic process of growth involving increasing returns
to scale.2 We shall rather be concerned, however, with a disequilibrium approach
to the problems of markets, that is, international trade, competition and division
of labor, and a disequilibrium interpretation of what economists now refer to as
“increasing returns to scale.” We shall start this disequilibrium analysis from a study
of Smith’s theory of international trade. Smith explained international trade by the
existence of disequilibrium, that is, surplus, and was criticized by Ricardo from the
point of view of the equilibrium theory.
To add a new interpretation to the already rich variety of interpretations of
different aspects of the WN is certainly troublesome for those who wish to know
what Smith really meant. But our purpose here is to examine the history of
economics as a rich source from which we can obtain hints, suggestions, and
encouragement to develop new economic theories. Thus we are asking Adam Smith
to support our attempt to develop disequilibrium theories of economics against the
modern equilibrium economic theory.
1 For Smith’s theory of economic growth, see Kurz and Salvadori (2003), which emphasizes
the classical traditions in the recent equilibrium theories of endogenous growth in the current
mainstream economics.
2 See for such approaches, Arrow (2000) and Lavezzi (2003).

and Economics 2, DOI 10.1007/978-4-431-54433-3__18, © Springer Japan 2014;
Reprinted from Vivienne Brown (ed.) (2004) The Adam Smith Review Volume 1:30–39,
with permission of Routledge.
142 18 Adam Smith and Disequilibrium Economic Theory
The plan of this article is as follows. We begin in Sect. 18.2 by explaining the
difference between equilibrium theory and disequilibrium theory. Section 18.3 deals
with Smith’s theory of international trade. While the modern equilibrium theory
assumes that the comparative advantages between trading countries are exogenously
given from the differences in climate, technology, factor endowment, and so on,
Smith’s theory can explain endogenously the comparative advantages between sim-
ilar countries. We shall emphasize, however, that Smith’s international trade theory
can be interpreted only as a disequilibrium theory. Then, in the following Sect. 18.4,
we shall argue that Smith’s proposition on the division of labor and the extent of
the market—which economists interpret as increasing returns to scale—can also be
fully demonstrated only as a disequilibrium theory. Our aim is to argue that the
division of labor is advanced by the existence of excess supply which large markets
are more likely to generate. Finally, Sect. 18.5 provides a summary and conclusion.
18.2 Equilibrium Theory and Disequilibrium Theory
According to Schumpeter, modern economic theory has inherited its equilibrium

theory from Book I, Chap. vii, of Adam Smith’s WN which analyzes natural and
market price:
The rudimentary equilibrium theory of Chap. vii, by far the best piece of economic theory
turned out by A. Smith, in fact points toward Say and, through the latter’s work, to Walras.
The purely theoretical developments of the nineteenth century consist to a considerable
degree in improvements upon it. Market price . . . is treated as fluctuating around a ‘natural’
price . . . which is the price that is sufficient and not more than sufficient to cover ‘the whole
value of the rent, wages, and profit, which must be paid in order to bring’ to market that
quantity of every commodity ‘which will supply the effectual demand,’ that is, the demand
effective at that price.
(Schumpeter 1954, p. 189)
When the price of any commodity is neither more or nor less than what is sufficient to pay
the rent of the land, the wages of the labour, and the profits of the stock employed in raising,
preparing, and bringing it to market, according to their natural rates, the commodity is then
sold for what may be called its natural price.
(WN, Book I, Chap. vii, p. 4)
The natural price is the equilibrium price to which actual market prices are
attracted. When the quantity of a commodity which is brought to market falls short
of the effective demand, the market price will rise more or less above the natural
price, at least one component part of the price (the rent, wages, and profit) must rise
above its natural rate, and more land, labor, and stock are used in raising, preparing,
and bringing to market the commodity so that the quantity brought to market is
sufficient to supply the effectual demand. This implies, then, that the market price
falls to the natural price and each component part to its natural rate, respectively.
Similarly, when the quantity brought to market exceeds the effectual demand, the
market price sinks below the natural price and at least one component part falls
18.3 Smith’s Theory of International Trade 143
below its respective natural rate, with the result that the quantities of labor, land,
and stock used are diminished, the quantity of the commodity is equalized to the
effectual demand, and the natural price is regained with the natural rates of its
component parts. In other words, the equilibrium price of a commodity is defined
by the equality of the demand for it and supply of it in the market3; and it is reached
through the adjustment of disequilibrium market price which rises if the demand is
larger than the supply, and reduced if the supply is larger than the demand (WN,
Book I, Chap. vii, pp. 3–16).
Being based on the given natural rate of rent, wages, and profits, the natural
price of a commodity remains unchanged by changes in the market price through
which it is established. In this sense, we may call Smith’s theory of the natural
price an equilibrium theory. It implies that disequilibrium soon disappears without
any effects left on the equilibrium eventually established. As was pointed out by
Schumpeter (1954, p. 189), Walras’ theory of equilibrium is the one of the most
refined versions of such a theory (Walras 1954). Walras considered two different
solutions in his theory of equilibrium, that is, the mathematical solution and the
practical solution. The former solution is to confirm the equality of the number of
unknowns (the equilibrium prices) with the number of equations (the conditions
of supply and demand). The equilibrium is obtained without any consideration of
the behavior of markets in disequilibrium. The latter solution is the tâtonnement,
which explains how the problems of equilibrium are solved practically in the
markets by the mechanism of competition. Walras assumed that no actual exchange
transactions take place at disequilibria where prices are being changed according
to the law of supply and demand, so as to make the practical solution identical
to the mathematical one (Negishi 1989, pp. 253–4, 263–4). This ensures that the
equilibrium outcome is independent of the disequilibrium path of adjustment.
If disequilibrium does not disappear soon and if its effects on the equilibrium
eventually established are important, however, we have to give up the equilibrium
theory and consider disequilibrium theory. In the latter, the equilibrium eventually
established is not independent of the path in disequilibrium through which it is
reached. Even though Smith’s theory of the natural price can be considered as
equilibrium theory, we can also find examples of disequilibrium theory in the WN.
18.3 Smith’s Theory of International Trade
Although Schumpeter mentioned only Say and Walras in the earlier passage,
Ricardo and J. S. Mill also followed Smith’s equilibrium theory in the develop-
ment of the classical school of economics in England (see Negishi 1986 for the
equilibrium theory of J. S. Mill). Thus, Ricardo even criticized Smith’s theory
3 The natural rate of each component part of the price of a commodity, that is, the equilibrium price
of a factor of production, is also determined by its demand and supply. See Negishi (1993).
of international trade from the point of view of Smith’s equilibrium theory of

natural and market price. The core of Smith’s theory of international trade is based,
however, not on the equilibrium analysis of natural and market price first presented
in Book I, Chap. vii, but on an analysis of the historical and natural progress of
capital investment in different sectors of the economy in Book II.
When the produce of any particular branch of industry exceeds what the demand of the
country requires, the surplus must be sent abroad, and exchanged for something for which
there is a demand at home. Without such exportation, a part of the productive labour of the
country must cease, and the value of its annual produce diminish. The land and labour of
Great Britain produce generally more corn, woollens and hard ware, than the demand of
the home-market requires. The surplus part of them, therefore, must be sent abroad, and
exchanged for something for which there is a demand at home. It is only by means of such
exportation, that this surplus can acquire a value sufficient to compensate the labour and
expense of producing it.
(WN, Book II, Chap. v, p. 33)
One would be led to think by the above passage, that Adam Smith concluded we were under
some necessity of producing a surplus of corn, woollen goods, and hardware, and that the
capital which produced them could not be otherwise employed. It is, however, always a
matter of choice in what way a capital shall be employed, and therefore there can never, for
any length of time, be a surplus of any commodity; for if there were, it would fall below its
natural price, and capital would be removed to some more profitable employment. No writer
has more satisfactorily and ably shown than Dr. Smith, the tendency of capital to move from
employments in which the goods produced do not repay by their price the whole expenses,
including the ordinary profits, of producing and bringing them to market.
(Ricardo 1951, p. 291, n.)
Clearly, Smith’s theory is a disequilibrium theory, since he tried to explain the

equilibrium of international trade from the existence of surplus, that is, excess
supply, in the domestic markets that are out of equilibrium. He thus went beyond
his equilibrium theory of the natural price by taking account of market disequilibria
and framing an explanation of international trade on the existence of domestic
disequilibria. Ricardo, on the other hand, stuck to Smith’s theory of the natural price
and assumed away the effects on the final equilibrium outcome of what happened
during the disequilibrium process of adjustment, and so the classical theory of
international trade developed by Ricardo (1951, pp. 128–149) and J. S. Mill (1909,
pp. 574–606) was based on the equilibrium approach. Smith’s disequilibrium theory
of international trade is explained in more detail as follows:
Between whatever places foreign trade is carried on, they all of them derive two distinct
benefits from it. It carries out that surplus part of the produce of their land and labour for
which there is no demand among them, and brings back in return for it something else for
which there is a demand. It gives a value to their superfluities, by exchanging them for
something else, which may satisfy a part of their wants, and increase their enjoyments. By
means of it, the narrowness of the home market does not hinder the division of labour in
any particular branch of art or manufacture from being carried to the highest perfection.
By opening a more extensive market for whatever part of the produce of their labour may
exceed the home consumption, it encourages them to improve its productive powers, and
to argument its annual produce to the utmost, and thereby to increase the real revenue and
wealth of the society.
(WN, Book IV, Chap. i, p. 31)
18.3 Smith’s Theory of International Trade 145
Myint (1958) argued that there are two different and independent theories of
international trade in the above passage, that is, the “productivity theory” and the
“vent for surplus” theory. The former theory points out the possibility that, by
widening the extent of the market, international trade improves the division of labor
and raises the general level of productivity within the country, so that costs fall. The
latter theory is considered by Myint as appropriate for analyzing underdeveloped
economies rather than advanced economies with well-developed markets.4 The
difficulty with Myint’s explanation here as an account of Smith’s intention is that
Smith himself did not restrict the applicability of his theory in this way: “In every
period, indeed, of every society, the surplus part both of the rude and manufactured
produce, or that for which there is no demand at home, must be sent abroad in order
to be exchanged for something for which there is some demand at home” (WN,
Book III, Chap. i, p. 7). Furthermore, the productivity theory has a difficulty as
a complete theory since, as pointed out by Bloomfield (1975, p. 469), Smith did
not insist on the converse proposition that increasing returns are a cause of trade,
independent of the existence of a domestic surplus before trade.
A better interpretation would be to regard the productivity theory and the vent for
surplus theory as two parts of Smith’s single disequilibrium theory of international
trade, rather than as separate independent theories.5 Consider two countries very
similar with respect to size, factor endowments, technology, and demand patterns.
The natural price of commodities would be similar in the two countries and there
would be no international trade between them, given the existence of transportation
costs. Even if it is assumed that industries are subject to increasing returns to
scale, the equilibrium theory could not explain international trade between two such
countries. No one can tell which of the two identical countries can enjoy the fruits
of increasing returns through international trade. Even in advanced countries with
well-developed and highly competitive markets, nay particularly in such countries,
however, there often arises a disequilibrium, say, excess supply, in some domestic
market. This is Smith’s starting point of a surplus of some kinds of produce. The
market price then falls below the natural price. In Smith’s equilibrium theory of
natural price, the supply will be decreased through the shift of factors of production
out of such an industry. At the same time, the surplus of such a commodity
beyond domestic demand can also be exported. But there is also the ‘productivity’
aspect of Smith’s theory of international trade. When equilibrium is recovered,
the natural price can be lower than before if there has been an improvement
in the division of labor.6 The final equilibrium outcome thus depends on what
4 Myint (1977) admitted that the vent for surplus theory cannot be applied to the highly advanced
“commercial nations” such as Holland and Hamburg even in the eighteenth-century Europe. See
Elmslie (1998) for a recent survey on vent for surplus.
5 Haberler (1959, p. 9) seems to suggest that the vent for surplus is part and parcel of the
productivity theory.
6 One might ask whether the final equilibrium price can be higher if final exports were less than
initial excess supply. Such seems unlikely, however, since increasing returns in Smith’s theory is
due to the division of labor which, once made, might not be lost.
happens to productivity during the disequilibrium process of adjustment, since the

equilibrium price achieved will be determined by changes in productivity introduced
during the disequilibrium process of adjustment. The vent for surplus theory is,
therefore, together with the productivity theory, an indispensable part of Smith’s
disequilibrium (or path-dependence) theory of international trade.7 But how might
productivity be improved during the disequilibrium process of adjustment? This
question is considered in the following section.
18.4 The Division of Labor and the Extent of the Market
One of the best-known propositions in the WN is that the division of labor depends
on the extent of the market, so that if the size of the market increases, the general
level of productivity rises and costs fall. As seen in Sect. 18.3, this gives the
foundation to ‘the productivity theory’ of Smith’s theory of international trade.
Smith gave two different kinds of division of labor: one is concerned with the intra-
firm subdivision of different operations to produce a given product, the extent of
which is limited by the demand for output of a firm or a plant, and the other is
concerned with an inter-firm division of labor or the specialization of firms in the
same industry, the extent of which is limited by the demand for the industry as a
whole.
If an increase in the demand for the industry not only induces the entry of new
firms, but also expands the scale of production of each firm, then equilibrium theory
shows that this causes changes in the intra-firm division of labor and the price of
the product is reduced.8 This is because “the owner of the stock . . . necessarily
endeavours, for his own advantage, to make such a proper division and distribution
of employment, that they [the labourers] may be enabled to produce the greatest
quantity of work possible . . . in a particular workhouse” (WN, Book I, Chap. viii,
p. 57). Smith argued, however, that “what takes place among the labourers in a
particular workhouse, takes place, for the same reason, among those of a great
society. The greater their number, the more they naturally divide themselves into
different classes and subdivisions of employment.” If this implies the inter-firm
division of labor in the same industry, it is difficult to explain it by the ordinary
equilibrium theory. The reason is that there exists no one who endeavors to make
such a proper division of labor among firms “that they may be enabled to produce
the greatest quantity of work possible” for the industry as a whole. In this case, the
final equilibrium division of labor is not independent of divisions of labor made in
disequilibrium situations of the adjustment process. Furthermore, as was studied in
7 An example of vent for surplus export in a developed country is the tendency of exports to increase
in recessions, which has been observed in Japan since the end of the 1950s and called the export-
drive effect of a recession. See Komiya (1990, p. 357).
8 On the comparative static analysis of a maximizing equilibrium, see Negishi (2000).
18.4 The Division of Labor and the Extent of the Market 147
Negishi (2000), it is by no means clear how and why we can have increasing returns
to scale from such divisions of labor.9
Most modern interpreters of Adam Smith refer to the following passage from the
WN as a summary statement of Smith’s theory that the division of labor is limited
by the extent of the market (e.g., Hollander (1973, p. 212), Sylos-Labini (1976,
pp.205–6), and Richardson (1975, pp. 353–4)).
The increase of demand, besides, though in the beginning it may sometimes raise the price
of goods, never fails to lower it in the long run. It encourages production, and thereby
increases the competition of the producers, who, in order to undersell one another, have
recourse to new divisions of labour and new improvements of art, which might never
otherwise have been thought of.
(WN, Book V, i.e. 26, p. 748)
It is somewhat curious that this passage is not from Book I, Chap. III, of the WN,
entitled “That the division of labour is limited by the extent of the market,” but from
Book V, Chap. i, entitled “Of the expenses of the sovereign or common-wealth”
where a problem of the East India Company is discussed.10 It is very clear, however,
that here what advances the division of labor directly, and hence pushes down prices,
is not the increase of demand itself, but the resulting excess supply11 through the
increased competition which forces the suppliers to undersell each other. In this
sense, it is not the increase of demand but the excess supply that can more directly
advance the division of labor. If the enlarged size of a market, where the number of
firms is larger, can reduce the equilibrium natural price, this is because the larger
market is more competitive and is more likely to give rise to excess supply than a
smaller market.12 Since the final equilibrium outcome in Smith’s analysis depends
on what happens in disequilibrium when there is excess supply, Smith’s analysis is
a typical disequilibrium theory.
Why are larger markets more unstable13 in that they are more likely to give rise
to excess supply? In so far as the increase of demand induces the entry of new firms,
the larger markets are more competitive in the sense that the number of firms is
9 To explain rather than to assume increasing returns is very difficult, even for the case of
equilibrium economics, in view of Robinson’s (1933, p. 219) criticism against Pigou (1932, p. 338).
10 Smith criticized the East India Company which insisted that the increase of demand in the Indian
market raised the price of Indian goods.

11 Even if the market is not perfectly competitive, excess supply will appear, since firms perceive
subjective demand curves, like Chamberlin’s dd curves, which are more elastic than the true
objective ones. See note 15.
12 Vassilakis (1987) stated that Adam Smith, among others, formulated the proposition that the
division of labor is limited by the stability of the market in the sense that a reduction in demand
uncertainty is equivalent to an increase in market size and reduction in uncertainty will increase the
degree of division of labor. We cannot agree with this statement in this otherwise highly instructive
survey, since it is the demand uncertainty and the instability of the large competitive market which
will increase the degree of division of labor.
13 The use of the term “unstable” is not in the Walrasian sense. It is in the sense of Vassilakis (1987).
See note 12.

greater there.14 So the question is, why is a greater number of firms in the market
more likely to lead to excess supply? Perhaps, the modern theory of competition can
answer the question as follows. Suppose for a moment that the cost of production
is reduced. The equilibrium in an oligopoly market where the number of firms is
small is likely to remain unchanged. This is because each firm does not increase
its supply, taking seriously into consideration the reaction of other firms (see, e.g.,
Sweezy 1939). In a monopolistically competitive market where the number of firms
is very large, however, the equilibrium is unstable in the sense that excess supply is
very likely to appear, since each firm perceives its own demand curve (dd) as more
elastic than the true demand curve (DD) and so expands its supply excessively for
the given cost reduction.15 Thus the larger the number of firms, the more likely it is
that individual firms overestimate the elasticity of their own demand curve and so
overestimate the returns from expanding production, whatever the original stimulus
to do this. If all firms do this, however, there is excess supply and firms are then
forced to find ways of cost-cutting in order to survive.
Although Smith was not referred to explicitly, it was Charles Babbage who
developed the disequilibrium theory further and argued that “one of the natural and
almost inevitable consequence of competition is the production of a supply much
larger than the demand requires” (Babbage 1835, p. 231) and that,
The effect of gluts in producing improvement in machinery, or in methods of working . . .
by the diminution of profit which the manufacturer suffers from the diminished price, his
ingenuity will be additionally stimulated; that he will apply himself to discover other and
cheaper sources for the supply of his raw materials . . . that he will endeavour to contrive
improved machinery which shall manufacture it at a cheaper rate . . . or try to introduce new
arrangements into his factory, which shall render the economy of it more perfect.
(Babbage 1835, p. 233; see also Karayiannis 1998)
18.5 Summary and Conclusion
We have argued that Smith’s theory of international trade may be reconstructed

using both the vent for surplus theory and the productivity theory. The former is
clearly a disequilibrium theory which explains the cause of international trade in
excess supply, while the latter explains the effects of trade on an economy in the
productivity gains arising from the divisions of labor which are made possible only
in the large international markets. Why, then, is the division of labor limited by the
extent of the market?
14 “For Adam Smith, as well as for other classical economists, competition is characterized by free
entry” (Sylos-Labini 1976, p. 200).

15 The firm’s dd curve is perceived under the supposition that other firms’ supply remains
unchanged (Chamberlin 1948, pp. 90–4). Roughly speaking, a demand curve is more elastic if
the percentage increase in quantity demanded is larger for a given percentage reduction in price.
Bibliography 149
Referring to the argument of Smith in the WN (Book V, Chap. i), it is suggested

that “new divisions of labour” are introduced by producers who, facing excess
supply in the market, must “undersell one another.” It can be argued that larger
markets with many competitors (e.g., monopolistically competitive markets with
free entry) are more likely to produce such excess supplies than the small markets
with a few competitors (e.g., oligopolistic markets).
If firms facing excess supplies in the market try to introduce “new divisions
of labour” in order to “undersell one another,” then we may shed new light on
the relation between the two aspects of Smith’s theory of international trade. It is
the surplus theory (the existence of excess supply) which necessarily leads to the
productivity theory (“increase in the real revenue and wealth of the society”).
In other words, it is the existence of the initial surplus as such in a country, not
the international trade induced, which leads it to “improve its productive powers.”
Thus, the crux of Smith’s disequilibrium economics is the market excess supply
which causes individual producers to introduce “new divisions of labour” so as to
survive in the competition to “undersell one another.”
Acknowledgments For comments or other forms of help, the author is grateful to the editor,
referees and Professor Hiroji Nakamura. The usual caveat applies.
Bibliography
Arrow, K. J. (2000). Increasing returns: historiographic issues and path dependence. European
Journal of the History of Economic Thought, 7, 171–180.
Babbage, C. (1835)[1832]. On the economy of machinery and manufactures. London: Charles
Knight.
Bloomfield, A. I. (1975). Adam Smith’s theory of international trade. In A. S. Skinner, and T.
Wilson (Eds.), Essays on Adam Smith. Oxford: Oxford University Press.
Chamberlin, E. H. (1948). The theory of monopolistic competition. Cambridge, MA: Harvard
University Press.
Elmslie, B. T. (1998). Vent for surplus. In H. D. Kurz, & N. Salvadori (Eds.), The Elgar companion
to classical economics. L-Z, Cheltenham: Edward Elgar.
Haberler, G. (1959). International trade and economic development. Cairo: National Bank of
Egypt.
Hollander, S. (1973). The economics of Adam Smith. Toronto: University of Toronto Press.
Karayiannis, A. D. (1998). Supply-push and demand-pull factors of technological progress in the
early decades of the 19th century (1800–1840). History of Economic Ideas, 6, 45–68.
Komiya, R. (1990). The Japanese economy: Trade, industry and government. Tokyo: University of
Tokyo Press.
Kurz, H. D., & Salvadori, N. (2003). Theories of economic growth: old and new. In H. D. Kurz, &
N. Salvadori (Eds.), The theory of economic growth, classical perspective. Gloucester: Edward
Elgar.
Lavezzi, A. (2003). Smith, Marshall and Young on division of labour and economic growth.
European Journal of the History of Economic Thought, 10, 81–108.
Mill, J. S. (1909)[1848]. Principles of political economy. London: Longmans, Green & Co.
Myint, H. (1958). The “classical theory” of international trade and the underdeveloped countries.
Economic Journal, 68, 317–37.
Myint, H. (1977). Adam Smith’s theory of international trade in the perspective of economic
development. Economica, 44, 231–248.
Negishi, T. (1986). Thornton’s criticism of equilibrium theory and Mill. History of Political
Economy, 18, 567–577.
Negishi, T. (1989). History of economic theory. Amsterdam: North-Holland.
Negishi, T. (1993). A Smithian growth model and Malthus’ optimum propensity to save. European
Journal of the History of Economic Thought, 1, 115–127.
Negishi, T. (2000). Adam Smith’s division of labour and structural changes. Structural Change
and Economic Dynamics, 11, 5–11.
Pigou, A. C. (1932). The economics of welfare. London: Macmillan.
Ricardo, D. (1951)[1817]. In P. Sraffa, (Ed.). On the principles of political economy and taxation.
Cambridge: Cambridge University Press.
Richardson, G. B. (1975). Adam Smith on competition and increasing returns. In A. S. Skinner, &
T. Wilson (Eds.), Essays on Adam Smith. London: Oxford University Press.
Robinson, J. (1933). Economics of imperfect competition. London: Macmillan.
Schumpeter, J. A. (1954). History of economic analysis. New York: Oxford University Press.
Smith, A. (1976)[1776]. In R. H. Campbell, & A. S. Skinner (Eds.), An inquiry into the nature and
causes of the wealth of nations. Oxford: Oxford University Press.
Sweezy, P. M. (1939). Demand under conditions of oligopoly. Journal of Political Economy, 47,
568–573.
Sylos-Labini, P. (1976). Competition: the product markets. In T. Wilson, & A. S. Skinner (Eds.),
The market and the state. Oxford: Oxford University Press.
Vassilakis, S. (1987). Increasing returns to scale. In J. Eatwell, M. Milgate, & P. Newman (Eds.),
The new palgrave, vol. 2. London: Macmillan.
Walras, L. (1954). Elements of pure economics (W. Jaffé, Trans.). Homewood: Irwin.
Chapter 19
Complete Specialization in Classical Economics
19.1 Interpretations of Specialization
The so-called modern interpretation of Ricardian theory of comparative advantage

results in the drastic conclusion that each country (England or Portugal) specializes
entirely in the production of a single commodity (cloth or wine). But Ricardo
himself was merely concerned with marginal adjustments of production to the given
terms of trade in his famous theory of gains from foreign trade. Ricardo has nothing
to do with the complete specialization. It was J. S. Mill, however, who used the
assumption of the complete specialization skillfully to determine the terms of trade
uniquely in his theory of the reciprocal demand. Classical economists, including
Bastable and W. T. Thornton, critically discussed many important aspects of Mill’s
theory, but they did not seem to raise the objection to Mill’s assumption of entire
specialization. It was Pareto, a neoclassical economist, who presented a numerical
example for which the assumption is inappropriate.
19.2 Ricardian Theory of Competitive Advantage
Young Samuelson at Harvard was asked by his friend Stanislaw Ulam, a mathemati-
cian, to name one proposition in all of the social sciences which is both true and
nontrivial. Although he failed to reply at that time, some 30 years later Samuelson
was convinced that an appropriate answer is the Ricardian theory of comparative
advantage (Samuelson 1972, p. 683).
England may be so circumstanced, that to produce the cloth may require the labour of 100
men for one year; and if she attempted to make the wine, it might require the labour of 120
men for the same time. England would therefore find it in her interest to import wine, and

and Economics 2, DOI 10.1007/978-4-431-54433-3__19, © Springer Japan 2014;
Reprinted from John Vint, J. Stanley Metcalfe, Heinz D. Kurz, Neri Salvadori, Paul A.
Samuelson (eds.) (2010) Economic Theory and Economic Thought: Essays in honour of
Ian Steedman:82–97, with permission of Routledge.
152 19 Complete Specialization in Classical Economics
to purchase it by the exportation of cloth. To produce the wine in Portugal, might require
only the labour of 80 men for one year, and to produce the cloth in the same country, might
require the labour of 90 men for the same time. It would therefore be advantageous for her
to export wine in exchange for cloth . . . Thus England would give the produce of the labour
of 100 men, for the produce of the labour of 80.
(Ricardo 1951, p. 135)
Let us consider a typical example of the modern interpretation of this classical

theory of the comparative advantage, which was developed by Samuelson and
others.
Portugal can divert resources from food to clothing production and in effect convert one
unit of food into one unit of clothing; England, on the other hand, can convert one unit
of food into two units of clothing. Almost certainly Portugal will specialize completely in
food, England completely in clothing . . . Both countries will be better off than if they do
not specialize.
(Dorfman, Samuelson, and Solow 1958, p. 31)
Mathematically, then, it is an example of linear programming. England will

maximize her National Product Z,
Z = px1 + x2 (19.1)
being subject to her resource constraint
2x1 + x2 ≤ C (19.2)
where p (1 < p < 2) is the internationally given relative price of food in terms
of clothing, x1 (≥ 0) and x2 (≥ 0) are total output of food and that of cloth in
England, respectively, and C is the given resource (say, labor population). Similarly,
Portugal will maximize her National Product Z defined as (19.1), being subject to
her resource constraint
x1 + x2 ≤ C (19.3)
where x1 (≥ 0) and x2 (≥ 0) are total output of food and output of cloth in Portugal.
It can easily be seen that the solution is x1 = 0, x2 = C for England, and x1 =
C and x2 = 0 for Portugal. England specializes entirely in cloth production and
Portugal in food production (Dorfman et al. 1958, pp. 31–32).
19.3 Ricardian Model of the Economy
In the so-called Ricardian model of the modern standard interpretation of the

comparative advantage theory of Ricardo, only the labor is the factor of production
(like C in (19.2) and (19.3) in the above) and there exists neither land nor capital.1
1 Steedman called this interpretation “text book Ricardian theory” (Steedman 1971, p. 14, 1979,
pp. 7–8).
19.3 Ricardian Model of the Economy 153
Therefore, the only cost of production is the labor cost, and all the commodities
produced are distributed among laborers so that the GNP is composed only of
the wage income. According to Ricardo’s numerical example, however, the labor
productivity is higher in Portugal than in England not only in the production of
cloth but also in that of wine. Thus, Samuelson is quite right to accuse Ricardo of
his odd economic geography.
Writing in the heyday of England’s industrial revolution, which country do you think
Ricardo made out to be the most productive? Obviously I should not have raised the question
if Ricardo had not selected Portugal as the superior of England in every respect, having a
real per capita G.N.P. in Colin Clark units that is somewhat between one-ninth and one-half
greater depending upon whether you are a drunkard or a dandy. Why this odd economic
geography?
(Samuelson 1972, p. 679)2
What is odd is, however, not Ricardo’s economic geography, but the so-called
Ricardian model of the modern interpretation of his comparative advantage theory.
Ricardo declared in the Preface to his Principles (1817, 1951a) that the principal
problem of the political economy is to determine the laws which regulate the
distribution of GNP among landowners, capitalists, and laborers under the names
of rent, profit, and wages.
The produce of the earth—all that is derived from its surface by the united application of
labour, machinery, and capital, is divided among three classes of the community; namely,
the proprietor of the land, the owner of the stock or capital necessary for its cultivation, and
the labourers by whose industry it is cultivated. . . . To determine the laws which regulate
this distribution, is the principal problem in Political Economy.
(Ricardo 1951a, p. 5)
In the true Ricardian model of the economy which exists behind the four numbers
of Ricardo’s theory of comparative advantage, therefore, there must exist land
and capital, in addition to labor, as the factors of production which require the
remuneration.
Economic growth implies, for Ricardo, the accumulation of capital and the
increase of labor population. Since land is given, however, the marginal productivity
of capital and labor declines as a result of the economic growth. In other words, these
marginal productivities are lower than average. Now Ricardo’s four numbers must
be interpreted to show the level of the marginal productivities of labor, rather than
the average productivity. As a result of economic growth the marginal productivity
of labor becomes lower in England than in Portugal, but the average productivity is
high. Since the population of a country is largely dominated by the labor population,
therefore, per-capita GNP is higher in England than in Portugal. This is because the
large land rent income results, as land becomes more scarce, from the difference
between the average and marginal productivities of labor. GNP consists not only of
2 If
you are a drunkard who measures GNP in terms of wine, a real per-capita GNP in Portugal is
1/2 greater than in England since 1/2 = (120 − 80)/80. Similarly, if you are a dandy, it is 1/9
greater since 1/9 = (100 − 90)/90.
the wage income but also of profit and rent incomes. Ricardo’s economic geography
is not odd, therefore, from the point of view of the true Ricardian model of the
economy.
19.4 Ricardo’s Theory of Gains from Foreign Trade
Ricardo’s numerical example of the comparative advantage should be interpreted as

the adjustments in the marginal land, where the labor productivity is lowest, not only
from the point of view of the agricultural fertility but from that of, for example, the
locational convenience for the agricultural and manufacturing industries. Ricardo
himself was merely concerned with marginal adjustments of production to the given
terms of trade in his famous theory of gains from foreign trade. In other words,
Ricardo has nothing to do with the complete specialization. In fact, Mizuta (2004)
has already emphasized this interpretation, by making reference to the following
arguments by Ricardo himself.
It will appear then, that a country possessing very considerable advantages in machinery
and skill, and which may therefore be enabled to manufacture commodities with much less
labor than her neighbors, may, in return for such commodities, import a portion of the corn
required for its consumption, even if its land were more fertile, and corn could be grown
with less labour than in the country from which it was imported.
(Ricardo 1951a, p. 136)3
If things were allowed to take their own course, we should undoubtedly become a great
manufacturing country, but we should remain a great agricultural country also. Indeed, it
was impossible that England should be other than an agricultural country: she might become
so populous as to be obliged to import part of her food.
(Ricardo 1952, pp. 816–5)
[A]n objection which is frequently made against freedom of trade in corn. . . . This objection
is founded on the supposition that we should be importers of a considerable portion of the
quantity which we annually consume. . . . I differ with those who think that the quantity
which we should import would be immense; . . . Poland and Germany. . . . To raise a larger
supply, too, those countries would be obliged to have recourse to an inferior quality of land,
and it is the cost of raising corn on the worst soils in cultivation requiring the heaviest
charges, which regulates the price of all the corn of a country.
(Ricardo 1951b, pp. 264–265)
3 Referringto this quotation from Ricardo, Steedman and Metcalfe already recognized that Ricardo
had considered the case of incomplete specialization (Steedman & Metcalfe 1973; Steedman 1979,
p. 108).
19.5 Mill’s Assumption of the Single Factor of Production 155
19.5 Mill’s Assumption of the Single Factor of Production
Perhaps it was J. S. Mill who, by the assumption of a single factor of production

(i.e., labor), begins the modern complete specialization interpretation of Ricardian
theory.
It is established, that the advantage which two countries derive from trading with each other,
results from the more advantageous employment which thence arises, of the labour and
capital—for shortness let us say the labour—of both jointly. The circumstances are such,
that if each country confines itself to the production of one commodity, there is a greater
total return to the labour of both together; and this increase of produce forms the whole of
what the two countries taken together gain by the trade.
(Mill 1874, p. 5)
The problem for Mill was “to inquire, in what proportion the increase of produce,
arising from the saving of labour, is divided between the two countries” (Mill 1874,
p. 5; Hollander 1985, p. 322). In other words, his problem is the determination of the
terms of trade which Ricardo simply assumed as given. Mill’s theory is, of course,
that of reciprocal demand developed in Chap. 18 of his Principles.
This Law of International Value is but an extension of the more general law of Value, which
we called the Equation of Supply and Demand—the supply brought by the one constitutes
his demand for what is brought by the other. So that supply and demand are but another
expression for reciprocal demand: and to say that value will adjust itself so as to equalize
demand with supply, is in fact to say that it will adjust itself so as to equalize the demand
on one side with the demand on the other.
(Mill 1909, pp. 592–593)
Mill’s theory of the reciprocal demands is nothing but the international version
of the general equilibrium theory. According to Mill, however, the theory is still not
perfect, since it fails to determine uniquely the equilibrium terms of trade.
[I]ntelligent criticism (chiefly those of my friend Mr. William Thornton), and subsequent
further investigation, have shown that the doctrine stated in the preceding pages, though
correct as far as it goes, is not yet complete theory of the subject matter—several different
rates of international value may all equally fulfill the conditions of this law.
(Mill 1909, pp. 596–597)4
Even in the modern general equilibrium theory, however, while the existence of
a general equilibrium is proved under fairly reasonable assumptions, the study on
the uniqueness of the equilibrium has not yet been developed. It is natural, then, that
in the period of classical economics Mill had to assume some stringent simplifying
conditions to solve this problem in the last three sections of his Chap. 18, which
Edgeworth (1894) called “superstructure.”
To demonstrate the existence of unique terms of trade, which equate reciprocal
demands between two countries, Mill made, first, assumptions which led to entire
4 What is ironical, however, is that Mill misunderstood Thornton’s criticism (see Negishi 1998,
2001, pp. 63–70, 2002).

specialization. A set of sufficient conditions for complete specialization are, as

already seen in Sect. 19.2 (the so-called modern interpretation of Ricardo), the
assumption of the single factor of production, labor, and that of constant input
coefficients in the production of two commodities.
The supposition was, England could produce 10 yards of cloth with the same labour as 15
of linen, and Germany with the same labour as 20 of linen; that a trade was opened between
the two countries; that England thenceforth confined her production to cloth, and Germany
to linen.
(Mill 1909, p. 597)
Second, even though, or since, he does not know “any laws of the consumption
of wealth as the subject of a distinct science” (Mill 1874, p. 132; Hollander 1985,
p. 268),5 Mill simply assumes a unit own-elasticity of demand with respect to
price, zero cross-elasticities of demand with respect to price and a unit income
elasticity of demand that “any given increase of cheapness produces an exactly
proportional increase of consumption; or, in other words, that the value expended
in the commodity, the cost incurred for the sake of obtaining it, is always the same,
whether that cost affords a greater or smaller quantity of the commodity” (Mill
1909, p. 598). In other words, the proportion in which the total income is to be spent
on each commodity is a given constant, irrespective of the level of income and the
prices of commodities.
19.6 The Two-Country, Two-Good Case
For the two-country (England and Germany), two-good (cloth and linen) case, then,
Mill can demonstrate that the relative international value (the terms of trade) is
uniquely determined. Let us assume that England (Germany) has the comparative
advantage in the production of cloth (linen), and England (Germany) is specialized
in the production of cloth (linen) after trade.
The terms of trade t (the price of cloth in terms of linen after trade) is solved
from
pm
n= (19.4)
t
where m is “the cloth previously i.e., before trade or in autarky required by Germany
(at the German cost of production),” n is “the quantity of cloth which England can
make with the labour and capital withdrawn from the production of linen [after
trade]” and p is “the cost value of cloth (as estimated in linen) in Germany” (Mill
1909, pp. 600–601).6
5 Inaddition, “the inclinations and circumstances of consumers cannot be reduced to any rule”
(Mill 1909, p. 587; Hollander 1985, p. 324).
6 As was pointed out and corrected by Chipman (1979), however, Mill made a slip and could not
derive (19.4) correctly.

19.6 The Two-Country, Two-Good Case 157
Fig. 19.1 The situation in

Germany
1 1
Figure 19.1 describes the situation of Germany.7 The quantity of linen is

measured vertically, and that of cloth horizontally. The maximum quantity of
linen Germany can produce is OB1 and that of cloth, OA1 . Point G indicates the
production and consumption of cloth and linen in Germany before trade (at autarky).
Thus Mill’s m is equal to OC. The slope of the line A1 B1 is p, i.e., p = OB1 /OA1 .
Similarly, Fig. 19.2 describes the situation in England. The maximum quantity of
linen England can produce is OB2 and that of cloth, OA2 . Let us suppose that point
E indicates the production and consumption of cloth and linen in England before
trade (at autarky). If England specializes in the production of cloth after trade, Mill’s
n is equal to DA2 .
Thus (19.4) can be explained as follows. German expenditure on cloth before
trade is pm in terms of linen, since p is also the before-trade price of cloth in terms
of linen there. Now German demand for English cloth after trade is pm/t, from the
assumption of the unit-own elasticity of demand with respect to price which has now
changed from p to t, while German income in terms of linen remains unchanged
before and after trade at OB1 (see Fig. 19.1). German demand should be equal to the
after-trade supply of cloth from England, which is equal to n by definition. Equation
(19.4) expresses the equality of demand and supply of cloth in the international
market. We can solve (19.4) for the terms of trade which will prevail after trade
(i.e., t, from the data available to us before trade, namely p, m, and n) if we can
assume that each country is specialized entirely after trade.
7 Figures 19.1 and 19.2 are reproduced from Negishi (2001, pp. 55 and 56).
Fig. 19.2 The situation in England
It is clear that the assumption of specialization played important roles in Mill’s

demonstration. It ensures that German income in terms of linen remains unchanged
before and after trade, which, in conjunction with Mill’s simplifying assumptions on
consumption, explains the German demand for English cloth on the left-hand side
of his (19.4). Similarly, it makes English income in terms of cloth and her demand
for cloth unchanged before and after trade, which explains why n on the right-hand
side of (19.4) signifies English supply of cloth to Germany.
19.7 Bastable on Reciprocal Demand
Even Mill, however, did not take full advantage of the assumption of complete
specialization. This is why, to demonstrate that the terms of trade are determined
uniquely, Mill considered that “we must take into consideration not only, as we have
already done, the quantities demanded in each country of the imported commodities;
but also the extent of the means of supplying that demand which are set at liberty
in each country by the change in the direction of its industry” (Mill 1909, p. 597).
In (19.4), to determine the terms of trade t, we need not only the quantity demanded
of the commodity to be imported, m, but also n, which is the quantity of a commodity
to be produced from means of production set at liberty by the change in the direction
of the industry. On this, Mill was criticized by Bastable:
The attempt made by Mill to amend his theory by introducing the additional element of the
amount of capital set free for the production of exports is, as he even admits, a failure; for,
in the case of two countries and two commodities, the amount of free capital, or, as I should
19.8 Thornton on Supply and Demand Theory 159
prefer to say, “productive power,” is evidently determined by reciprocal demands, so that

nothing is gained by the laborious and confusing discussion in Sects. 6, 7, 8 of Chap. xviii.
(Bastable 1900, p. 29)
As Chipman (1979) showed,
m = a 1 A1 , n = b 2 A2 (19.5)
where A1 (A2 ) is the maximum quantity of cloth which Germany (England) can
produce, and a1 (b2 ) the constant proportion in which expenditure is assumed to be
devoted to cloth (linen) in Germany (England). This is because the German national
income in terms of cloth is OA1 before trade; as is seen in Fig. 19.1, the English
national income in terms of cloth is OA2 before trade in Fig. 19.2, and b2 = 1 − a2 ,
where a2 is the constant proportion in which expenditure is assumed to be devoted
to cloth in England. Thus (19.4) may be written as
tb2 A2 = a1 B1 (19.6)
where B1 is the maximum quantity of linen Germany can produce, as is shown in

Fig. 19.1, since p = B1 /A1 . The right-hand side of (19.6) is the demand for cloth
from Germany and the left-hand side is the demand for linen from England, both in
terms of linen. The terms of trade t can be uniquely determined by the equation of
reciprocal demands (19.6) and there is no need to introduce “the quantity of cloth
which England can make with the labour and capital withdrawn from the production
of linen,” i.e., n. So far Bastable’s criticism seems to be right.
Bastable went too far, however, to deny the important significance of Sects. 6,
7, and 8 of Chap. xviii of Mill’s Principles in which Mill tried to demonstrate the
uniqueness of the equilibrium terms of trade. The theory of the reciprocal demand
in general, developed in Sects. 1–5 of the same chapter, cannot ensure the unique
solution, as Mill argued in Sect. 6.8 Even though it is not the given value of n in
(19.4), it is necessary to introduce additional assumptions to prove the uniqueness.
Such assumptions are, as we have seen, those which lead to complete specialization
and those simplifying ones on consumption.
19.8 Thornton on Supply and Demand Theory
Chipman (1979) evaluated Mill’s solution of t from (19.4) very high, as historically
it is the first demonstration of the equilibrium price by the use of the equality of
demand and supply. To this classical equilibrium theory, however, W. T. Thornton
(1866, 1869, 1870) was very critical.
8 Therefore, my previous argument (Negishi 2001, pp. 56–57) should be amended in this respect.
Even if it were true that the price ultimately resulting from competition is always one at
which supply and demand are equalized, still only a small portion of goods offered for sale
would actually be sold at any such price, since a dealer will dispose of as much of his stock
as he can at a higher price, before he will lower the price in order to get rid of the remainder.
(Thornton 1869, p. 53, 1870, p. 65)
At first, Mill simply denied the significance of such trades at disequilibrium

prices.
Limitations such as these affect all economical laws, but are never considered to destroy
their value. As well might it be called an significant truth that there is a market price of a
commodity, because a customer who is ignorant, or in a hurry, may pay twice as much for
the thing as he could get it for at another shop few doors farther off.
(Mill 1869, p. 639)
Later, however, even Mill had to admit that “there has been some instructive
discussion on the theory of Demand and Supply—by which additional light has
been thrown on these subjects” (Mill 1871, p. xxxi).
Thornton is concerned with the non-uniqueness of the final equilibrium price,
which is due to shifts in demand and supply curves caused by exchanges at
non-equilibrium prices. If Mill’s “superstructure” aimed to reply to Thornton’s
criticism of demand and supply equilibrium theory, Mill should have dealt with
this problem. Thanks to the assumption of complete specialization, which implies
that the production is already finished, the new problem to be solved is the
changes in consumers’ demand due to the exchanges at disequilibrium prices.
Mill’s simplifying assumption on consumers’ demand is, fortunately, also helpful
in dealing with this problem. In addition to Mill’s assumption, let us assume that
consumers in two countries, Germany and England, have identical taste, so that
the world demand for both commodities, cloth and linen, is independent of the
distribution of income between countries. In other words, the changes in demand
for any commodity of any country caused by a redistribution of world income are
offset by those of the other country completely.
In the model used in Sect. 19.6 above, Germany specializes the production of
linen B1 (see Fig. 19.1) and England that of cloth A2 (see Fig. 19.2). The world
income is then B1 + tA2 , where t is the international price of cloth in terms of linen.
The condition for the demand and supply equilibrium for cloth in the world market is
a(B1 + tA2) = A2t (19.7)
where a is the constant proportion in which the expenditure is devoted to cloth

(identical for both countries). The equilibrium terms of trade t may be solved as
aB1
t= (19.8)
bA2
where b = 1 − a is the constant proportion in which the expenditure is devoted

to linen. It is independent of any redistribution of the world income, caused by
exchanges made at non-equilibrium prices.
19.9 Pareto on Complete Specialization 161
Thus, Mill’s model in his superstructure can deal with Thornton’s criticism of
demand and supply theory, if identical taste is assumed for both countries. This
additional assumption does not seem to be a stringent one, in view of assumptions
which have usually been made in the history of international trade theory. Thus, a
unique rate of international value can be determined by the principle of reciprocal
demands equation, even if demand and supply curves of individual commodities
are shifted as a result of exchange transactions at other rates of international value
(Negishi 1998; 2001, pp. 66–67).
19.9 Pareto on Complete Specialization
Bastable, Thornton, and other classical economists did not seem to be critical of
Mill’s use of the assumption of complete specialization to demonstrate the theory of
reciprocal demands. In spite of the fact that Ricardo himself emphasized incomplete
specialization as we saw in Sect. 19.4 above, Mill and the contemporary classical
economists admitted the assumption of the complete specialization to demonstrate
the principle of reciprocal demands. However, Pareto, a neoclassical economist after
the marginal revolution, raised the objection to the classical assumption of complete
specialization.
Ricardo’s reasoning is good only to illustrate one possible case. Let A and B be
the two goods which Ricardo talks about, and assume that in one day the less skillful
worker produces 1 of A, or 1 of B. According to Ricardo’s example, the more skillful
worker will in one day make six-fifths of A or four-thirds of B. This is shown by the
following table in which I and II identify the workers.
I II
A 6/5 1
B 4/3 1
Let us assume that the two workers each work 30 days producing A, 30 days
producing B, and that their wants are satisfied. We will have:
I II Total quantities
A 36 30 66
(α )
B 40 30 70
Then, still following Ricardo, let us assume that I produces only B, and II only A;
we will have:
I II Total quantities
A 60 60
(β )
B 80 80
The total quantity to be divided between the two people is greater for B, but it is smaller for
A, and we do not know whether, taking account of the tastes of the individuals, there is, or
there is not, compensation. . . . For example, if A is bread and B coral ornaments, it could
very well happen that the deficit of 6 bread will not be compensated by 10 more coral.
(Pareto 1906; 1971, pp. 369–370)
19.10 Pareto’s Two-Country, Two-Good,

One-Factor (Labor) Model
Let us consider a two-country, two-good, one-factor (labor) model of the economy

suggested by Pareto’s numerical example. The production possibility line9 of
Country I is
y1 = b 1 − a 1 x1 (19.9)
where y1 and x1 signify, respectively, the output of good A and that of good B. Since
(y1 = 36, x1 = 40) and (y1 = 0, x1 = 80) must satisfy (19.9), a1 = 9/10 and b1 = 72,
i.e., we have
9
y1 = 72 − x1 . (19.9)
10
Similarly the production possibility line of Country II is
y2 = b 2 − a 2 x2 (19.10)
where y2 and x2 signify, respectively, the output of good A and that of good B. Since
(y2 = 30, x2 = 30) and (y2 = 60, x2 = 0) must satisfy (19.10), a2 = 1 and b2 = 60,
i.e., we have
y2 = 60 − x2. (19.10)
Figure 19.3 shows the production possibility line of Country I, i.e., (19.9), where
good A is measured vertically and good B horizontally. The country is at autarky
at point a and incompletely specialized in good B, for example, at point b, which
9 An example of the production possibility line is line B1 A1 in Fig. 19.1.

19.10 Pareto’s Two-Country, Two-Good, One-Factor (Labor) Model 163
Fig. 19.3 Case example,

Country I
Fig. 19.4 Case example, Country II
implies that she produces larger amounts of good B than in the autarky, but still
also produces some of good A. If the terms of trade coincide with the slope of
the production possibility line of this country, she can still enjoy the consumption
indicated by point a by exporting good B and importing good A.
In Fig. 19.4, where again good A is measured vertically and good B horizontally,
first, the production possibility of Country II, i.e., (19.10), is shown by the line
ac and second, the line segment ab is transferred from Fig. 19.3. From (19.9),
therefore, its slope is
dy 9
=− . (19.11)
dx 10
Country II is at autarky at point d on line ac. If she is completely specialized in

the production of good A and exports it in exchange for good B, she can enjoy the
consumption indicated by point b. She can enjoy the gains from trade, provided that
point b is preferred to her autarky point d.
To assure the gains from trade for Country II, without introducing consumers’
preferences, point b should be located above and to the right of autarky point d in
Fig. 19.4. At point a, x = 0 and y = 60, and at point d, x = y = 30. First, therefore,
the import of x should not be less than 30, which, in view of (19.11) implies that the
export of y should be larger than 33. Second, the export of y should not be larger
than 30, which, again in view of (19.9), implies that the import of x should not be
larger than 100/3. As was pointed out by Pareto, therefore, the entire specialization
of Country I is not permissible, since it implies that the import of x of Country II is
40, much larger than 100/3.
Thus, contrary to Gandolfo’s interpretation (Gandolfo 1994, pp. 17–18), we
evaluate Pareto’s numerical example, not as an objection to the theory of com-
parative costs in general, but as a counter-example to the general use of the
assumption of complete specialization. If two countries are very different in size,
it may be impossible to assume complete specialization for these two countries to
demonstrate the theory of comparative costs. Even then, however, it is possible
to show that international trade is “Pareto” superior to autarky, provided that
incomplete specialization is permitted in the larger country.10
Acknowledgments It is my great honor and pleasure to contribute my paper to this Festschrift for
Professor Ian Steedman, with whom I have shared a common interest in the history of economics
and international trade theory. I remember, with many thanks, his review of my book (Steedman
1995). I would also like to thank an anonymous reviewer for their comments.
Bibliography
Bastable, C. F. (1900) . Theory of international trade. London: Macmillan.

Chipman, J. S. (1965). A survey of the theory of international trade: part I: the classical theory.
Econometrica, 33, 477–519.
Chipman, J. S. (1979). Mill’s ‘superstructure’: how well does it stand up? History of Political
Economy, 11, 477–500.
Dorfman, R., Samuelson, P. A., & Solow, R. M. (1958). Linear programing and economic analysis.
New York: McGraw-Hill.
Edgeworth, F. Y. (1894). The theory of international values, III. Economic Journal, 4, 424–443.
Gandolfo, G. (1994). International economics I. Berlin: Springer.
Hollander, S. (1985). The economics of John Stuart Mill. Oxford: Blackwell.
Mill, J. S. (1874). Essays on some unsettled questions on political economy. London: Longmans,
Green, Reader, and Dyer.
Mill, J. S. (1869)[1976]. Thornton on labour and its claims, in idem. Essays on economics and
society (pp. 631–668). Toronto: University of Toronto Press.
10 See, for further details, Chipman (1965, pp. 488–489).

Bibliography 165
Mill, J. S. (1871)[1909]. Preface to the 7th edn, in idem. Principles of political economy. London:
Longmans, Green and Co.
Mill, J. S. (1909). Principles of political economy. London: Longmans, Green and Co.
Mizuta, K. (2004). Economic policy and economic liberalism: Ricardo’s case of international eco-
nomics (in Japanese). The Keizaigaku, Annual Report of Economic Society, Tohoku University,
65, 489–502.
Negishi, T. (1994). The history of economics. The collected essays of Takashi Negishi, II.
Aldershot: Edward Elgar.
Negishi, T. (1998). Mill’s superstructure, how it should have been. Aoyama Journal International
Politics, Economics and Business, 42, 27–39.
Negishi, T. (2000). Economic thought from Smith to Keynes. The collected essays of Takashi
Negishi, III. Aldershot: Edward Elgar.
Negishi, T. (2001). Developments of international trade theory. Boston, MA: Kluwer Academic.
Negishi, T. (2002). How Mill should have replied to Thornton. Transactions of the Japan Academy,
57, 38–40.
Pareto, V. (1906). Manuale d’economia politica. Milano: Società editrice libraria.
Pareto, V. (1971). Manual of political economy (A. S. Schwier, Trans.). New York: Kelley.
Ricardo, D. (1951a)[1817]. On the principles of political economy and taxation. Cambridge:
Cambridge University Press.
Ricardo, D. (1951b). Pamphlets and papers. Cambridge: Cambridge University Press.
Ricardo, D. (1952). Speeches and evidence. Cambridge: Cambridge University Press.
Samuelson, P. A. (1972). The collected scientific papers of Paul A. Samuelson, III. Cambridge,
MA: MIT Press.
Steedman, I. (1971). Trade amongst growing economies. Cambridge: Cambridge University Press.
Steedman, I. (Ed.). (1979). Fundamental issues in trade theory. London: Macmillan.
Steedman, I. (1995). The Manchester School of Economic and Social Studies, LXIII, 111–112.
Steedman, I. and Metcalfe, J. S. (1973). On foreign trade. Economica Internazionale, 26, 516–528.
Thornton, W. T. (1866). A new theory of supply and demand. Fortnightly Review, 6, 420–434.
Thornton, W. T. (1869). On labour: Its wrongful claims and rightful dues, its actual present and
possible future. London: Macmillan.
possible future (2nd ed.). London: Macmillan.
About the Author
Takashi Negishi
A member of the Japan Academy, professor emeritus of University of Tokyo,
former president of the Econometric Society, and former president of the Japanese
Society for History of Economic Thought. Major publications include General
Equilibrium Theory and International Trade (1972), History of Economic Theory
(1989), and Collected Essays of Takashi Negishi, three volumes (1994 and 2000).
Professor Negishi taught and conducted research at Stanford University, the
University of Tokyo, the University of New South Wales, the University of
Minnesota, and the London School of Economics.
His major contributions range from general equilibrium theory and welfare eco-
nomics (existence, optimality, and stability), theory of monopolistic competitions
(kinked demand curves), and micro foundations of Keynesian macroeconomics to
the history of economics. For the details of these contributions, see the International
Journal of Economic Theory, volume 4, number 2, June 2008: A Special Issue on
Social Welfare, Market Equilibrium and Stability in Honor of Professor Takashi
Negishi.

and Economics 2, DOI 10.1007/978-4-431-54433-3, © Springer Japan 2014
Name Index
A F
Arrow, K. J., 141, 149 Fellner, W., 109, 114
Fischer, S., 10, 13, 21, 32, 96, 98, 100, 103
B
G
Babbage, C., 148, 149
Gandolfo, G., 25, 27, 32, 66, 79, 80, 89, 91, 164
Bastable, C. F., 36, 42, 43, 158, 159, 161, 164
Bauer, O., 55, 59
Bhagwati, J. N., 113–115, 126 H
Blanchard, O. J., 96, 98 Haberler, G., 145, 149
Blaug, M., 4, 7, 8 Harberger, A. C., 134, 136
Bloomfield, A. I., 18, 20, 145, 149 Harris, J. R., 96, 98
Boehm-Bawerk, E. V., 51, 55, 58, 59 Heckscher, E. F., 7, 8, 75, 80
Borch, K., 98 Hicks, J. R., 38, 42, 48
Hollander, S., 38, 42, 149, 164
Hume, D., 4, 9, 13
C Hutchison, T. W., 7, 8
Cantillon, R., 4, 9, 13
Chamberlin, E. H., 147–149
Chipman, J. S., 36, 38, 42, 164 I
Corden, W. M., 46, 48, 96, 98 Itoh, M., 96, 98–101, 103, 108, 113, 114
Cournot, A. A., 105, 106, 108, 114
J
Jevons, W. S., 21, 61
D Johnson, H. G., 118, 126
Dorfman, R., 164
Dornbusch, R. S., 10, 13, 21, 32, 100, 103
K
Karayiannis, A. D., 149
E Kemp, M. C., 10, 13, 45, 49, 69, 74, 120, 121,
Edgeworth, F. Y., 33, 42, 62, 66, 115, 126, 164 126, 127, 136
Ekelund, R. B., 38, 42 Keynes, J. M., 6, 8
Elmslie, B. T., 18, 20, 149 Kiyono, K., 99–101, 103
Emmanuel, A., 56, 59 Kobayashi, N., 5, 8, 16, 20
Engel, C. L. E., 10, 89 Kojima K., 27
Ethier, W., 127, 130, 136 Kojima, K., 24, 32

170 Name Index
Komiya, R., 146, 149 Picchio, A., 38, 42

Kreuger, A., 108, 114 Pigou, A. C., 147, 150
Krugman, P. R., 127, 130, 137
Kurz, H. D., 141, 149
R
Rae, J., 45
Ricardo, D., 21, 23–28, 32, 33, 53, 143, 144,
L
150, 161, 165
Lancaster, K. J., 120, 126
Kojima on, 27–29
Lavezzi, A., 149
modern interpretation, 23, 33, 37
Leontief, W., 88, 92
original Ricardo model, 23
Lerner, A. P., 99, 103
theories, 151
Lipsey, R. G., 120, 126
Richardson, G. B., 19, 20, 150
Robinson, J., 147, 150
Rybczynski, T. M., 81, 85, 86, 125
M
Man, T., 4
Marshall, A., 61, 66, 127, 129, 131, 137 S
Marx, K., 5, 8, 51, 53, 55 Salvadori, N., 141, 149
Mathews, R. C. O., 127, 137 Samuelson, P. A., 10, 13, 21, 23, 32, 66, 81, 86,
McCulloch, J. R., 4, 8 100, 103, 151–153, 164, 165
Meade, J. E., 61, 64, 66 Schumpeter, J. A., 3, 8, 16, 20, 142, 143, 150
Melvin, J., 127, 137 Shibata, H., 108, 114
Menger, C., 21, 51, 61 Shubik, M., 106, 114
Metcalfe, J. S., 165 Smith A., 15
Metzler, L. A., 82, 86 Smith, A., 3–5, 8, 9, 15–20, 33, 43, 127, 141,
Mill, J. S., 42 144, 146, 147, 149, 150
Mill, J. S., 33, 34, 36, 38, 43, 44, 49, 53, 59, Smith, M. A. M., 126
61, 66, 143, 144, 149, 151, 155–161, Solow, R. M., 164
164 Sraffa, P., 20, 132, 137
Minhas, B. S., 91, 92 Stackelberg, H. V., 109, 110, 113, 114
Mizuta, K., 25, 165 Staley, C. E., 10, 13
Morishima, M., 57, 59 Steedman, I., 165
Mundell, R. A., 45, 49 Steuart, J., 5
Mussa, M., 11, 13 Stolper, W. F., 81, 86
Myint, H., 18, 20, 145, 149 Sweezy, P. M., 56, 59, 133, 137, 148, 150
Sylos-Labini, P., 148, 150
N
Negishi, T., 17, 19, 20, 25, 32, 42, 49, 59, 98, T
120, 126, 136, 137, 147, 150, 165 Takayama, A., 64, 66
Newman, P., 62, 66 Thornton, W. T., 33, 38–40, 42, 160, 161, 165
Todaro, M. P., 96, 98
O V
Ohlin, B., 75, 80, 120, 126 Vassilakis S., 17
Ono, Y., 108, 113, 114 Vassilakis, S., 20, 147, 150
Viner, J., 10, 13
P
Pareto, V., 161, 162, 165 W
Petrella, F., 4, 8 Walras, L., 20, 21, 61, 142, 143, 150
Phelps, E. S., 123, 126 Whitaker, J. K., 20
Subject Index
A competition, 19, 105, 160

absorption, 7, 11 Sraffian, 20, 128, 131–133
accumulation Walrasian, 19, 128
capital, 153 consumer
agricultural methods preferences, 164
industry, 154 consumers’ surplus, 43, 44, 46, 48, 117, 127
agriculture, 5, 17, 25, 96 consumption, 154, 156, 158, 159, 163, 164
auction, 38, 39 of cloth and linen, 157
autarky, 22, 34, 95, 96, 127, 130 of wealth, 156
corn
cost of raising, 154
B corn law, 25
balance of trade, 6, 9, 28, 65, 93, 99, 102, 136 cost
box-diagram, 42, 73, 84 comparative, 164
C D
capital, 16, 51, 55, 69, 153 deduction, 87
accumulation, 5, 17, 23 demand, 158
free, 158 and supply, 158–160
-labor ratio, 69, 76 elasticity of demand, 156, 157
mobility, 25, 54, 56 for cloth, 158, 159
set free, 158 diminishing returns, 18, 37
stock, 153 discounted present value, 48, 55
withdrawn, 156 disequilibrium
capitalist, 153 exchange, 38, 41
CES production function, 90, 91 theory, 143, 146
cheap government, 18 division of labor, 15, 17, 146, 147, 149
classical school, 3, 5, 9, 21, 33, 42, 43, 51 domestic distortion, 93, 99
cloth sector, 151, 156, 157 duopoly, 106, 109, 113
Cobb–Douglas production function, 73, 80, 91
commodity, 154, 156, 158, 160 E
imported, 158 economic
production, 151 geography, 23
comparative advantage, 19, 21, 22, 44, 75, 153, growth, 23, 81, 115, 153
156 history, 16
comparative cost, 21, 75 policy, 5, 16

172 Subject Index
economics H
classical, 155 Heckscher–Ohlin model, 81
economy, 153 Heckscher–Ohlin theorem, 75, 88, 89
effective demand, 6 historical school, 5
effectual demand, 16 homothetic, 10, 13
efficiency wage, 96
empirical refutation, 87, 91
Engel curve, 10, 13 I
England, 151, 152, 156, 160 imported commodities, 164
income, 158, 159 income, 156
revolution, 153 distribution, 160
situation, 157 level, 156
equilibrium national, 159
classical theory, 159 increasing returns, 18, 47, 127
prices, 159 to scale, 141
terms of trade, 155, 159, 160 induction, 87
theory, 142, 146 industrial revolution, 7, 23
unique, 155 industry, 154, 158
exchange, 160, 164 change, 158
exhaustive distribution, 71, 73, 77 infant industry, 43, 45
expansion line, 71, 84 internal economy, 18, 45, 127, 131, 133
expected income, 96, 97 international trade, 142, 144, 146, 148, 164
exploitation, 51, 53, 58 theory, 161
export invisible hand, 18
-drive, 146 isoquant, 69, 73
subsidy, 98–100
J
F Japanese economy, 99, 103, 132, 134
factor endowments, 75, 79
factor intensity reversal, 89, 91
factor price equalization, 77, 78, 130 K
foreign exchange, 132, 133 Keynesian economics, 7, 133, 135
foreign investment, 6, 120 kinked demand curve, 133
foreign trade
gains, 151, 154
free trade, 18, 25, 43, 75, 108, 115, 119, 123, L
126, 154 labor, 151–156, 162
power, 51
productivity, 23, 54, 63, 153
saving, 155
G (trade) union, 39, 83
gains from competition, 108 value theory, 27, 52
gains from trade, 18, 23, 44, 53, 93, 108, 115, laboring classes, 153
127, 129 laissez-faire, 17
general equilibrium theory, 16, 61, 62, 155 land
General Theory, 6, 8 marginal land, 154
Germany, 154, 156, 157, 160 law, 153
cloth production, 159 economical, 160
English cloth, 157 of indifference, 10
national income, 159 of International Value, 155
gold, 4, 27, 31 Leontief paradox, 88, 91
goods, 160 linear homogeneity, 71, 73, 76
Subject Index 173
M producers’ surplus, 43, 46, 127

Malthusian population principle, 24 production, 156, 158, 162
manufacture, 5, 16, 25, 26, 96 factors, 152, 155
marginal land, 24, 26 function, 69, 75
marginal productivity, 23, 69, 75 increased, 155
market marginal adjustments, 151, 154
prices, 160 of linen, 160
money, 4, 6, 9, 11, 15, 28, 43, 48 of one commodity, 155
quantity theory of, 29 productivity
monopolistically competitive markets, 149 average, 153
monopoly, 105, 106 profit, 16, 23, 24, 43, 46, 53, 153
N Q
natural order of investment, 16, 17 quantity theory of money, 29
natural price, 16, 142 quota, 108, 110, 111, 113
neo-classical
economists, 161
general equilibrium theory, 16, 61, 62
school, 20, 21 R
non-traded goods, 12, 13, 103 rate of interest, 6, 51, 53
non-uniqueness, 34, 38, 41, 77 reciprocal demand, 27, 33, 37, 41, 151, 155,
159, 161
rent, 153
O capital, 70, 71, 76
offer curve, 38, 61–66 income, 153
oligopoly, 105, 108, 148 land, 16, 23
optimal saving, 123, 124 -seeking, 108
organic composition of capital, 55, 56 resources, 152
output revolution, 161
of good, 162 Ricardian model, 23, 27, 37, 100
rigid wage, 96
risk, 7, 25
P Rybczynski theorem, 81, 85, 125
Pareto efficiency, 42
partial equilibrium analysis, 61, 62
per-capita GNP, 23, 89, 102, 130 S
perfect competition, 105, 108, 127, 128 social (aggregate) indifference curve, 10, 61,
physiocracy, 17 93
political economy, 153 specialization, 151, 154–158, 160–162, 164
Political Economy Club, 38 complete, 11, 22, 25, 34, 36, 37, 78, 125,
population, 152 131
Portugal, 151–153 incomplete, 75, 80, 123, 164
pre-trade, 157, 159 specie-flow mechanism, 4, 9, 10, 27
price, 156, 160 specific factor, 63, 71, 72
higher, 160 Sraffian competition, 20, 128, 131–133
of cloth, 156 St. Petersburg paradox, 98
of food relative, 152 stability, 11, 13, 66, 136
primitive accumulation, 5, 6 of foreign exchange, 133
Principles stock, 16
Marshall, 61, 127, 137 Stolper–Samuelson theorem, 81
Mill, 33, 36, 38, 42, 53, 155, 159 subsistent wage, 24, 53, 56
Ricardo, 21, 23, 25, 26, 32, 153 supply and demand, 155
174 Subject Index
T V
tariff, 44, 81–83, 99, 108, 111, 113, 117 value
technology, 69, 75 destruction, 160
terms of trade, 10, 23, 27, 30, 31, 33, 34, 42, international, 155, 156
61, 65, 97, 99, 100, 115, 155, 159 variable capital, 51
trade, 151, 154–158, 160 vent for surplus, 18
gains, 164
trade indifference curve, 64, 65, 93, 99
trading countries, 155 W
transformation curve, 64, 72, 73, 93, 95 wage, 16, 23, 39, 51, 70, 71, 77, 94–96, 130,
two-good model, 156, 162 131, 153
wage–rental ratio, 69, 70
wages fund, 39
U Walrasian competition, 19, 128, 131, 132
under-development, 5, 18 Wealth of Nations, 3, 8, 15, 16, 19, 141–144,
unemployment, 6, 7, 93, 95, 96 146, 147, 149
urban unemployment, 96 world economy, 160

(Advances in Japanese Business and Economics 2) Takashi Negishi (Auth.) - Developments of International Trade Theory-Springer Japan (2014)

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

(Advances in Japanese Business and Economics 2) Takashi Negishi (Auth.) - Developments of International Trade Theory-Springer Japan (2014)

Загружено:

Авторское право:

Доступные форматы

Advances in Japanese Business and Economics 2

For further volumes:

Editorial Board Members:

ISSN 2197-8859 ISSN 2197-8867 (electronic)

© Springer Japan 2001, 2014

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

This book is based on my lectures on international trade in 1993–2000 at the School

January 2001 Takashi Negishi

It is my pleasure to be able to publish the second enhanced edition of this book.

Tokyo, Japan Takashi Negishi

Part I Historical Progress of International Trade Theory

8 Marshall and Offer Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 61

Part II Modern Theory and Recent Developments

Part III Historical Appendix

18 Adam Smith and Disequilibrium Economic Theory.. . . . . . . . . . . . . . . . . . . 141

Name Index .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 169

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 171

According to Schumpeter,1 the construction of scientific economics was started in

T. Negishi, Developments of International Trade Theory, Advances in Japanese Business 3

If we emphasize mercantilists’ wrong view that the wealth of a country consists

where Y and M denote, respectively, the domestic production of commodities and

1.1. Mercantilists considered that the supply of money in a country is directly

Blaug, M. (1985). Economic theory in retrospect. Cambridge: Cambridge University Press.

T. Negishi, Developments of International Trade Theory, Advances in Japanese Business 9

Let us construct a drastically simplified version of the model used by Dornbusch,

Fig. 2.1 The specie-flow mechanism

solution of (2.3). If M > B = wL/V , M decreases through time and if M < B =

2.1. Suppose the utility function of a consumer is

T. Negishi, Developments of International Trade Theory, Advances in Japanese Business 15

1 Smith defined capital as the stock which is used to obtain profit.

or factor endowments are important to explain the comparative advantage.4 Trade

argued by Negishi (2000).

1 Revolution made in 1870s by W. S. Jevons (1835–1882) in Manchester, C. Menger (1840–1921)

T. Negishi, Developments of International Trade Theory, Advances in Japanese Business 21

Fig. 4.1 The situation of

interpreters conclude, Ricardian theory of international trade is an incomplete theory

of capital in the former. Competition between capitals induces capitals in England

5I owe this point to Professor K. Mizuta of Higashi Nihon Kokusai University.

4.2 Appendix: Kojima on Ricardo

Table 4.1 Labor cost of

Table 4.2 Natural price

7 Table 4.2 corresponds to Table 3 in Kojima (1951).

Table 4.3 Labor cost of

Table 4.4 Natural price

Suppose first that gold is produced only in Portugal.10 Then,

and from (4.2), (4.3), and (4.4)

(agp v p G − L p )ve ace

P (agp v p G − L p)ve ace

without introducing reciprocal demands of cloth and wine.

must be satisfied. In view of (4.5), then, this condition requires that

(agp v p G − L p)ve agp 12

J. S. Mill (1806–1873)s Principles of Political Economy (1848) was written as

T. Negishi, Developments of International Trade Theory, Advances in Japanese Business 33

Fig. 5.1 Germany

Fig. 5.2 England

where B1 is the maximum quantity of linen Germany can produce, as is shown

5.1. Consider the maximization of utility function U = a log x + (1 − a) log y, being

5.2 Appendix: Mill and Thornton

The so-called Mill’s superstructure in Chap. 18 of his Principles is constructed so

If Thornton (1866) is ever concerned with the non-uniqueness of equilibrium

a(B1 + tA2) = A2t (5.4)

where a is the constant proportion in which the expenditure is devoted to cloth