D E V E L O P M E N T E C O N O M I C S

This page intentionally left blank

DEVELOPMENTECONOMICSFrom the Poverty to the Wealth of Nations

Third Edition

YUJIRO HAYAMI

Y O S H I H I S A GODO

OXPORDUNIVERSITY PRESS

OXPORDUNIVERSITY PRESS

Great Clarendon Street, Oxford 0X2 6DPOxford University Press is a department of the University of Oxford.It furthers the University's objective of excellence in research, scholarship,and education by publishing worldwide inOxford New YorkAuckland Bangkok Buenos Aires Cape Town ChennaiDar es Salaam Delhi Hong Kong Istanbul Karachi KolkataKuala Lumpur Madrid Melbourne Mexico City Mumbai NairobiSao Paulo Shanghai Taipei Tokyo TorontoOxford is a registered trade mark of Oxford University Pressin the UK and in certain other countriesPublished in the United Statesby Oxford University Press Inc., New York Yujiro Hayami and Yoshihisa Godo, 2005The moral rights of the author have been assertedDatabase right Oxford University Press (maker)First published 2005All rights reserved. No part of this publication may be reproduced,stored in a retrieval system, or transmitted, in any form or by any means,without the prior permission in writing of Oxford University Press,or as expressly permitted by law, or under terms agreed with the appropriatereprographics rights organization. Enquiries concerning reproductionoutside the scope of the above should be sent to the Rights Department,Oxford University Press, at the address aboveYou must not circulate this book in any other binding or coverand you must impose this same condition on any acquirerBritish Library Cataloguing in Publication DataData availableLibrary of Congress Cataloging in Publication DataData availableISBN 0-19-927270-0 (hbk.)ISBN 0-19-927271-9 (pbk.)1 3 5 7 9 10 8 6 4 2Typeset by Newgen Imaging Systems (P) Ltd., Chennai, India.Printed in Great Britainon acid-free paper byBiddies Ltd., King's Lynn, Norfolk.

Preface to the Third Edition

The first edition in 1997 of this book, single authored by Yujiro Hayami, was atranslation (with revisions) from a Japanese version under the title KaihatsuKeizaigaku published by the Sobunsha Publishing Company in Tokyo in1995, which was later translated into Chinese and published by the SocialScience Documents Publishing House in Beijing. The second edition in 2001was also first published in Japanese in 2000. In contrast, this edition jointlyauthored by Yujiro Hayami and Yoshihisa Godo was prepared in English foran international audience from the beginning.

This edition aims to render a perspective on the problems in developingeconomies in the new millennium. For this goal, most data are updated to2000 or more recent years wherever possible, while the previous edition used1995 as the baseline for data comparisons across countries. In particular,Chapter 2 is completely restructured with the new data set.

During the decade centring in 2000, a major change occurred in thecurrent of development thought. At the time the second edition was prepared,international development assistance policies were still dominated by thevoice of economists in the International Monetary Fund, the World Bankand the US Department of Treasury advocating the use of free markets forthe development of developing economiesthe so-called 'WashingtonConsensus.' In the less than ten years which followed, however, this view waslargely replaced by the so-called 'Post-Washington Consensus' advocatinggreater roles for the public sector and civil society in reducing poverty. Thisprocess was outlined in Chapter 8 (Sections 8.5 and 8.6). A major factorunderlying this paradigm change was the rising emphasis on poverty reduc-tion as the direct objective of development policies, as epitomized by theUnited Nations' Millennium Development Goals. In the previous edition,issues concerning poverty were discussed in Chapter 7 as a part of the problemof income distribution. In this edition, however, the measurement and ana-lysis of poverty in relation to economic development are treated moresquarely, with the title of Chapter 7 changed from 'Income Distribution andEnvironmental Problems' to 'Income Distribution, Poverty, and Environ-mental Problems.'

Further, in response to comments from several instructors and students whoused the previous edition as a text, two appendices are added: (B) on thePigou theorem of equivalence between tax and subsidy in removing negativeexternality and (C) on the theory of agricultural land tenure choice.

vi Preface to the Third Edition

The preparation of this edition was supported by the Foundation forAdvanced Studies on International Development (FASID), Tokyo. We wouldlike to express sincere appreciation for both financial/logistic support fromthe FASID administration and academic input from the members of the FASIDgraduate faculty including Keijiro Otsuka, Tetsushi Sonobe, KaliappaKalirajan, Debin Ma, Kei Kajisa, Takashi Yamano, and Futoshi Yamauchi.Also invaluable were the comments on the previous editions as well as on newmaterials that were received from Robert Allen, Masahiko Aoki, KennethArrrow, Randolph Barker, Kaushik Basu, Abdul Bayes, Partha Dasgupta,Robert Evenson, Avner Greif, Shigeki Hakamada, Koichi Hamada, RobertHerdt, Mahabub Hossain, Hall Hill, Takashi Inoguchi, Takenori Inoki, ShigeruIshikawa, Bruce Johnston, Michael Kevane, Masao Kikuchi, Taejong Kim,Hirohisa Kohama, Ryutaro Komiya, Takashi Kurosaki, Laurence J. Lau, JustinLin, Masahiro Matsushita, Ryosin Minami, Watsuji Nakagane, TakashiNegishi, Douglas North, Masahiro Okuno-Fujiwara, Elinor Ostrom, Jean-Philippe Platteau, Agnes Quisumbing, Gustav Ranis, Vernon Ruttan, YasuyukiSawada, T. N. Srinivasan, Paul Streeten, Akira Suehiro, Juro Teranishi, ErikThorbecke, Henry Wan, Jr., and Yasukichi Yasuba. Technical assistance fromSuzanne Akiyama and Yue Yaguchi is gratefully acknowledged.

The biggest impact on us that may have made this edition distinct from theprevious edition has come from the students in the graduate programme oninternational development studies organized jointly by FASID and theNational Graduate Institute of Policy Studies (GRIPS). As our students aremainly sent from development agencies (including NGOs) in Africa and Asiaas well as Japan, their motivation for mastering development economics isextremely high. Intensive interactions with them for the past four years haveconstantly forced us to try to make this volume a truly useful guide for thedesign of development policies for their nations in the future. In gratitude forthe stimulus received from them, this edition is dedicated to the students, bothpresent and past, in the FASID-GRIPS joint graduate programme.

Yujiro Hayami and Yoshihisa GodoApril 2004

Contents

Detailed Contents viiiList of Figures xvList of Tables xvii

Introduction 11. A Theoretical Framework for Economic Development 92. A Comparative Perspective on Developing Economies 313. Population Growth and the Constraint of Natural Resources 634. Breaking the Natural Resource Constraint 925. Capital Accumulation in Economic Development 1226. Patterns and Sources of Technological Progress 1617. Income Distribution, Poverty, and Environmental Problems 1918. Market and State 2429. The Role of Community in Economic Development 310

10. Tradition and Modernization: A Concluding Remark 349

Appendices 362Bibliography 383Index of Names 415Index of Subjects 421

Detailed Contents

List of FiguresList of Tables

IntroductionScope of development economicsOrganization of the book

1. A Theoretical Framework for Economic Development1.1 Development of the Social System

1.1.1 A model of dialectic social development1.1.2 A historical example1.1.3 Marx and new institutionalism

1.2 The Theory of Induced Innovation*1.2.1 Induced technological innovation1.2.2 Induced institutional innovation1.2.3 Logic of political market1.2.4 Historical path dependency

1.3 Developing Economies in the Light of the TheoreticalFramework

2. A Comparative Perspective on Developing Economies2.1 Economic Growth and Structural Change

2.1.1 Per capita GDP and its growth2.1.2 Changes in industrial structure

2.2 Sructure of Capital Accumulation2.2.1 Capital formation and savings in economic growth2.2.2 External debt and inflation

2.3 Accumulation of Human Capital2.3.1 Measurement of human capital2.3.2 Human capital investment and economic growth

2.4 Population, Natural Resources, and Foods2.4.1 Population pressure on natural resources2.4.2 Population growth vs. food supply

* General readers not interested in technical detail may wish to skip sections marked with anasterisk (*)

xv

xvii

125

9

99

12141616202125

27

31323237

424345

495051545456

Detailed Contents ix

3. Population Growth and the Constraint of Natural Resources3.1 Population Growth in Economic Development

3.1.1 Historical changes in world population3.1.2 Demographic transition3.1.3 The case of India

3.2 Economic Theories of Population Growth3.2.1 The Malthus model3.2.2 The household utility maximization model*

3.3 Theories of Resource Constraint on Economic Growth3.3.1 From Malthus to the Club of Rome3.3.2 The Ricardo model*3.3.3 The dual economy model*

4. Breaking the Natural Resource Constraint4.1 Potential of Science-Based Agriculture4.2 A Perspective on the Green Revolution

4.2.1 Development and diffusion of modern varieties4.2.2 Conditions of technology transfer4.2.3 External and internal land augmentation

4.3 Barriers to Induced Innovation4.3.1 Problems in Africa4.3.2 Whither the Green Revolution?

4.4 Development via Natural Resource Slack4.4.1 Colonialism and the vent-for-surplus theory4.4.2 The staple theory4.4.3 The Dutch disease

5. Capital Accumulation in Economic Development5.1 From Adam Smith to Marx

5.1.1 Capital in Adam Smith5.1.2 Ricardo revisited5.1.3 The Marx model of capitalist development*5.1.4 The Marx model and the efficiency wage theory*

5.2 Development Theories and Policies after World War II5.2.1 The theory of balanced growth5.2.2 Application of the Harrod-Domar model*5.2.3 The model of low-equilibrium trap*5.2.4 Development theories and policy choice

636364677073

737478788085

92

92

969799104

107109111

115116117119

122

123123125126131

133134135136138

x Detailed Contents

5.3 Neoclassical Production Function and Growth Model5.3.1 Different assumptions of the production function5.3.2 The Solow-Swan model*

5.4 Growth Accounting Test5.4.1 The growth-accounting equation5.4.2 Sources of modern economic growth

5.5 Changes in the Pattern of Economic Growth5.5.1 A historical extension of growth-accounting5.5.2 A trap in the Marx-type growth

6. Patterns and Sources of Technological Progress6.1 The Marx vs. the Kuznets Pattern of Economic Growth

6.1.1 Stylization of the two patterns6.1.2 Trends in the rates of saving, interest, and wages

6.2 Technological Conditions of the Two Growth Patterns6.2.1 The shift in the industrial technology regime6.2.2 The shift in the demand structure6.2.3 Borrowed technology and the Marx-type growth

6.3 Searching for the Sources of Technological Progress6.3.1 Accounting for TFP growth*6.3.2 Schooling and economic growth6.3.3 Increasing returns and the endogenous growth model*6.3.4 Schumpeter and centrally planned economies6.3.5 Institutional conditions of borrowing technology

7. Income Distribution, Poverty, and Environmental Problems7.1 Inequality and Poverty

7.1.1 Concepts and measurement of income distribution7.1.2 Concepts and measurement of poverty7.1.3 Patterns of changes in inequality and poverty

7.2 Causes of Inequality7.2.1 Changes in factor shares7.2.2 The dual economic structure7.2.3 Agriculture-non-agriculture income differential7.2.4 On the redistribution of incomes and assets

7.3 Economic Stagnation and Poverty7.3.1 Income distribution effects of the Green Revolution7.3.2 A comparison of two villages in Indonesia

139139141

145145148

151152155

161

161162164

168168169171

173173176181184188

191

191191195198

209209210211213

216217220

Detailed Contents xi

7.4 Environmental Problems in Economic Development7.4.1 The core of environmental problems7.4.2 Rural poverty and environmental destruction7.4.3 Industrialization and environmental pollution7.4.4 Lowering the peak of the inverted-U-shape curve7.4.5 Towards global coordination

8. Market and State8.1 The Economic Functions of the Market and the State

8.1.1 Efficiency of the competitive market8.1.2 Market failure8.1.3 Government failure8.1.4 On the choice of economic system

8.2 Around the Infant Industry Protection Argument8.2.1 Market failure in dynamic economy8.2.2 Ricardo vs. List8.2.3 The Listian trap8.2.4 The import-substitution industrialization policy

8.3 The Rise and Fall of Developmentalist Models8.3.1 The limit of information and the role of ideology8.3.2 Defeat of the old developmental market economies8.3.3 Collapse of the centrally planned economies8.3.4 Trap of populism

8.4 Success and Failure of the New DevelopmentalMarket Economies8.4.1 The system of new developmental market economies8.4.2 The source of success8.4.3 Beyond achieving the catch-up goal

8.5 Resurgence of Market Liberalism and its Consequences8.5.1 The structural adjustment policy of the IMF and

the World Bank8.5.2 Recurrent crises in Latin America8.5.3 Financial crisis in East Asia

8.6 From the Washington Consensus to thePost-Washington Consensus8.6.1 Criticisms of the Washington Consensus8.6.2 Poverty reduction as an immediate objective8.6.3 The post-Washington Consensus prospect

223224226228232235

242

242244245247249

250251252253254

257258259263265

268268270277

280

280282288

295295298303

xii Detailed Contents

9. The Role of Community in Economic Development9.1 The Economic Functions of Community

9.1.1 Prisoner's dilemma9.1.2 Trust as a social capital9.1.3 Supply of local public goods

9.2 Rural Organization in Developing Economies9.2.1 Dominance of peasants9.2.2 Management of common-property resources9.2.3 Landlord-tenant relations

9.3 Economic Rationality in Community: A Perspective fromPhilippine Villages9.3.1 Labour hiring by peasants9.3.2 Income and work-sharing9.3.3 Changes in the sharing system9.3.4 The role of community norm9.3.5 Egoism and altruism

9.4 The Community in Market Development9.4.1 Ethnic networks and guilds9.4.2 From the putting-out to the modern

subcontracting system9.4.3 Overcoming the community failure

9.5 Towards an Optimal Combination of the Community,the Market, and the State

10. Tradition and Modernization: A Concluding Remark10.1 Institutional Innovation for Technology Borrowing10.2 The Experience of Japan10.3 Multiple Paths to Economic Modernization

Appendix A: Theoretical Supplements to Technological ProgressA. 1 Increases in the capital-labour ratio and shifts in

production functionA.2 The classification of technological changeA.3 Changes in the trends of factor prices and

factor sharesA.4 Possibilities for induced innovationA.5 Interpretation by the meta-production functionA.6 Mathematical analysis of changes in factor shares

Appendix B: The Pigou Theorem on Equivalence betweenTax and Subsidy in Removing Externality

310311312313316317317321324

328329331332335338

339340

341343

346

349349351355

362

362363

367368370372

374

Detailed Contents xiii

Appendix C: Theories on the Choice of Land Tenure System

Bibliography

Index of NamesIndex of Subjects

378

383

415421

This page intentionally left blank

List of Figures

1.1 Interrelated developments in the social system 101.2 A model of induced innovation 181.3 A model of political market for a public good 232.1 International comparison between per capita GDPs in current US dollars

converted by exchange rate and purchasing power parity, 2000 352.2 International comparison of the average annual growth rates of

GDP (converted by exchange rate, 1965-2000 averages) andratio of capital formation to GDP (1965-2000 averages) 45

2.3 Changes in GDP per capita and I-S gap from the 1965-80averages to the 1981-2000 averages 46

2.4 International comparison of the average number of yearsof schooling and average life expectancies at birth, 2000 52

2.5 International comparison of the average annual growth ratesof per capita GDP from 1965 to 2000: and (a) increases inthe average number of years of schooling from 1965 to2000; (b) the average life expectancies at birth from 1965 to 2000 53

2.6 International comparison of the average annual growth rates ofper capita GDP from 1965 to 2000 and percentage increases infood production per capita from 1965 to 2000 57

3.1 Changes in the birth- and death-rates in the UK, 1750-1970,nine-year moving averages 68

3.2 Changes in the birth- and death-rates in India, 1901-2000,ten-year averages 71

3.3 The Malthusian population theory and its revision 743.4 A household utility maximization model on the determination of

the number of children 753.5 The Ricardo model of economic development 833.6 The dual economy model of the Lewis-Ranis-Fei type 874.1 Long-term changes in real prices (deflated by 1967-standard CPI)

and yield per hectare of corn and wheat in the USA 954.2 Increases in paddy yield per hectare corresponding to diffusion

of the modern varieties with different characteristics ofresistance to brown planthopper Biotypes I and II 99

4.3 Changes in rice yield per hectare (in brown rice) in Japan,Taiwan, and Korea, five-year moving averages, semi-log scale 101

4.4 Changes in farmland area per worker, percentages in areaimproved by land infrastructure development projects, andarea planted to improved rice varieties in Japan and the Philippines 103

xvi List of Figures4.5 Relationship between marginal costs of agricultural production

from new land openings and from irrigation construction4.6 Paddy yields per hectare harvested in selected Asian countries,

1953-2000, five-year moving averages4.7 Rice price and public investment in irrigation, Philippines and

Sri Lanka, 1960-985.1 The Marx model of capitalist economic development5.2 The model of low-equilibrium trap5.3 Comparison between the Solow-Swan model and

the Harrod-Domar model6.1 Movements in the ratios of domestic saving to GDP and

national saving to GNI in Japan, 1883-2001, five-yearmoving averages

6.2 Movements in the real rate of interest in Japan, 1882-20026.3 Movements in the real wage rate in Japan (1934-36 = 100),

1886-2002, seven-year averages, semi-log scale6.4. The Japan/USA and Korea/USA ratios in average schooling,

per capita GDP, and capital-labour ratio7.1 Lorenz curves for Bangladesh, Brazil, and Japan7.2 International comparison of the Gini coefficients7.3 International comparison of absolute poverty7.4 GDP growth and poverty indexes in Thailand, 1962-20017.5 Cumulative percentage of farms in three size classes adopting

modern varieties and tractors in thirty villages in Asia7.6 International comparison of carbon dioxide emission9.1 The pay-off matrix of the prisoner's dilemma gameA. 1 Elements of growth in labour productivityA.2 Classifications of technological progress and substitution between

labour and capitalA.3 Income shares of labour and capitalA.4 Possibilities for induced technological innovationA. 5 Factor substitution along meta-production functionB.1 The Pigou model on the equivalence between tax and subsidy in

achieving social optimality under externalityC.1 Model of land tenancy

105

112

115127137

143

165166

167

178194201205208

219229312363

364366369371

374379

List of Tables

2.1 Major development indicators in selected economies2.2 Changes in the sectoral shares of GDP in selected economies2.3 Changes in the structure of merchandise export and the

competitive performance of industry in selected economies2.4 Investment, saving, external debt, and inflation in

selected economies2.5 Improvements in education and health in selected economies2.6 Population, land, and food production in selected economies3.1 World population, 1000-20503.2 Population in India, 1871-20015.1 Growth rates of output, input, and productivity in

selected developed countries5.2 Accounting for long-term growth in labour productivity in

the USA and Japan5.3 Accounting for growth in labour productivity in the

former Soviet Union5.4 Comparisons in the growth rates of labour productivity and

TFP between newly industrializing economies (NIEs) anddeveloped industrial economies

6.1 Stylized facts in the two phases of modern economic growth6.2 Sources of growth in national income per person employed in

selected developed economies7.1 Cumulative shares of household incomes, by quintile

class of households, Bangladesh, Brazil, and Japan7.2 Estimates of the Gini coefficient in Japan, 1890-19987.3 Estimates of regression equations to explain the Gini coefficients7.4 Estimates of regression equations to explain absolute poverty7.5 Historical changes in agriculture-manufacturing relative

labour productivity, agriculture-industry terms of trade,farm-non-farm household relative income in Japan, 1885-2000

7.6 Mexican wheat acreage as percentage of all wheat acreage bysize and tenure of holdings: 1969-70 post-monsoon season inLyallpur, Sahiwal, and Sheikhupura districts, Pakistan

7.7 Economic changes in a survey village in Indonesia in whichmodern rice varieties failed to be adopted, 1968-71 to 1978

7.8 Economic changes in a survey village in Indonesia in whichmodern varieties of rice were successfully adopted, 1968-71 to 1978

7.9 Estimates of regression equations to explain carbon dioxide emission

3339

40

4451556571

150

152

155

157162

174

194200202207

212

218

221

222230

xviii List of Tables9.1 Labour inputs per hectare of rice crop area in Japan,

Philippines, and Indonesia9.2 Comparisons between the actual revenue of harvesters and

the imputed cost of harvesting labour under the hunusan andthe gama contracts in a survey village (Village E) in thePhilippines, 1976 wet season

9.3 The cost-return structures of rice production, estates vs. peasants inVillage W, Philippines, 1977 dry season

C. 1 Orders of magnitudes in risk, contract enforcement cost andenterpreneurial opportunity associated with alternative contracts

330

334

337

381

Introduction

The world today is characterized by extremely large income inequality amongcountries. According to the World Bank's World Bank Indicators 2003,average per capita income in 2000 ranged from the level exceedingUS$ 25,000 in high-income countries belonging to the Organization forEconomic Cooperation and Development (OECD), to the meagre level of onlyUS$ 280 among the least developed countries (39 countries in Sub-SaharanAfrica according to the United Nations' definition).

In that year, the total population amounted to 6.1 billion, of which thepopulation in high-income countries with per capita income above US$ 9,200numbered only 950 million. Yet this 16 per cent of the global populationreceived more than 80 per cent of world income. In contrast, 2.5 billionpeople, or nearly 40 per cent of world population, in low-income countrieswith per capita income below US$ 750 were entitled to only about 3 per centof the world income.

These per capita income comparisons are made in terms of the UnitedNations' estimates of gross national income (GNI) converted to US dollarsusing the official exchange rates. Such comparisons tend to underestimate thelevel of economic welfare being enjoyed by people in low-income relative tothose of high-income economies because of differences between exchangerates and purchasing power parities, as well as incomplete enumeration ofnon-market goods and services in GNI statistics. Yet, even after this statisticalbias is corrected, it is certain that an extremely wide gap in the levels of realincome and living remains between low-income and high-income countries,though the gap might be reduced from an order of one to a hundred to anorder of one to several tens.

In addition, there are many indicators other than national income statisticsto show poverty and destitution in low-income economies. For example, theUnited Nations' Food and Agriculture Organization (FAO) estimates thatchronically undernourished people in 1997-99 amounted to 815 million orabout one-quarter of population in low-income countries, as high as one-third in Sub-Saharan Africa. Another indication is the high infant mortalityrate, with as many as 105 out of 1,000 newly born babies dying beforereaching age 1 in Sub-Saharan Africa in 2001, in contrast to only 5 in high-income OECD countries.

The escape from such destitution and misery through economic develop-ment must be the common national goal of low-income countries. Indeed,

2 Introduction

developing countries that achieved independence after World War II havealmost unanimously undertaken ambitious development programmes aimedat catching up with high-income economies. Several success stories havebeen recorded. Especially remarkable are the so-called 'Newly IndustrializedEconomies' (NIEs) in Asia, such as Korea, Taiwan, Hong Kong, and Singapore.Starting the early post-World War II period with per capita income levels notmuch different from those of low-income countries today, these NIEs havenow joined the ranks of high-income economies. Following the NIEs, severaleconomies in East Asia, including China, Malaysia, Thailand, and Vietnamhave been growing much faster than high-income countries. However, therates of growth in low-income economies, especially in Sub-Saharan Africa,have been lower than in high-income economies, with the result beingwidening worldwide differentials in per capita income.

It should not be difficult to imagine how such growing inequality in theworld economy has been exacerbating tension in international relations. Forabout four decades after World War II, the confrontation between the North(high-income developed economies) and the South (low-income developingeconomies) represented one of the two major axes for mapping internationalrelations, together with the confrontation between the West (capitalist marketeconomies) and the East (socialist centrally planned economies). Since theend of the cold war, the global confrontation between two superpowers inthe East and the West has been replaced by multidimensional ethnic andlocal conflicts involving civil wars and terrorism. These relatively small butnumerous and pervasive conflicts, if amplified by growing internationaleconomic disparity, will likely result in major instability in the world politicalsystem. Emancipation of people in developing countries from poverty is,therefore, not only desirable on humanitarian grounds but also necessary fordeveloped countries whose peace and prosperity hinge critically on the sta-bility of the international order.

Scope of development economics

The major task of development economics is to explore the possibility ofemancipation from poverty for developing economies. It should be stronglyfocused on low-income developing countries where poverty is especiallyacute. How can low-income economies in the world today be set on the trackof sustained economic development for the immediate goal of reducingpoverty and the long-run goal of catching up to the wealth of developedeconomies? The ultimate goal of development economics is to obtain ananswer to this question.

Introduction 3

In order to achieve this goal, it is of course necessary to understand thestructure and mechanism of low-income economies. However, the char-acteristics of low-income economies cannot be properly understood withoutcomparisons with those of high-income economies. A key to identifying thecauses of poverty and stagnation in low-income economies may be found inthe experiences of economies that escaped from the same trap. It was throughthe process of economic development over a 200-year period since theIndustrial Revolution that the majority of people in developed countries in theWest were emancipated from poverty. The process was shortened to less than100 years in Japan, and to less than forty years in Asian NIEs.

An effective theory of development economics should be based onunderstanding the similarities and differences of these histories comparedwith current situations in low-income economies. For this understanding it isvital to learn the theories of economic development by great economists in thepast, who aimed to identify effective policies to promote and sustain devel-opment in their ages. Indeed, 'an inquiry into the nature and the causes of thewealth of nations' (Adam Smith, 1776) is equivalent to the inquiry into thecauses of poverty and underdevelopment.

While it is critically important to learn from the experience of successfuldevelopment, it is equally useful to learn from cases of failure. A dramaticexample in our day was the recent collapse of centrally planned economies,which until only three decades ago had been considered by many to representan effective model for developing economies to catch up and even surpassadvanced market economies. Identifying the factors underlying both thefailure of centrally planned economies as well as the relative stagnation ofsome developing economies that tried to adopt the central planning model,would be a vital step towards understanding the sustainable developmentmechanism.

It is relatively common to distinguish the term 'economic development'from 'economic growth,' though they are used interchangeably in some cases.'Economic growth' has a connotation of quantitative expansions in economicvariables, especially aggregate and per capita national incomes as measuredby such statistics as GDP and GNI. Therefore, the analysis of economic growthis concerned mainly with measuring growth in economic variables andidentifying their interrelationships such as between the national incomegrowth rate and the speed of capital formation.

On the other hand, 'economic development' is usually conceived as aprocess involving not only quantitative expansions but also changes in non-quantitative factors such as institutions, organizations, and culture underwhich economies operate. If we follow this usage, economic growth is

4 Introduction

considered a quantitative aspect of economic development. If so, in additionto the analysis of economic growth, the study of economic development mustinvestigate the influences of institutional and cultural factors on economicgrowth as well as the impacts of economic growth on those factors.

Since this book is focused on the development of low-income economiestowards catching up with high-income economies, the range of eco-nomic growth concerned is so wide that major cultural and social changes arenecessarily involved. Thus, it is inevitable that this book is intended to be atreatise of economic development as its title suggests. To be effective, how-ever, development economics must incorporate the achievements of eco-nomic growth analysis to the maximum extent.

Among the many issues and subjects pertaining to development economics,this book is strongly focused on the role of technology borrowing as a majormeans for low-income economies to catch up with advanced ones. A criticalcondition for the transfer of foreign technology is development of appropriateinstitutions. For new institutions to function effectively, they must be con-sistent with people's value system in the recipient economy. Thus, a majoragenda of this book is to investigate the potential of developing economiesendowed with different social and cultural heritages to achieve institutionalinnovations needed for effective technology borrowing. The overall aim is toidentify possible means to facilitate this process.

An equally strong focus is placed on the choice of economic system fordevelopment. In this book, this issue is posed as a question of what would be theoptimum combination of market, state, and community. These three organ-izations coordinate the division of labour among peoplethe market by meansof competition, the state by means of coercion, and the community by means ofcooperation. They have both merits and demerits in coordinating people'seconomic activities in a socially desirable direction. How to combine market,state, and community in the economic system for maximizing growth in socialproductivity, under the unique cultural and institutional conditions in eacheconomy will be the ultimate question addressed by our investigation.

While special focus is placed on low-income economies, the book coversbroadly'developing economies' at various stages of development. However, noconsensus exists on the definition of 'developing economies'. Until recently, acommon practice was to classify as developing economies all countries otherthan OECD members, high-income oil exporters, and centrally planned eco-nomies in Eastern Europe and the Soviet Union, while it was customary toinclude the centrally planned economies in East Asia such as China andVietnam in this category as well. Since the collapse of the socialist bloc,ex-socialist economies in Eastern Europe and Soviet Union are now also often

Introduction 5

classified as developing economies. For more specific analyses, the UnitedNations' classification of 'low-income economies' (with per capita GNI lessthan US$ 745), 'lower-middle-income economies' (with per capita GNIbetween US$ 745 and US$ 2,975), 'upper-middle-income economies' (withper capita GNI between US$ 2,975 and US$9,206) in 2001 will be used(World Development Indicators 2003).

On the other hand, 'high-income economies' (more than US$ 9,206) in theUN definition include not only OECD members but also high-income oilexporters (such as United Arab Emirates, Kuwait, and Brunei) and others(such as Hong Kong, Israel, and Singapore) in 2002. Yet, in this book the term'high-income economies' is used to represent the countries that joined OECDbefore 1995 (except Greece, Turkey, and Mexico). Also it is used inter-changeably with 'developed economies' and 'advanced economies.'

Organization of the bookThis book is organized in the following manner. Chapter 1 aims to establish atheoretical framework for the whole volume. As a basic framework, devel-opment of the social system is considered as a process of interactions betweenthe economic subsystem and the cultural-institutional subsystem. The eco-nomic subsystem consists of activities combining economic resources (labour,capital, and natural resources) through technology to produce goods andservices useful for human living. These activities expand through accumu-lation of resources and progress in technology to result in economic growth.People's economic activities are coordinated and controlled by institutions(which here means the rules of society) and culture (which represents people'svalue system). As relative endowments of economic resources changeforexample when natural resources like land become scarcer relative to labourowing to population growthnew agricultural technology may be required tosave land relative to labour. For this technology to be developed and adopted,a new set of institutions may become necessary. A model is developed toconceptualize how such technological and institutional changes are inter-related with each other, how they respond to changes in resource endow-ments, and how such responses are governed by cultural traditions.

Chapter 2 tries to develop a bird's-eye view on the current status andgrowth potential of developing economies by means of highly condensedinternational comparative statistics in order to postulate broad hypotheses forthe analyses in the subsequent chapters. The development pattern thus drawnis far different from that of the growth stage theories a la Rostow (1966) in

6 Introduction

which countries are supposed to advance linearly to higher developmentalstages according to the order and sequence of their economic 'take-offs'. Adramatic contrast to Rostow's model is that per capita income in Argentina,which used to be one of the wealthiest nations in the period immediately afterWorld War II, has recently been surpassed by that of Korea, which rankedamong the poorest in the early post-war years. Similar examples are abundantif not quite so dramatic. It is evident that an apparent 'take-off does notguarantee sustained growth. It is also clear that wide differences in economicgrowth rates among developing countries are due not so much to differencesin natural resource endowments, but may instead be explained mainly byinvestment in both physical and human capital. Data seem to support ahypothesis that the magnitude of such broadly defined capital formation doesnot depend very much on the level of per capita income. If so, it should bereasonable to hypothesize that even poverty-stricken economies can be set onthe track of rapid economic development depending on the policies adopted.

Chapters 3 and 4 analyse the effects of explosive population growth andresulting relative scarcity in natural resources in the low-income economiesthat are characterized by high dependency on the production and export ofprimary commodities. Chapter 3 tries to identify the causes of 'populationexplosion' in developing countries after World War II in the light of demo-graphic histories in both developed and developing economies in order todraw future predictions. Further, development theories by classical econo-mists such as Malthus and Ricardo who incorporated population as anendogenous variable in the economic system relative to fixed naturalresources are examined to draw implications for developing economies today.Chapter 4 identifies the shift from resource-based to science-based agricultureas the basic force that prevented dismal predictions by Malthus and Ricardofrom being realized. This chapter investigates the process in which themechanism of science-based agriculture has now been transferred to devel-oping economies and how such process can be promoted and sustained. Asconcluded in the chapter, it is no longer possible today to sustain economicdevelopment through nineteenth-century-type natural resource exploitation,and any resource-rich economy is bound to stagnate in poverty without majorefforts to improve natural resource conservation and utilization efficiencies.

Chapters 5 and 6 examine the roles of capital accumulation and techno-logical progress in industrial development. Chapter 5 traces the major currentsin development thought and ideology after World War II which have resultedin the adoption by many newly independent nations of a strategy gearedtowards maximizing capital accumulation in the industrial sector by means ofgovernment planning and command. This strategy tends to consider 'capital'synonymous with large-scale machinery and equipment embodying modern

Introduction 7

labour-saving technologies developed in advanced industrial countries. Itthereby tends to overlook the importance of finding appropriate technologiesfor efficient use of scarce capital and abundant labour in developing eco-nomies. In more recent years, the basic defect of such strategy has becomeevident in the economic stagnation plaguing its faithful adherents. It has thusbecome recognized that accumulated capital cannot be an effective basis ofeconomic development unless it is combined with appropriate technologyand manpower under an appropriate organization. Chapter 6 examines insti-tutional conditions by which appropriate technology and human resourcesare developed for rapid industrialization. A conclusion is that governmentinvestment in scientific research and education as well as the organizationof competitive markets to facilitate innovations by Schumpeterian entre-preneurs are necessary conditions for sustained industrial development.

Chapter 7 examines the problems of inequality, poverty, and environ-mental degradation that developing economies are facing. In the early phaseof development, strong population pressure on limited land resources tends topush up land rent and pull down labour wage rates in the rural sector. In theurban sector the importation of labour-saving technologies from advancedcountries tends to increase returns to capital relative to labour. Altogether,income disparity between asset-owning and assetless classes widens. Con-currently, as farmlands become short to support growing rural population,people tend to open and cultivate fragile lands in hills and mountains whichwould be better conserved for forest and pasture, with the result being serioussoil erosion and flooding, and thus aggravating poverty in rural areas. Inmajor cities air and water pollution tend to worsen at an accelerating pacebecause early industrialization often proceeds with little investment in pol-lution control. There is a real danger that the growing inequality and dete-riorating environment might create so much social tension as to result inmajor social disruptions. Yet, importation of social welfare institutions suchas minimum-wage laws for the purpose of income transfer risks worseningthe lot of the majority of poor people who stake out a living in informalsectors lying outside the realm of such programmes. A solution should besought that is directed at counteracting the basic economic forces whichcreate problems instead of trying to cure only their apparent symptoms.

Chapters 8 and 9 discuss what kind of institutional set-up would beappropriate for promoting economic development. In Chapter 8, this problemis considered from the question of how to combine the market and the statefor the design of an economic system. In Chapter 9, the discussion is expandedto include the question of how to incorporate community relations into theeconomic system. The market is an organization coordinating competitionamong people seeking profits by impersonal means of prices. The state

8 Introduction

intervenes in matters of resource allocations through the use of coercive po-wer. The community organizes collective actions based on mutual trust withina small group characterized by intensive personal interactions. Theoretically, themarket is efficient in the supply of private goods. The community's com-parative advantage lies in the supply of 'local' public goods, of which thebeneficiaries are locally confined, whereas the supply of 'global' or 'pure'public goods such as basic scientific research and judicial systems should be liewith the state. However, in developing economies where markets are poorlyorganized and characterized by highly imperfect information, they tend to failto achieve efficient resource allocations even for private goods. Also, in somerural communities that had hitherto enjoyed free use of abundant naturalresources under sparse population, it would be difficult to develop the ability tomanage common-property resources at an adequate pace to cope with rapidlygrowing resource scarcity under accelerated population growth. In these casesit may appear necessary for the government to become involved in activitiessupplying private goods and local public goods. However, it must be recog-nized that in the economies where the market is undeveloped and local com-munities' resource-management capacity is low, the government'sadministrative and information-collecting capacity is also weak. Therefore, theexpansion of the scope of government activities for the correction of marketand community failures could well be subject to the high probability of gov-ernment failure, which could be much more costly to society.

What should be the right combination of community, market, and state forpromoting economic growth is thus the problem of high research priority indevelopment economics. There is no single optimum combination uniformlyapplicable to developing economies. Under different cultural and social tradi-tions, the efficiency of the market may be relatively higher in one economy,whereas the organizational ability of community is relatively stronger inanother. In the former it would be effective to increase the role of the market,whereas in the latter it would be better to expand the role of community. Forexample, in the course of modernization in Japan, a rather unique form ofeconomic organization has been created under a different cultural and socialtradition from that in the West. On the basis of this unique institutionalset-up, Japan was able to catch up with the economic power of WesternEurope and North America, though it has been turning out to be a negativeasset in the development stage after the completion of catching up. Chapter 10concludes with the argument that if developing economies today are to catchup with developed economies, they must develop effective economic systemseach suitable to their unique cultural and social traditions as well as theirdevelopment strategies.

1. A Theoretical Framework forEconomic Development

In this book we will examine the economic development necessary to bridgethe extremely wide gap in per capita income between the low-incomedeveloping economies and the high-income developed economies in theworld today. Such extensive economic growth cannot be realized withoutexamining the requisite major changes in social organizations and people'svalue systems. Understanding the process by which quantitative expansionsin economic variables (such as capital and labour force) interact with cultureand institutions to evolve a social system that supports major growth in percapita income should be the ultimate goal of development economics. As astep towards this goal, development of a theoretical framework for the analysisof complex relationships among economic, cultural, and institutional changesis presented in this chapter.

1.1 Development of the Social System

1.1.1 A model of dialectic social development

A broad conceptual framework for development of social systems is outlinedin Figure 1.1. This figure illustrates a model of the evolution in social systemsthrough dialectic interactions between economic and cultural-institutionalvariables. The lower section of this figure represents the economic sector as asubsystem of society. This subsystem consists of interactions between tech-nology and 'resources'broadly defined as 'factors of production', includingnatural resources, labour, and capital. Technology is the determinant on thevalue of product to be produced from a given combination of productionfactors, commonly called 'production function' in economics.

If we measure economic growth by the increase in average per capitaproduct (or income), it is realized through increases in per capita endowmentsof resources and/or 'progress in technology' defined as an increase in productfor given inputs of resources. 'Product' is defined here as economic valuenewly added to society by the inputs of labour, capital, and natural resourceswithin a period; this Value added' is distributed to owners of the resources tobecome their incomes, which are aggregated into the income of the society.

Increases in economic resources and progress in technology are not inde-pendent. For example, as the technology of controlling water-flows is

10 Framework for Economic Development

FIG. 1.1 Interrelated developments in the social system

developed and necessary investments are made for use of the technology-such as construction of irrigation canals and diversion damshitherto uselessbarren lands could be converted into economically useful arable lands. Ifmore food could be produced on the increased land resources, the food sur-plus might be stored, allowing a greater portion of labour input to be divertedfrom food production to capital formation activities in the next period.

Thus, while the progress of technology provides a basis of resource aug-mentation, it is promoted by purposive resource-using activities. For exam-ple, advances in irrigation technology are achieved through research on theidentification of water-flow patterns as well as the development of irrigationfacilities for adequate control of the water-flows through experiments ofvarious designs, be it done by scientists and engineers in modern researchlaboratories or by primitive trial and error by peasants on their farms. Thoseactivities use both human effort and capital for the addition of the stock ofengineering knowledge. Since this increase in knowledge has the sameoutput-increasing effect as investment in tangible capitalsuch as the con-struction of irrigation canals and damsresearch and development activitiescan be called 'investment in intangible capital'.

Framework for Economic Development 11Similar to the production of tangible capital, it is possible to formulate a

process of producing technical knowledge from the inputs of labour andcapital. A critical element in augmenting this knowledge production functionis 'investment in human capital', defined as enlargement of human capacityby such means as education, training, and health care. Investment in humancapital will increase the efficiency of knowledge production, which in turnwill improve the efficiency of production of economic value added fromgiven resources in the society. Thus, cumulative increases in average productper capita will result from investments in both tangible and intangiblecapital.1

The productivity of an economic subsystem, consisting of its resourceendowments and technology, is conditioned by culture and institutions insociety. Broadly defined, institutions as well as technology are a part ofculture. However, culture is here narrowly defined to imply the value systemof people in the society, while institutions are defined as 'rules sanctioned bythe members of the society' including both formally stipulated laws andinformal conventions. Cultures and institutions thus defined are inseparablyrelated. The rules that contradict the morals of people would not be sanc-tioned socially and, if stipulated formally, would not function effectively. Forexample, the institution of slavery to stipulate a person's property rights onother human beings could hardly be expected to function as a social insti-tution today as it is inconsistent with the culture of the modern world. Yet, itwas a perfectly legitimate and effective institution under different culturessuch as in ancient Greece and Rome.

Culture and institutions indicated in the upper section of Figure 1.1 ascomponents of the social system exert significant influences on the eco-nomic subsystem located in the lower section. For instance, an importantparameter to determine the rate of investment is the ratio of saving toincome; this parameter is determined largely by people's future preferenceover present consumption, which is a part of their value system. It hasbeen the tradition of modern neoclassical economics to analyse theworkings of the economic subsystem under the assumption of fixed pre-ferences. Such an approach would be effective for the analysis of asituation in which the upper subsystem was relatively constant. Yet, theapproach would be grossly inadequate for dealing with the wide range ofeconomic development within which major cultural and institutionalchanges inevitably occur. In this respect, the theory of Max Weber (1920)identifying the Protestant ethic as a source of modern capitalist develop-ment represents an important methodological suggestion, irrespective of itsempirical validity.

12 Framework for Economic Development

1.1.2 A historical example

While accumulation of resources and progress in technology are conditionedby culture institution changes in the latter are also induced by the former.Such a process of social development through dialectic interactions betweenthe economic and the cultural-institutional subsystems may be understoodmore concretely by tracing the transition from the hunting and gatheringeconomy to the agricultural (and pastoral) economy.

A basic force inducing this epochal change in human history was theincreased scarcity of natural resources under the pressure of populationgrowth.2 As long as population was sparse and land was felt to infinitely existlike air, the killing of wild animals and the harvesting of wild crops inunlimited amounts would have shown no sign of exhaustion. However, aspopulation grew (though very gradually), it was inevitable that the day wouldcome when exploitation of the wild resources began to exceed their repro-ductive capacity and, thereby, the hunting-gathering economy could not besustained.

To avoid the subsistence crisis that arose from this resource exhaustion, itbecame imperative for hunters to augment/increase the reproduction processby raising animals instead of killing and eating them immediately, and forgatherers to plant nuts and cereals for future harvests. An economic basis ofthe increased reproduction was the accumulation of capital. A limited list ofcapital items was required for hunting and gathering, such as stones, knives,clubs, and bows and arrows. A larger capital stock was required for shifting tothe agriculture-based system, especially in the forms of reared animals, stand-ing crops and trees, and opened and cultivable farmlands. Capital requirementincreased further as the agricultural production system advanced to the stageat which it began to rely heavily on man-made land infrastructure, such asirrigation and drainage facilities.

To convert animals and plants to productive capital, it was necessary toaccumulate knowledge to identify useful animals and plants for domestica-tion as well as the appropriate methods to feed and grow them. Countlessefforts of primitive producers to advance agricultural technology throughtrial and error were the major source of investment in intangible capital. Theseefforts to enlarge the reproduction process under the growing scarcity ofnatural resources are likely to have been induced by the producers' need forsurvival.

While such advancement in technical knowledge was necessary, it was notsufficient for the development of the agriculture-based economic system. Thisdevelopment required a major institutional change: establishment of property

Framework for Economic Development 13rights on productive resources. A basic rule in ideal primitive hunting andgathering economies was free access to natural resources, under which all theresources were the property of everyone but no one person's property inparticular. Under this rule anyone could capture and consume any usefulanimals and plants as they found them. As long as this rule prevailed, a personwho attempted to engage in agricultural production had to face the difficultyof preventing others from taking away the animals and crops he raised. In suchcircumstances there would have been little incentive for anyone to startagricultural production by investing in livestock and standing crops. There-fore, the requisite for the formation of an agricultural economy was theestablishment of a new social order of clearly defined property rights by whichthe person who made efforts to invest in productive capital could excludeothers from its use (Demsetz, 1967; Alchian and Demsetz, 1973). In the courseof this development of agrarian civilization, property rights were first assignedto livestock and standing crops, and later extended to cover agricultural lands.

Those who were assigned property rights on land would have beenequipped with strong incentives to invest in improving the quality of the land,from removing stones and tree roots, fencing and terracing, to irrigation anddrainage. The form of property rights also evolved from communal ownershipby tribe or village to private ownership by household or individual, witha stronger power of exclusion and, hence, a stronger incentive for privateinvestment.

Common to all institutions, stipulation and enforcement of property rightsentail costs. The most profitable situation for an individual is for him to breakthe rules (e.g. steal others' properties) while others are observing the rules(e.g. do not steal others' properties). Thus, the temptation is always high foranyone to become a 'free-rider' who tries to gain from breaking the rules. Tothe extent that people's propensity to become free-riders is high, it is costly toenforce the property rights by such means as police and courts. It is the ethicsas a part of culture that reduces the cost of enforcing the rules of society.Indeed, 'thou shall not steal' is a unanimous moral code in the commandmentsof the great religions that coincided with the development of agrarian civil-izations. It seems reasonable to hypothesize that such a religious doctrine wasboth the cause and the consequence of establishment of the agriculturallybased economic system.

Economic and social development through such interactions betweeneconomic forces and cultural-institutional elements have been repeatedover history. For instance, the patent system that was established with thedevelopment of modern industrial society was aimed at assigning propertyrights on engineering knowledge and information, thereby promoting private

14 Framework for Economic Developmentinvestment in this critical component of intangible capital (Evenson andWestphal, 1995). Negotiations in the GATT (General Agreement on Tariffsand Trade) Uruguay Round followed up by WTO (World Trade Organization)on intellectual property rights represented an attempt to establish inter-nationally uniform rules on the protection and the transactions of propertyrights over a wide range of knowledge and information including computersoftware. This attempt was a response to the growing need of the world todayin which the role of knowledge and information, as a factor of economicproduction, has been rising faster than that of tangible capital. Likewise, theestablishment of the International Law of the Sea creating exclusive economiczones over 200 nautical miles from each country's coast was an attempt tomobilize conservation efforts for marine resources at the national level inresponse to growing scarcity and high prices of fish and other marine prod-ucts (Hannesson, 1991). These are among the efforts to achieve the institu-tional innovation of the same nature as developing property rights onlivestock, crops, and lands in the prehistoric initiation of agriculture.

7.7.3 Marx and new institutionalism

The theoretical framework outlined above has a basic similarity with theperspective on evolution of the social system described by Karl Marx andFriedrich Engels.3 The economic subsystem and the cultural-institutionalsubsystem in Figure 1.1 correspond broadly with what they term 'infra-structure' and 'superstructure', respectively. In their system, the core of thesuperstructure is the property-rights relations of production factors (so-called'production relations'), while infrastructure is the technology needed todetermine the capacity of material production from available resources. Whilethe institution is believed to determine realization of the technology's pro-duction potential, technology is identified as the basic force in structuring theinstitution; at the origin the institution is so structured as to best exploitthe potential of material production. This view on the formation of institu-tions in response to economic demand is analogous to the theory of inducedinstitutional innovation.

Marx and Engels assumed a major time-lag between increases in materialproduction capacity and changes in institutions; this made changes in thesocial system discontinuous and abrupt. In their perspective technicalknowledge and tangible capital are accumulated gradually to bring aboutcontinuous growth in productive capacity. In contrast, institutions cannotadjust immediatelythey must be stable over time so that the rules of society

Framework for Economic Development 15for structuring people's stable expectations in dealing with others couldeffectively function.

Moreover, the core institution in the Marx-Engels theory is the property-right assignment of a key production factor at each stage of economicdevelopmentsuch as slaves in the ancient classical world, land in medievalfeudalism, and capital in modern industrial capitalism. Changes are bound totake time as it will be strongly resisted by the prestige class to whom propertyownership is exclusively bestowed. As a result, even though the institutionwas originally designed to best exploit the productive potential of society, asit becomes inconsistent with the changed conditions of material productionresulting from technological progress and capital accumulation, it tends tosurvive. In other words, the institution that was once a carrier of economicdevelopment over time turns out to be the 'fetter' against further developmentunder a new technology regime. Marx and Engels theorized that this gapbetween the institution and the production potential would be ultimatelyclosed through a violent political revolution. This perspective was forciblymarshalled in a classic statement by Marx:

The mode of production of material life determines the general character of social,political and spiritual processes of life. At a certain stage of their development, thematerial forces of production in society come into conflict with the existing relationsof production, orwhat is but a legal expression for the same thingwith the propertyrelations within which they had been at work before. From forms of development ofthe forces of production these relations turn into their fetters. Then comes the periodof social revolution. With the change of the economic foundation the entire immensesuperstructure is more or less rapidly transformed. (Marx [1859], 1904: 11-12)

Marx considered technological progress and capital accumulation decisivein determining the productive capacity of society and denied the importanceof natural resources relative to population. In this respect, our perspectivediffers from Marx's and is closer to that of new institutional historians inemphasizing the influence of changes in relative resource endowments andprices due to population growth and other factors (North and Thomas, 1973;North, 1981). We also consider that institutions are not quite as inflexible asto make violent revolution inevitable for major institutional changes. There isconsiderable historical evidence to support the hypothesis that the basicinstitutional framework, including property relations, changed throughcumulative adjustments by such means as informal agreements and reinter-pretations of laws and codes (Davis and North, 1970).

However, there is no guarantee that such cumulative adjustments aresufficiently rapid and responsive to emerging social needs. The cost of

16 Framework for Economic Developmentincremental change in one institution can be prohibitively high as this par-ticular institution is inseparably intertwined with others. Its change therebydemands a change in the total institutional framework that has been his-torically determined (see Section 1.2.4 on this historical path dependency).Due to fear of social sanctions, such as ostracism, against the deviation byindividuals from established norms and conventions, even obviously ineffi-cient institutions like castes are often difficult to change (Akerlof, 1984).Because a future gain from an institutional reform is uncertain, and itsdistribution among various social groups is difficult to predict relative to theobvious loss to a specific group, opposition to reform tends to be stronglyorganized, while support is only weakly so (Fernandez and Rodrik, 1991), interms of the logic of the political market (Section 1.2.3). It is therefore notuncommon to observe that a society continues to be trapped in economicstagnation and poverty under a dysfunctional system bound by strong socialinertia for the preservation of established institutions (Basu et al, 1987).

Thus, it is likely that changes in institutions and, more so, in culture lagsignificantly behind changes in the material production base, and that theresulting contradictions could often create strong social and political tension,culminating in major disruptions, as Marx and Engels envisioned.

1.2 The Theory of Induced Innovation*

The theoretical framework developed in the previous section is general butnot very operational for economic analysis in the sense that the impliedhypotheses are too broad for empirical testing. In the following section wewill construct an operational economic model by extracting some elementsfrom the general model, on which development economics must focus. Forthis purpose it is necessary to use technical terms specific to economics.

7.2.7 Induced technological innovation

First, our focus will be placed on a causal relationship within the economicsubsystem in Figure 1.1, in which changes in resource endowments inducechanges in technology. A standard economic theory on this relationship iscalled the theory of 'induced technological innovation' in the tradition ofJohn R. Hicks (1932).

The Hicksian theory presupposes a mechanism in which, as the endowmentof one factor (e.g. capital) becomes more abundant relative to another factor

* Readers not accustomed to the technical analysis of economics may wish to skip this section.

Framework for Economic Development 17(e.g. labour), a change in technology is induced towards using more capitaland saving labour for given relative factor prices (for a more exact definition,see Appendix A.2). Such a biased change in technology stems from the effortsof profit-seeking entrepreneurs to reduce production costs by substitutingrelatively more abundant (hence cheaper) resources for scarcer (hencedearer) resources. The induced innovation theory within the framework ofneoclassical economics has assumed a competitive market by which relativeabundance and scarcity of factors are reflected in factor prices used as data forentrepreneurs' production plans. However, this theory can be applicable tosubsistence-orientated non-market economies also, if it is assumed thatrelative resource scarcities are recognized by producers, even very roughly, interms of shadow prices reflecting the social opportunity costs of the resources.

Based on such assumptions, Figure 1.1 is a model explaining the process oftransition from the hunting-gathering economy to the agricultural economy,as well as subsequent advances in the technology of agricultural production.With some modification, this model can be used to explain a transition to theindustrial economy also.

Figure 1.2 represents the production relation (production function) ofproducing a single commodity (e.g. food) from inputs of three factors: labour(I), capital (K), and land (A) representing natural resources. Capital is hereassumed to be produced mainly by past labour input.

The upper A-L quadrant in Figure 1.2 represents the substitution betweenland and labour in terms of isoquant for producing one unit of product (unitisoquant). On the other hand, the 0-Z line in the lower L-K quadrantrepresents the complementary relationship of capital with labour in the eventof substituting labour for land. For example, as long as a farmer engages inslash-and-burn shifting cultivation, he can cultivate a large area using hisown labour with very little capital consisting of such small items as a hatchet,a digging stick, and a stock of seeds. However, if he attempts to shift to a morelabour-intensive, land-saving system under settled agriculture, he must buildup large capital by improving farmlands (removing roots and stones, terrac-ing and fencing) and acquiring a greater variety of farming tools andimplements than those needed for shifting agriculture. Thus, the substitutionof labour for land through such intensification of land-use should beaccompanied by exponential growth in the capital-labour (K/L) ratio. Toillustrate this relationship, the 0-Z line is drawn in a concave form.

The /-curve in the A-L quadrant represents the 'innovation possibilitycurve', defined as an envelope of unit isoquants corresponding to all thepossible technologies that could have been developed with the knowledge andhuman capacity available at a particular period. This curve shifts over time

18 Framework for Economic Development

FIG. 1.2 A model of induced innovation

(from 70 in period 0 to /j in period 1 as indicated in Figure 1.2) correspondingto the accumulation of knowledge and the improvement in human capacity.According to the theory of induced technological innovation, a particulartechnology as represented by i0 is developed and adopted for period 0,because it is this technology that minimizes the cost of production for theprice ratio between land and labour (Po) reflecting relative scarcities of thesefactors in this period. In other words, i0 is developed through the effort ofproducers to reach the cost-minimizing point a within an available set ofpossibilities (/0).

Assuming complementarity between labour and capital in their substitu-tion for land, as explained earlier, the land-labour ratio at point a(OA0/OI0)

Framework for Economic Development 19corresponds with the capital-labour ratio at point d(OKo/OLo). Since in thisparticular case it is assumed that capital is the product of past labour alone,the price of capital relative to the price of land can be considered to movelargely parallel with the labour-land price ratio (P0). This assumption ofcomplementarity between labour and capital is adopted for the sake of sim-plicity to represent the three-dimensional relation in a two-dimensionaldiagram. This simplification might be permissible as an approximation tofacilitate understanding the characteristics of technological progress in pre-industrial economies. For the analysis of industrial economies in Chapter 6,the substitution between labour and capital as well as the substitutionbetween tangible and intangible capital will be treated as a central problem.

Assume that, as time passed from period 0 to period 1, relative scarcity ofland increased with the result of lowering the relative price of labour to landfrom P0 to PI . Meanwhile, the innovation possibility curve would have shiftedtowards the origin from 70 to /], reflecting the increased capacity of society toproduce a unit of food with a smaller input of factors. Corresponding to thesechanges, it now becomes optimum for producers to reach point c by choosinga technology represented by ii, over other possibilities embraced by /].

However, until the new ii technology is actually developed, producers willhave to continue using the old i0 technology and, hence, can move only frompoint a to b. It is through producers' efforts in repeated trial and error, as wellas organized scientific research and development (in the case of modernsociety), that the new ii technology will become available. The basic premissin the theory of induced technological innovation is that the expected gain (orreduction in cost) for producers, as measured by the distance between PI andPI, in the move from point b to c, will induce them to make efforts fortechnological development with the result of changing technology from i0towards i\.

The move from hunting and gathering to agriculture may be explained interms of this theory as follows: When the availability of usable land appearedto be limitless relative to sparse population and, therefore, the relative scarcityof land to labour (P0) was very low, collection of foods from wild animals andplants (IQ) could well have been an optimum technology in the sense that itproduced food at a minimum cost. Even if population grew, and the relativescarcity of land rose (P0 to PI), there would have been little scope to increasefood supply by applying more labour to limited land (a to b) as long ashunting and gathering were the sole option for food production. However, iffarming technology (ij) became available, people would be able to producemuch more food from given land resources (b to c) at a lower cost. Thispossibility would have worked as a driving force for primitive hunters and

20 Framework for Economic Developmentgatherers to search for ways to increase reproduction of useful animalsand plants.

7.2.2 Induced institutional innovation

The theory of induced technological innovation is explained above in terms ofproducers' cost-minimizing behaviour in the tradition of neoclassical eco-nomics. Such a theoretical structure appears to be relevant to modern marketeconomies in which technological innovations are carried out mainly by largefirms with research and development capacities, though theory has been asubject of heated theoretical discussion.4 Major modifications are needed toapply the theory to the analysis of transformation within subsistence-orientedeconomies and transition from subsistence-oriented to market-orientedeconomies.

The reason is not, as once commonly thought, because small subsistence-oriented producers in premodern economies are ignorant and bound bytradition, and therefore, unable to search for and adopt profitable crops andcultural practices. On the contrary, accumulated evidence shows thatsubsistence-oriented small farmers in developing economies allocate resourcesrationally and respond effectively to profitable economic opportunities(T. W. Schultz, 1964; Hopper, 1965; Yotopoulos, 1968; Barnum and Squire,1979; Rosenzweig, 1984; Tiffen and Mortimore, 1994). This trait would beshared not only by farmers but by hunters and gatherers as well.

It is not reasonable, however, to assume that they anticipate a wide range ofinnovation possibilities along the /-curve and move linearly towards point cin response to changes in relative factor scarcities and/or innovation possib-ilities. It is more reasonable to assume an evolutionary process of the Nelson-Winter (1982) type, namely as food production per capita decreased forhunters and gatherers, corresponding to growing population pressure onnatural resources, they were forced to search for ways to increase food supplythrough trial and error. Only those who happened to reach the i\ curve(agriculture) were able to survive. With this modification, induced techno-logical innovations are thought to produce technological change in thedirection that the traditional theory predicts. However, some economies maynot be able to survive because they continue to be trapped in the old tech-nology (i0). Some may be able to survive as they adopt better technology thani0. However, there is no guarantee that they can reach the best technology asrepresented by ii.

A major modification required for the theory to cover both primitive andhigh stages of development would be to combine the theory of technological

Framework for Economic Development 21innovation with the theory of institutional innovation. For whatever highprofit a technological innovation may be expected to produce, and howeverrational a producer may be, it may not be possible for him alone to carry outthe innovation. As explained earlier, the development from hunting andgathering to agriculture involves the process of capital accumulation in theform of livestock, standing crops, and prepared farm fields, for which prop-erty rights need to be established. However, assignment and protection ofproperty rights can hardly be achieved by individual efforts but need col-lective action by people in the society. Collective action is required not onlyto create institutions for promoting private investment incentives, but also toundertake large-scale investment in social overhead capital, such as floodcontrol of rivers and building of gravity irrigation systems. Appropriateinstitutions must be prepared to organize people effectively for such collect-ive action.

Then what mechanism should we assume to organize collective action tofacilitate technological progress and capital accumulation in a sociallyoptimum direction? The most naive model would be to assume that collectiveaction is organized when aggregate social profit from the move from pointb to c (Figure 1.2) exceeds the cost of organizing the collective action toenable such a move. This naive model could well be valid in broad terms ofprogress in human history in which property rights have been strengthenedand institutions have developed to mobilize collective action for buildinginfrastructure (such as irrigation) corresponding to growing populationpressure on natural resources.

However, if such a naive mechanism of induced institutional innovationalways operated, all the economies would have grown smoothly and no greatincome gap would ever have emerged between developed and developingeconomies. Thus, to understand the causes of the poverty and under-development versus the wealth and development of nations in today's worldit is necessary to understand the conditions under which the mechanism ofinduced institutional innovation fails to operate effectively.

7.2.3 Logic of political market

The supply of public goods in response to social needs is determined throughpolitical process at equilibria between demands for and supplies of thosepublic goods from various interest groups, which might be called 'politicalmarkets' in analogy with economic markets for ordinary goods and services.The problem is that the mechanism of the political market does not guaranteethe optimum supply of public goods in terms of economic well-being in

22 Framework for Economic Developmentsociety. As Mancur Olson (1965) predicted, collective action is usually muchless organized than a socially optimum level, because only part of its profitaccrues to those who shoulder the cost of organizing the action. This is thebasic cause of a general undersupply of public goods.

Social rules (such as property rights) and social overhead capital (such asroads) bear the properties of'non-rivalness' and 'non-excludability' commonto public goods. Non-rivalness is the property of a good to be utilized jointlyby many, and non-excludability is the property of a good where utilization bythose who do not pay for the cost of its supply is possible (Musgrave, 1959;Stiglitz, 2000). For example, once an irrigation canal is dug by the collectivework effort of villagers, all those who engage in farming along this canal canutilize its water jointly. The problem is that it is difficult and costly to preventsomeone from using (or stealing) water who did not contribute labour for theconstruction of the canal. For this latter property (non-excludability), temp-tation is high for anyone to become a free-rider in the use of public goods;this applies equally to the enforcement of social rules, such as property rights,as explained in the previous section.

For the supply of public goods someone must take charge of organizingcollective action. Collective action is organized at various levels, includingvoluntary cooperation in the local community and the religious group. Forthe supply of 'global public goods' widely applicable to a large number ofpeople in society, however, it often becomes necessary to set up a mechanismof coercion in the form of 'state'. The collective action aimed to form andmanipulate the coercive power of state is called 'polities' or 'political move-ments'. The organizer of political movements is called a 'political leader' or'politician', whether from small local communities, nation states, or inter-national arenas. The leader must apply major efforts to bring people togetherin an agreement on collective action and enforce it with persuasion, intimida-tion, bribery, or violence. Economic benefits expected from the public goodproduced by organized collective action for society may far exceed the costpaid. This benefit is not usually appropriated by the political leader. Forexample, the stipulation and protection of property rights on livestock mayenable primitive hunters to engage in agriculture (as represented by a movefrom point b to c in Figure 1.2). However, the economic benefit from thisprovision of public good, as measured by P^P^, is appropriated by individualproducers who shifted from hunting to agriculture.

Returns to the leader for his cost of organizing collective action for thesupply of a public good (e.g. property-rights protection) would be thestrengthening of his power base due to increased support from people whocapture economic gains from the public good. Unless the increment in his

Framework for Economic Development 23utility arising from his strengthened political power was expected to exceedhis cost, he would not attempt to organize the collective action.

Such behaviour of the political leader is modelled in Figure 1.3, in thetradition of public choice theory or the economics of politics (Downs, 1957;Buchanan and Tullock, 1962; and Breton, 1974). Line MR representsdecreasing marginal revenue of the leader for increasing the supply of apublic good. Marginal revenue for the politician is defined as the marginalincrease in his utility from the strengthening of his power base (increasedvotes in the case of parliamentary democracy) expected from a unit increasein the public good provision. Line MR is drawn as a downward slope since itseems reasonable to assume that the marginal social productivity of a publicgood tends to decrease as its supply increases, with a resulting decrease in themarginal gain in political support from the beneficiaries.

On the other hand, the leader's marginal cost (MC) is defined as the mar-ginal disutility of his time and effort in organizing the collective action. LineMC is upward-sloping because the cost of preventing 'free-riders' rises pro-gressively as a greater number of people will have to be organized for anincreased supply of the public good.

Because the vertical distance between MR and MC measures the marginalnet utility or marginal profit (revenue minus cost) of the political leader, hisprofit will be maximized by the level of public good supply at the intersection

FIG. 1.3 A model of political market for a public good

24 Framework for Economic Developmentof these two lines. There is no guarantee at all that this optimum for thepolitician coincides with the optimum for society. In general, it is probablethat the supply of public good is below the social optimum because onlya fraction of social benefits will accrue to the politician.

If the political leader's marginal revenue and cost are located in an initialperiod (0) at MR0 and MC0, 000 is the optimum supply of the public good forthe political leader. If, towards the next period (l), changes occur in relativeresource scarcities and in technological possibilities (as represented respect-ively by P0 to PI and 70 to /j in Figure 1.2), a shift from the old to the newtechnology (IQ to ij) would become profitable for a large number of producersin the society. Then these potential beneficiaries from the new technologywould render stronger support for the politician who would act to provide thepublic good (such as the protection of property rights) that is needed forthe adoption of the new technology. The result would be the moving up of thepolitician's marginal revenue curve from MR0 to MR^. The mechanism ofinduced innovation in technology and institution would thus work throughsuch an inducement mechanism for the supply of public goods in the 'politicalmarket'.

The problem lies in how efficiently this inducement mechanism wouldwork in terms of the economic welfare maximization criteria for the society.How much the supply of a public good would increase in response to anincrease in social demand depends, in part, on how efficiently the increasedsocial demand is translated into the upward shift in the politician's MRcurve. This efficiency tends to be low, especially for the type of public goodwhose social benefit is large in aggregate but is distributed thinly over a largenumber of private producers, and, hence, not visible enough to mobilizepolitical support (or lobbying) activities. This is the basic dilemma that resultsfrom major under-investment in the public goods with high social pay-offs(Olson, 1965).

Another factor determining the efficiency of the political inducementmechanism is the slope of the MC curve. The increase in the supply of publicgood in response to a given shift in the marginal revenue curve from MR0 toMRi is larger for a relatively flat marginal cost curve such as MC0 than for asharply rising curve such as MC'0. A major determinant of the location andshape of the MC curve is people's value system. For example, the marginalcost of strengthening property rights would be high in a society in which thetheft of animals had not been recognized as a major crime.

Conflict of interests among various groups in a society would also sharpenthe slope of the MC curve. For example, the establishment of property rightson land should produce major benefits to those undertaking the change to

Framework for Economic Development 25settled agriculture. It would be opposed, however, by hunters and nomadswho would be excluded from the use of the land to which property rights areassigned. If this opposition is well organized, the marginal cost of strength-ening property rights on land would rise sharply so that the supply of thispublic good would be severely limited relative to increased social need.

How can the efficiency in the translation of social demand to the politi-cian's marginal revenue curve be improved? How can his marginal cost curvebe lowered? To a large extent, these tasks were facilitated in premodernsocieties by the religious developments that changed people's moral per-ceptions. What ideologies would be an effective substitute for this role ofreligion in modern societies? How can modern education and informationmedia promote efficiency of the induced innovation mechanism involv-ing political processes? This problem is one of the most difficult and mostimportant agendas in development economics (to be discussed in detail inChapters 8, 9, and 10).

7.2.4 Historical path dependency

A major constraint on the effective working of the induced innovationmechanism would be scale economies in an institutional set-up corres-ponding to a particular technological regime. Such scale economies wouldmake incremental changes difficult in an economic system that was histor-ically formed. For example, in the process of transition from nomadismto agriculture, it may have been difficult for a small number of farmers toestablish arable farming with their collective action, even if they agreed torespect each others' property rights on lands and crops. They could hardlyprevent nomads from grazing animals on their croplands because of thecustoms of nomadic society. Thus, the transition to settled agriculture inducedby population pressure on land resources could have been disrupted by thebinding power of traditional nomadic culture and institutions. However, if forsome historical reason (such as a large-scale migration of agriculturalists likehomesteading in the US West) a majority of land happened to be enclosed,nomads may have found it difficult to continue their traditional way of lifeand would have been compelled to move to settled agriculture, therebyeliminating the nomadic system.

This example illustrates the possibility of multiple equilibria (e.g., domina-tion of nomadism versus domination of settled agriculture) for a societyto reach in a manner similar to the world of 'new growth theory' withthe assumption of increasing returns based on externality (Romer, 1987;Murphy etal, 1989;Krugman, 1991; Grossman and Helpman, 1991), which is

26 Framework for Economic Developmentdiscussed later (Section 6.3.1). To which equilibrium a society will movedepends to a large extent on its 'historical path' (David, 1985; Arthur, 1988;North, 1990: ch. l). Useful insights on the emergence of multiple equilibria arealso provided by evolutionary game theory (Mailath, 1992; Kandori, Mailath,and Rob, 1993; Matsui, 1996; Aoki, 2001). According to this theory humanbeings can perceive future possibilities only within a narrow range based ontheir own past experience, and also they tend to be conce