AWORLD BANK COUNTRY ECONOMIC REPORT Korea A World Bank Country Economic Report 127- Kofea Pmincid Bo..d.,ies spmi.1 citje NORTH KOREA 0 20 40 60 80 100 KILOMETERS 0 20 40 60 MILES GADIG W Seoul EONG GI ,Ls- f-I .. \.-. " j' CHUNG /1" BUG CHUNG  NAM GYEONG Daejeon -L: BUG 3 Gumi JEON Daegu BUG GYEONG NAM Changwon Mas; n JEON NAM CHINA K RE Korea Policy Issues for Long-Term Development The report of a mission sent to the Republic of Korea by the World Bank Parvez Hasan and D. C. Rao Coordinating Authors Published for the World Bank The Johns Hopkins University Press Baltimore and London Copyright © 1979 by The International Bank for Reconstruction and Development THBV WORLD BANK 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America The views and interpretations in this book are the authors' and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their behalf. The map serving as frontispiece has been prepared for the convenience of readers of this book; the denominations used and the boundaries shown do not imply, on the part of the World Bank and its affiliates, any judgment on the legal status of any territory or any endorsement or acceptance of such boundaries. Library of Congress Catalog Card Number 78-21399 ISBN 0-8018-2228-9 ISBN 0-8018-2229-7 paperback Foreword THIS IS THE SEVENTEENTH IN THE CURRENT SERIES of World Bank country economic reports, all of which are listed on the following page. They are published, in response to a desire expressed by scholars and practitioners in the field of economic and social development, to facilitate research and the interchange of knowledge. The Bank regularly prepares economic reports on borrowing countries in support of its own operations. These surveys provide a basis for dis- cussion with governments and for decisions on Bank policy and opera- tions. Many governments use the reports as an aid to their economic planning, as do consortia and consultative groups of governments and institutions providing assistance in development. All Bank country reports are subject to the agreement of-and several have been published by-the governments concerned. Korea: Policy Issues for Long-Termn Development results from the work of a number of missions that visited the Republic of Korea during 1976,whfen tGe=Korean government was preparing its fourth five-year economic development plan for 1977-81. These missions had three basic purposes: to review and analyze economic developments during 1961-76; to assess the fourth plan; and to identify issues that would continue to be relevant in the 1980s, the period of the fifth and sixth development plans. The findings of these missions were consolidated in a report for discussion with the Korean government in early 1977; that report has now been prepared for publication. Because the Korean economy has maintained a rapid pace of growth in recent years, it is more than ordinarily important to point out that the book does not take into account the developments since early 1977. If anything, however, the general optimism of this study has been reinforced by events since that time. The first three chapters of this book constitute the main report. These chapters synthesize the findings of the several studies and present v Vi POREWORD a broad analysis of development issues in the fifteen years before 1976 and the fifteen years thereafter; they draw heavily on the detailed presentation and analyses in subsequent chapters and appendixes. This book thus extends and complements the analysis in Korea: Problems and Issues in a Rapidly Growing Economty, published for the World Bank by The Johns Hopkins University Press in 1976. HOLLIs B. CHENERY Vice President for Development Policy The World Bank Washington, D.C. March 1979 WORLD BANK COUNTRY EcONoMIc REPORTS Published for the Bank by The Johns Hopkins University Press Korea: Policy Issues for Long-Term Development The Commonwealth Caribbean: The Integration Experience Ivory Coast: The Challenge of Success Korea: Problems and Issues in a Rapidly Growing Economy Kenya: Into the Second Decade Yugoslavia: Development with Decentralization Nigeria: Options for Long-Term Development Economic Growth of Colombia Published by the World Bank Papua New Guinea: Its Economic Situation and Prospects for Development The Philippines: Priorities and Prospects for Development Lesotho: A Development Challenge Turkey: Prospects and Problems of an Expanding Economy Senegal: Tradition, Diversification, and Economic Development (also pub- lished in French) Chad: Development Potential and Constraints (also published in French as Le Developpement du Tchad: Possibilites et Limites) Current Economic Position and Prospects of Peru Current Economic Position and Prospects of Ecuador Employment in Trinidad and Tobago Contents Foreword v Acronyms xv Acknowledgements xix Overview PARVVEZ HASAN AND D. C. RAO 3 The Pattern of Rapid Economic Growth, 1961-76 3 Can Rapid Economic Growth Continue} 7 PART ONE THE MAIN REPORT 1. The Pattern of Economic Growth, 1961-76 D. C. RAO 15 Exports 16 Manufacturing 20 Agriculture 23 Infrastructure 24 Urbanization 25 Investment and Financing 27 Eimployment and Equity 32 2. Korea's Fourth Plan, 1977-81 D. C. RAO 45 Objectives of the Fourth Plan 49 Exports 52 Investment Program 57 Domestic Resource Mobilization 60 External Resource Requirements 62 Employment and Wages 68 Social Development 71 3. Development Issues in the 1980s PARVEZ HASAN AND D. C. RAO 75 Export Prospects 76 Changes in Industrial Structure 78 Export Dependence 80 vii viii CONTENTS Policy Implications 82 Resource Requirements 84 Compulsions for Growth 85 The Development of Agriculture 95 Government's Changing Role in Investment Programming 103 PART Two HUMAN RESOURCES 4. Population Growth, Programs, and Policies CATHERINE S. PIERCE 109 Demographic Trends 110 Population Projections 118 The National Family Planning Program 120 Determinants of Changing Fertility 129 Policies Beyond Family Planning 132 5. Social Development EREDRICK GOLLADAY AND TIMOTHY KING 135 Health Conditions and Services 136 Major Issues of Health Policy 142 Education and Skill Formation 151 Major Issues of Educational Policy 160 Social Development Priorities 168 6. Labor Availability and Employment RICARDO MORAN 173 Historical Trends and Pattems 175 Projections of Labor Supply 181 Changing Composition of Employment 192 PART THREE MAJOR SECTORS 7. Agriculture RANDOLPH L. P. HARRIS 205 Historical Perspective 206 Agriculture under the Fourth Plan 209 Major Issues of Agricultural Policy 212 8. Manufacturing LARRY E. WESTPHAL 233 Incentives to Industrial Expansion 236 Trade Expansion and Industrial Development 240 Depth of Industrial Structure 251 Manufacturing under the Fourth Plan 253 Major Issues of Industrial Policy 260 9. Energy PHILIPPE BOURCIER 281 Growth and Pattern of Energy Consumption 282 Development of the Fuel and Power Industry 286 Energy under the Fourth Plan 294 Major Issues of Energy Policy 297 CONTENTS ix 10. Transport TILLMAN NEUNER 303 Sectoral Structure 304 Pricing and Finance 307 Licensing System for Private Transport 311 Transport under the Fourth Plan 313 Major Issues of Transport Policy 318 11. Public Finance MARINUS VAN DER MEL 327 Composition and Size of the Public Sector 327 Past Trends of Public Savings 333 Public Savings under the Fourth Plan 337 Major Issues of Fiscal Policy 354 Mobilizing Financial Savings 379 Major Issues of Financial Policy 394 PART FOUR APPENDIXES A. Macroeconomic Model SYAMAPRASAD GUPTA 407 Structure 408 Assumptions 415 Sector Classification 417 The Equation System 418 B. Structure of Exports and Imports JAYATI DATrA-MITRA 429 Export Structure 430 Import Growth and Structure 439 C. External Capital GEORGE KALU 443 Sources of Medium- and Long-term Capital 445 Short-term Borrowing 451 Debt Service 451 External Finance under the Fourth Plan 453 D. Economic and Social Statistics GEORGE KALU 457 Index 525 FIGURES 6.1 Indexes of Manpower Use, 1963-74 176 6.2 Growth of Selected Categories of the Working-age Population, 1963-74 177 6.3 Annual Change in the Real Agricultural Wage and the Real Earnings in Manufacturing, 1962-75 180 6.4 Real Value Added per Worker and the Real Average Wage, 1963-74 200 X CONTENTS 12.1 Nominal Rates on Time and Savings Deposits, 1962-76 381 A.1 Simplified Flow Chart of the Model's Relations 410 TABLES 1.1 Composition and Growth of Gross National Product, 1961-76 16 1.2 Composition of Gross Manufacturing Output, 1961-75 21 1.3 Direct Contributions to Growth in Manufacturing, 1960-73 22 1.4 Urban Population, 1960-75 26 1.5 Composition of Investment, 1962-76 27 1.6 Financing of Investment, 1962-75 29 1.7 Extrabudgetary Operating Deficits, 1973-76 30 1.8 Debt Outstanding and Debt Service, 1966-76 31 1.9 Manpower Use, 1963-75 33 1.10 Composition of Employment, by Sector, 1963-75 33 1.11 Employment and Productivity, by Sector, 1963-75 34 1.12 Farmers' Terms of Trade, 1963-76 39 1.13 Grain Price Support by Government, 1968-76 40 1.14 Rural and Urban Income Levels, 1963-76 41 2.1 Balance of Payments, Current Account, 1973-76 46 2.2 Composition and Growth of Gross National Product, 1971-81 51 2.3 Export Targets, by Commodity Group, 1975-81 56 2.4 Allocation of Investment, by Sector, 1972-81 58 2.5 Investment and Savings, 1975-81 61 2.6 Gross National Savings and Income, 1967-81 62 2.7 Savings, by Sector, 1975-81 62 2.8 Unit Value and Volume of External Trade, 1972-76 63 2.9 Terms of Trade, 1974-81 64 2.10 Composition of Commodity Imports, by Group, 1975-81 64 2.11 Sources and Uses of External Capital, 1977-81 67 2.12 Labor Force, 1976-8 1 68 2.13 Employment, by Sector, Projections for 1976-81 69 2.14 Investment Program for Social Development, by Service, 1972-81 72 3.1 Export Ratios, Korea and Selected Countries, 1960-70 81 3.2 Population of Korea, Projections for 1975-2000 86 3.3 Age Composition of the Population, 1960-75 and Projections for 1975-2000 87 3.4 Indicators of Growth and Equity, Projections for 1976-90 91 3.5 Food Demand, Projections for 1975-90 96 3.6 Cultivable Land under Two Levels of Investment, Projections for 1975-9 0 98 CONTENTS Xi 3.7 Foodgrain Production, Demand, and Import Requirements, Projections for 1990 99 3.8 Per capita Disposable Income of Households, by Sector, Projections for 1976-90 103 4.1 Population of Korea, 1949-75 110 4.2 Age Composition of the Population, 1955-70 111 4.3 Selected Measures of Changes in Age Composition, 1955-70 111 4.4 Vital Rates and Expectation of Life at Birth, Estimates for 1950-75 112 4.5 Age-specific Fertility Rates, 1950-70 113 4.6 Age-specipc Marital Fertility Rates, 1950-70 114 4.7 Geographic Distribution of Population, by Province, 1955-75 116 4.8 Net Migration to Urban Areas, by Province, 1960-70 117 4.9 Total Population, Projections for 1975-2000 119 4.10 Selected Measures of Changes in Age Composition, Projections for 1975-2000 119 4.11 Selected Measures of Changes in Age Composition, by Area, Projections for 1975-2000 120 4.12 Demographic Indicators, Korea and Selected Countries, Projections for 2000 122-23 4.13 Achievements of National Family Planning Program, 1962-76 127 4.14 Selected Findings of the 1973 Family Planning and Fertility Survey 128 4.15 Family Planning Acceptors, by Method, 1962-75 129 4.16 Projected Number of Contraceptors Necessary to Reach Replacement Fertility by 2000 130 5.1 Investment Program for Social Development, by Service, 1972-81 136 5.2 Cause-specific Morbidity for Selected Subpopulations 138 5.3 Average Korean Diet, by Selected Nutrient, 1975 140 5.4 Supply of Health Manpower, Projections for 1974-81 143 5.5 Professional Capacity and Needs, by Province, Estimates for 1973 149 5.6 Educational Enrollment, by Level, 1950-75 151 5.7 Enrollment Ratios, by Age and Sex, 1975 153 5.8 Educational Attainment of the Employed Population, by Occupation, 1960-70 154-55 5.9 Enrollment in Public and Private Institutions, 1965-75 156 5.10 Enrollment in Tertiary Institutions, by Field of Study, 1965-75 160 5.11 Investment Program for Education and Manpower Development, 1977-81 162 6.1 Labor Force and Employment, 1963-75 174 6.2 Manpower Availability and Deployment, Projections for 1976-81 183 Xii CONTENTS 6.3 Average Labor Force Participation Rates, by Age, Sex, and Place of Residence, 1963-71 185 6.4 Growth of Labor Force, Simulation A, Projections for 1975-2000 189 6.5 Growth of Labor Force, Simulation B, Projections for 1975-2000 189 6.6 Urban Labor Force Participation Rates, by Age and Sex, Korea and Selected Countries 190 6.7 Labor Force, by Sex and Place of Residence, Simulations A and B, Projections for 1975-2000 192 6.8 Value Added, Employment, and Labor Productivity, by Sector, 1964-74 194-95 6.9 Sectoral Employment, Projections for 1976-90 198 6.10 Growth of Productivity and the Average Wage, 1963-74 199 6.11 Earnings Differences, by Occupational Category, 1971-74 201 7.1 Distribution of Farms, by Size, 1974 206 7.2 Value Added in Agriculture, Forestry, and Fishing, 1966-76 207 7.3 Average Area, Yield, and Production of Foodgrains, 1965-76 208 7.4 Investment Program for Agriculture, 1972-81 210 7.5 Targets of the Fourth Plan for Agriculture, 1975-81 211 7.6 Food Consumption, Korea and Japan, Selected Years 212 7.7 Food Consumption, Projections for 1975-90 213 7.8 Parameters of Additional Land Development, 1977 215 7.9 Cumulative Area of Improved Land to be Available for Cropping under Three Levels of Investment, 1977-86 217 7.10 Increase in Annual Grain Production under Alternative Yields 218 7.11 Rice Production under Three Levels of Investment, Projections for 1981 and 1990 220 7.12 Foodgrain Production, Demand, and Import Requirements, Projections for 1990 221 7.13 Composition of Farm Household Receipts, by Source, 1969-75 225 7.14 Agricultural Labor Requirements, by Month, Estimates for 1975-90 225 7.15 Annual Income of Farm and Urban Households, 1963-75 227 8.1 Observed Structural Shares in Korea, 1955 and 1972 241 8.2 Norm Structural Shares 241 8.3 Sources of Growth of Aggregate Output, by Sector, 1955-68 243 8.4 Direct Contributions to Growth in Manufacturing, by Sector, 1960-73 245 8.5 Structural Indicators, 1965-81 254-55 8.6 Composition of Planned Fixed Investment in Manufacturing, by Sector, 1967-81 256 8.7 Composition of Actual and Planned Investment in Manufacturing, by Sector, 1972-8 1 256 CONTENTS Xlii 8.8 Composition of Manufacturing Value Added, by Sector, 1960-81 257 8.9 Composition and Growth of Manufactured Output and Exports, by Sector, 1975-81 258 8.10 Investment and Value Added in Manufacturing, 1967-81 261 9.1 Comsumption of Commercial Energy, by Sector, 1966-74 283 9.2 Energy Consumption, Energy Intensity, and Gross National Product, 1966-74 284 9.3 Electricity Sales and Distribution, 1966 and 1974 290 9.4 Transmission and Distribution Facilities, 1974 291 9.5 Consumption of Petroleum, by End-use Sector, 1966-74 292 9.6 Consumption of Commercial Energy, by End-use Sector, 1974-81 295 9.7 Investment Program for Energy, 1972-81 297 10.1 Volume of Domestic Traffic, 1965-75 305 10.2 Railway Survey, 1965-75 306 10.3 Highway Survey, 1965-75 307 10.4 Shipping Survey, 1965-75 308 10.5 Volume of Domestic Traffic, 1971-81 315 10.6 Investment Program for Transport, 1977-81 317 10.7 Composition of Transport Investment, by Mode, 1972-81 318 10.8 Transport Investment by the Public Sector, 1975-81 319 11.1 Revenue and Expenditure of General Government, 1963-75 330-31 11.2 Public Savings and Related Factors, 1962-76 334-35 11.3 Budget Expenditure of General Government, 1968-76 338-39 11.4 Government Savings, 1976-81 340 11.5 Government Savings, 1967-81 340 11.6 Consolidated Current Account of the Government Sector, 1976-81 342-43 11.7 Total Tax Revenue, 1976-81 344 11.8 Current Expenditure of Government, 1976-81 350 11.9 Government Expenditure on Capital Account and Its Financing, 1976-81 353 11.10 Tax Buoyancy Ratios, 1976-81 355 12.1 Structure of the Financial System, 1975 366-67 12.2 Specialized Banks: Functions and Sources of Funds 368-69 12.3 Distribution of Assets of Financial Institutions, 1963-75 371 12.4 Measures of Financial Development, 1963-75 373 12.5 Sources of Growth of Financial Institutions, 1963-75 375 12.6 Flow of Funds between Government and Financial Institutions, 1963-75 376 12.7 Public Offerings of Stocks and Bonds, 1968-76 378 Xiv CONTENTS 12.8 Financial Surpluses and Deficits, by Sector, 1963-76 380 12.9 Composition of Gross Financial Investments of Households, 1963-76 382 12.10 Composition of Sources and Uses of Corporate Funds, 1963-76 384 12.11 Financial Surpluses and Deficits, by Sector, 1977-81 392 12.12 Gross Financial Investments of Households, 1972-81 395 12.13 Growth of Assets and Liabilities of the Banking System, 1972-81 398 B. 1 Export Performance of Selected Commodities, 1965-75 432-33 B.2 Export Performance of Selected Commodities, 1967 and 1975 434 B.3 Export Performance of Selected Commodities, 1970 and 1974 436 B.4 Market Shares in Total Korean Exports, 1965-75 437 B.5 Commodity Exports, by End Use, 1965-75 440-41 C. 1 Foreign Savings in Relation to Investment and Gross National Product, 1962-76 443 C.2 Sources and Uses of External Capital, 1966-76 444 C.3 Commitments of Medium- and Long-term Capital, 1965-76 446-47 C.4 Total Inflows of Medium- and Long-term Capital, 1962-76 452 C.5 Composition of Public and Private External Debt, 1961-76 453 C.6 External Resource Requirements, 1976-81 454 C.7 Debt Service, 1977-81 455 D.1-D.50 See separate listing on pages 457-59 Acronyms APSF Agricultural Price Stabilization Fund BOK Bank of Korea CIF Cost, insurance, and freight CNB Citizens National Bank EEC European Economic Community EPB Economic Planning Board FF r- Fertilizer Fund FIR Financial intermediation ratio FOB Free on board GDP Gross domestic product GMF Grain Management Fund GNP Gross national product GRT Gross registered tonnage GSF Government Supply Fund ICOR Incremental capital-output ratio IDA International Development Association IEC Information, education, and communication IFC International Finance Corporation ILO International Labour Organisation IMF International Monetary Fund isic International Standard Industrial Classification ISWACO Industrial Site and Water Resources Development Corporation IUD Intrauterine device KAL Korean Air Lines HAP Knowledge, attitude, and practice KASS Korean Agricultural Sector Study KDA Korea Development Association KDB Korea Development Bank KDI Korea Development Institute KEB Korea Exchange Bank xv Xvi ACRONYMS rECO Korea Electric Company KEDI Korea Educational Development Institute KFW Kreditanstalt fur Wiederaufbau RHC Korea Highway Corporation KIC Korea Investment Corporation KIFP Korean Institute of Family Planning KITC Korea Investment Trust Company KMPA Korea Maritime and Port Authority KMPC Korea Mining Promotion Corporation KNR Korea National Railroad KSE Korea Stock Exchange KSFC Korea Securities Finance Corporation L/C Letter of credit LDC Less developed country LFPR Labor force participation rate MAF Ministry of Agriculture and Fisheries MCI Ministry of Commerce and Industry MER Ministry of Energy and Resources MIB Medium Industry Bank MOF Ministry of Finance MOHA Ministry of Home Affairs MOHISA Ministry of Health and Social Affairs MOT Ministry of Transport MST Ministry of Science and Technology NACF National Agricultural Cooperatives Federation NIF National Investment Fund NWP National Welfare Pension OECD Organisation for Economic Co-operation and Development OECF Overseas Economic Cooperation Fund OLA Office of Labor Affairs PL480 (U.S.) Public Law 480 PPFK Planned Parenthood Federation of Korea SDR Special drawing right SITC Standard International Trade Classification STFC Short-term finance company TCC Transport Coordination Committee TCO Transport Coordination Office TOE Tons of oil equivalent UMM Unorganized money market Unesco United Nations Educational, Scientific, and Cultural Organization ACRONYMS XVii USAID United States Agency for International Development usccc United States Commodity Credit Corporation VAT Value-added tax VTn Vocational training institute WGT Working group on transportation i Acknowledgements THIS REPORT, put into final form in February 1977, is the product of economic work done by the World Bank on Korea during 1975 and 1976. Preparatory analysis was undertaken in the first half of 1976 in conjunc- tion with missions on human resources by TimQthy. .i>g.athqrine Pierce, Fredrick Qolladay, and Ricardo Moran; industr by yAr_y_West-= phal, Vinod Prakash, and Manuel Penalver-Quesada; macroeconomic modeling by Syamaprasad Gupta and Ronald Padula; energy by Philippe Bourcier, 0. B. Fall, William Helt, and Saied Quraeshi; agriculture by George Ross Miller and George Tolley; transport by Tillman Neuner; public finance by Marinus van der Mel; and the financial sector by Agustin Que. The main economic mission visited Korea in June and July of 1976. Members of this mission were Paez Has_n, chief; ICLJs deputy chief; ayatiDatta-Mitra, general economist; Syamaprasad Gupta, econ- ometrician; George KAu, general economist; Timothy King, human resources specialist; Ronald Padula, research assistant; Robert Pearse, consulting agricultural economist; Manuel Penalver-Quesada, industrial economist; Agustin Que, financial sector specialist; Michael Sarris, gen- eral economist; Marinus van der Mel, public finance expert; Larry Westphal, industrial economist; and Virginia Menmuir, secretary. The principal authors of individual chapters are identified in the volume. The coordinating authors nevertheless accept responsibility for the entire report. The help and hospitality received in Korea are gratefully acknowl- edged. Special thanks are due to the Economic Planning Board and the Korean Development Institute. Valuable assistance was received from many individuals, and to all of them sincere thanks are extended. Special mention must nevertheless be made of Suh Suck Joon, Kim Mahn Je, and Kim Jae Ik. The work on various sectors benefited greatly from studies financed by the United Nations Development Programme under a technical xix xx ACKNOWLEDGMENTS assistance project to help the Korean government in the formulation of its development plan for 1977-81. Executed by the World Bank, the project enabled a far more detailed analysis of many sectors than would otherwise have been possible. David Loos, division chief for Korea in the World Bank, reviewed the entire manuscript and made valuable suggestions at all stages of preparation. William Bussink, senior economist, also was most helpful in the internal review. Bruce Ross-Larson edited the manuscript for publication. Brian J. Svikhart managed design and production of the book and Florence Robinson prepared the index. PARVEZ HASAN D. C. RAO Korea Parvez Hasan and D. C. Rao Overview EcONOMuIC ANALYSTS frequently use such adjectives as remarkable, ex- traordinarv, and spectacular to describe Korea's economic performance during the 1960s and early 1970s. One reason for this effusiveness is that the Korean economy has consistently achieved and often surpassed the ambitious targets set by planners for economic growth and develop- ment. A second reason is that Korea managed this performance despite considerable odds. The country has no significant reserves of natural resources; the population density is among the highest in the world; most of the physical infrastructure was destroyed during the Korean War. But its social homogeneitv, sound policies, and strong political leadership, combined with a favorable international environment and a firm commitment to growth, enabled Korea to overcome such obstacles and create a success story in economic development. The rates of economic growth in Korea have been matched only by the oil-exporting countries; the level of infrastructure development is among the highest in the developing world; and social services, left largely to market forces in the private sector, have raised substantially the standards of both rural and urban areas. This success has enabled govcrnment planners to turn from their dominant concern with economic growth and address a broad array of social and economic issues. Given the requirements for greater attention to considerations of equity, for structural changes to maintain the comparative advantage of Korean exports, and for new roles to be undertaken bv government in response to the changing domestic and external conditions expected in the late 1970s and the 1980s, the question now is whether Korea's rapid eco- nomic growth can continue. The Pattern of Rapid Economic Growth, 1961-76 In 1961 Korea was one of the poorest developing countries, with heavy dependence on agriculture and a weak balance of payments 3 4 OVERVIEW financed almost entirely by foreign grants. By 1976 it had become a semi-industrial, middle-income nation with an increasingly strong ex- ternal payments position and the prospect of eliminating its current account deficit in the late 1970s. GNP grew over the period at the average rate of more than 10 percent a year, and per capita income tripled in real terms. The manufacturing sector grew at 18 percent a year in con- stant prices, increasing its share in GNP from 14 percent to 30 percent. At the same time the share of agriculture fell from 40 percent to less than 25 percent. Exports, principally manufactured goods, rose from $41 million in 1961 to nearly $8 billion in 1976. The average real growth of exports was 33 percent a year, and exports in 1976 constituted 36 percent of GNP. Rapid industrial growth, relying increasingly on expansion of labor- intensive manufactured goods, contributed to high rates of growth of employment. More than 2 million jobs, constituting more than 40 percent of total additional employment, were created in industry during 1963-76. The overall rate of employment growth of 3.9 percent a year during this period was substantially higher than the growth in labor force, which also was rising rapidly. The significant improvement in the employment situation is further indicated by the decline in the open unemployment rate from 8.2 percent in 1963 to 3.9 percent in 1976 and the average increase in real wages of more than 7 percent a year during the period. The rising demand for labor in the industrial and urban sectors caused heavy outmigration from the rural areas. Conse- quently growth in agricultural employment, at 1 percent a year, was relatively slow. Meanwhile value added in agriculture grew at more than 4 percent a year, indicating a fairly respectable rate of productivity growth in this important sector, which still employs more than 40 per- cent of the labor force. This performance, combined with deliberate government policies after 1967 to improve the agricultural terms of trade, tended to ensure growth in per capita rural incomes more or less in line with the rest of the economy. In addition, substantial government efforts were directed toward improving the quality of rural life by providing such infrastruc- ture as roads, electricity, telephones, and piped water. The broad sharing of the benefits of growth in Korea is reflected in data on the distribution of income and the progress toward meeting basic needs. The distribution of income in Korea compares favorably with most developing countries. Although the share of the bottom 40 percent of income recipients ap- parently declined somewhat in recent years, their per capita consumption in absolute terms still increased by at least 75 percent over 1961-76. OVERVIEW 5 The improvement was also significant in nutritional standards, adult literacy, life expectancy at birth, and the infant mortality rate. What caused this transformation? Korea's outstanding economic and social gains are attributable to the interaction of several factors whose relative importance cannot easily be gauged. Political stability, strong leadership, and a firm commitment to development were the pre- requisites for economic advance. Large inflows of foreign grants and loans were important, particularly in the early stages of industrializa- tion. The social environment was also conducive to rapid economic progress. For various historical and cultural reasons the society of post- war Korea was more homogeneous and less structured than in most other parts of Asia. There were no strong regional or religious differ- ences, no deeply entrenched class or caste structures. Consequently the pursuit of economic opportunity was less inhibited by tradition and class, and the mobility and adaptability of labor were relatively un- restricted by social constraints. Furthermore Korea's labor force in the early 1960s was well educated, disciplined, and industrious. Perhaps most important, however, was the early recognition by Korean policy- makers of the capacity of trade in manufactured goods to compensate for the paucity of natural resources. This recognition led to the formu- lation of a development strategy concentrating on the expansion of labor-intensive manufacturing for export. The beneficial impact of the orientation toward labor-intensive ex- ports can be seen in the factor proportions and efficiency of factor use in manufacturing. The sector is typified by low capital-output and capital-labor ratios and high rates of productivity growth. Despite the increased importance of steel, petrochemicals, shipbuilding, and ma- chinery in the first half of the 1970s, there was no perceptible increase in the incremental capital-output ratio for the manufacturing sector. This pattern reflects not only the heavy emphasis on labor-intensive goods, but also low construction costs and high rates of capacity use which, in turn, reflect good management and organization. Even though the increase in the average capital employed per worker has been small, labor productivity grew at an average rate of about 7 percent a year during the 1966-76 period, indicating changes in output mix and pro- ductivity improvements in existing industries. The high and steadily increasing level of labor productivity has been a fundamental strength of the Korean development process. The point is worth stressing, however, that efficient use of capital and labor can be related only in part to the export orientation of Korean manufacturing. Indeed the contribution of exports to manufacturing 6 OVERVIEW growth and output during the 1960s was comparatively modest. The share of exports, though growing rapidly, was still quite small. Even in 1970 gross manufactured exports constituted only 15 percent of manu- facturing output. After 1970 the contribution of export expansion to manufacturing growth became much greater. By 1976 the ratio of man- ufactured exports to manufactured output had risen to 25 percent, but the fact that three-fourths of industrial output was for the domestic market suggests that productivity improvements in home-goods indus- tries must also have been widespread. One factor is sometimes overlooked in reviewing Korea's economic performance: agriculture already was well advanced by the early 1960s. This enabled government to devote more attention and resources to the urban industrial sector. Land reform, which in the late 1940s fixed a ceiling of three hectares on landholdings, was important to the mod- ernization process. This reform not only brought about greater social equality through the redistribution of three-quarters of cultivable land, it removed the feudal obstacles that so often hinder agricultural devel- opment. By 1961 more than 2 million hectares, or virtually all of Korea's cultivable land, were under intensive cultivation and producing 92 percent of the country's requirements for grain. Average yields of rice were about 2.3 tons a hectare, or nearly one-third more than yields now obtained in Southeast Asia. Starting with this strong base, the country's agricultural sector made tangible and impressive gains. The extensive use of modern farming techniques, particularly water management and the application of fer- tilizer, and the development of local adaptations of high-yielding rice varieties made possible a steady increase in yields. In 1976 average rice yields exceeded 4.3 tons a hectare and yields up to 6 tons were not uncommon. These yields were among the highest in Asia and were ex- ceeded only in Japan. The continued development of the agricultural sector was crucial in ensuring that rural and urban disparities did not substantially widen. Korean economic development has relied heavily on foreign financing of investment outlays and balance-of-payments deficits. The deficit in the current account of the balance of payments averaged 7.2 percent of GNP during the second plan period 1967-71 and 5.7 percent during the third plan period 1972-76. The proportion of investment financed from foreign inflows, excluding transfers from abroad, was close to 30 percent in the second plan period and 25 percent in the third plan period. The heavy investment required to sustain Korea's rapid growth thus was well beyond Korea's own saving capabilities. This condition was not, how- OVERVIEW 7 ever, the result of any failure of the national saving effort. Nor did large external inflows weaken the domestic mobilization of financial resources. Indeed the ratio of national savings to GNP rose from about 2 percent in the early 1960s to about 20 percent during 1975-76. The marginal saving rate averaged about 25 percent. Because investment demand was growing rapidly, reducing the reliance on external resources was not possible, even with a substantial domestic saving effort. Until the mid-1960s, a significant proportion of foreign capital was in the form of grants. Since then, however, most of the external pay- ments deficit has been financed by foreign borrowing. Korea's outstand- ing medium- and long-term debt rose from about $300 million at the end of 1966 to about $7 billion at the end of 1976. The real burden nevertheless remained moderate, mainly because of a thirtyfold increase in export earnings over the period. The ratio of debt service payments to exports of goods and nonfactor services, after averaging about 21 percent in the early 1970s because of the rather extensive use of supplier credits in the late 1960s, declined to an estimated 10 percent in 1976. The Korean strategy had an element of risk. If exports had faltered, or if resources had not been efficiently used, the influx of foreign resources to finance industrial development would almost certainly have slowed down. That influx did not slow down because economic policies gen- erally were extremely effective and targets for the growth of exports and GNP were almost invariably exceeded. Can Rapid Economic Growth Continue? In many ways Korea entered the fourth plan period 1977-81 with a stronger economic position than ever before. Despite the balance-of- payments difficulties of 1974 and 1975 and a much less favorable inter- national environment than expected, growth in GNP averaged 11 percent a year during the third plan period. Although the net growth in world trade in manufactures was sluggish during 1973-76, Korea's export volume nearly doubled. At the same time the success has been great in obtaining construction contracts in the Middle East. Net earnings from this source were close to $250 million in 1976 and growing. Reflecting these developments, the deficit in the current account of the balance of payments was sharply reduced to 1 percent of GNP in 1976 and is likely to remain within modest limits during the fourth plan period. Parallel to this reduction, the domestic saving rate has been rising 8 OVERVIEW sharply, in large part because of aggressive fiscal policies and a strong recovery in public savings. Resource management problems might nevertheless persist because of the tendency of Korean planners to underestimate investment require- ments and overestimate future domestic savings potential. Even if the fourth plan's targets for growth and saving are met, the need for gross external inflows will continue to be sizable. But problems of mobi- lizing external capital are likely to be less serious than in the past. Korea's creditworthiness has considerably improved. The burden of external debt is being steadily reduced. Moreover the relatively success- ful adjustment of the Korean economy to higher intemational oil prices and the world recession has increased the confidence of govemment and business in their own capabilities. The fourth plan projects that GNP will increase by 9.2 percent a year during the 1977-81 period. The share of manufacturing in GNP is pro- jected to rise from 28 percent to 40 percent; manufacturing is to provide 54 percent of incremental GNP. Aggregate merchandise exports are to grow at 16 percent a year, rising from $8 billion in 1976 to $14 billion in 1981 in constant 1975 prices. To bring about structural changes in exports and industrial output, Korean planners are focusing on machine- ry, electronics, and ships; the proportion of this group in total exports is expected to increase from 16 percent in 1975 to 31 percent in 1981. If an overall growth rate of 9 percent a year is to be sustained, estimates suggest that investment as a proportion of GNP must rise from 26 percent in 1976 to 27.5 percent in 1981. This investment will require an average marginal saving rate of 31 percent over the period. All these targets, while ambitious, are attainable. The chances are also good that Korea will be able to sustain a long-term GNP growth rate of 9 to 10 percent a year through the 1980s. Sustain- ing such a growth rate is nevertheless contingent on three fundamental conditions: that the world economy continues to grow at roughly the same rate as in the last decade; that planned efforts toward export diversification and structural change in industrv are successful and facilitate high rates of export growth; that government policies continue to attach high priority to a wider distribution of growth benefits so as to ensure a stable social order. If it is assumed that the international economic environment will remain reasonably favorable, the principal long-term issues for Korea are likely to revolve around exports and the distribution of income. Rapid expansion of exports will continue to be critical in determin- ing the economic growth path in the 1980s. If world trade in man- OVERVIEW 9 ufactures grows at 8.5 percent a year, compared with average growth of 9.5 percent a year during 1960-73, Korea's share in this trade will have to rise from nearly 1 percent in 1975 to nearly 3 percent in 1990 in order to sustain export growth of 16 percent a year. This share will not be easily achieved, but it is attainable provided government con- tinues its efforts to further diversify Korean exports and maintains a favorable domestic policy environment. Although exports of ships, steel, and metal products have become large, textiles, clothing, electronics, and footwear-all of which are labor-intensive-still constitute more than one-half of manufactured exports. Even though the labor force will continue to grow at the fairly high rate of 2.7 percent a year during the 1980s, Korea's comparative advantage will start to shift from products intensive in labor to products intensive in capital and skills. In the long run a changing pattern of exports, reflecting and respond- ing to shifting comparative advantage, is essential for sustained export expansion. Although the threat to Korea's international position in light manufactured goods is not immediate and the base of its products and markets is fairly secure and diversified, the rate of growth of this group will come down sharply in the 1980s as competition from other developing countries becomes more intense. Korea is also beginning to encounter limits to the expansion of several traditional manufactured exports because of its dominant position in some markets and because of trade barriers in the developed countries. Unless new exports are vigorously developed, particularly in machinery and electronics, Korea cannot hope to capture an increasing share of the world market of manufactured goods. Although the prospects for sustaining rapid export growth are favor- able, continuing reliance on a strategy of export-oriented growth might be criticized on the ground that it makes the economy excessively vulnerable to external shocks lying entirely outside Korea's control. But given the country's poor natural resource base and relatively small do- mestic market, it is inevitable that economic growth in the 1980s will be heavily dependent on the expansion of exports. True, exports will rise sharply as a proportion of GNP and may conceivably exceed GNP by 1990, but this ratio is not the best indication of the export dependence of the Korean economy. The ratio of manufactured exports to gross manu- factured output, a more appropriate indicator, is projected to rise to about 30 percent in 1981 and 40 percent in 1990. The problems associated with export dependence and vulnerability thus do not appear to be grave. The growing dependence on exports nevertheless has some major policy implications. The Korean economy will be increasingly suscep- 10 OVERVIEW tible to fluctuations in international demand and prices, implying that the problem of managing aggregate demand will be a major long-term concern for Korean policymakers. What is called for is the maintenance of realistic exchange rates, adequate international reserves, and more generally a reasonably flexible policy posture in the face of numerous uncertainties. Considerable care will be needed to avoid recurrent balance-of-payments problems which would disrupt the development process. A second major implication of the continuing export orienta- tion is the need to liberalize imports and adopt commercial policies in keeping with Korea's emerging position as an important trading nation. The use of protective barriers to establish import-substituting activities that are not internationally competitive could jeopardize the growth of Korean exports in two ways: by raising production costs of exporting industries, and by engendering resistance among Korea's trading part- ners. Liberalizing imports will enable Korea to relax some of the existing restraints on consumption that were imposed to protect the balance of payments. For Korea, continued rapid growth seems essential to absorb the large expected additions to the work force and to bring about further reduc- tions in the incidence of povertv. Indeed significant progress toward the basic equity goals of eliminating unemployment and absolute poverty cannot be achievcd at a GNP growth rate of, sav, 7 percent. Paradoxical though it may seem, a high growth rate of 9 to 10 percent a year-while generating greater employment and facilitating the eradication of pov- erty-could still result in a more uneven income distribution than now prevails. There are two main reasons for this: average income among various sectors of the economv will continue to diverge as a result of different rates of productivity growth; the dominant focus of economic activity will be shifting from the rural sector, where incomes are rela- tively evenly distributed, to the urban sector, where income inequalities are much greater. Because the distribution of income is more equal in Korea than in most developing countries, some widening of income disparities might not be a serious problem, particularly if average per capita incomes are rising, at 7 to 8 percent a year. The threat of a widen- ing gap betw.een rural and urban incomes could be more of a problem. Severe land and climatic constraints impose definite limits to further growth in agricultural productivity. The support for agricultural terms of trade already is a drain on public resources and may in the future result in excessive surpluses of rice. The basic problem has been the slow growth of nonagricultural income in the rural sector. Recent in- flows of workers' remittances and related construction activities never- OVERVIEW I I theless appear to have caused a sharp increase in rural income from nonagricultural sources. It seems likely that fiscal transfers will be necessary, but not sufficient, to improve the distribution of income in the 1980s and that other re- distributive measures will be required. Government clearly will have to be more active in directly intervening in matters of equity. At the same time, with the economy growing at 9 percent a year, government will have more resources to apply to such redistributive measures as direct transfers. These transfers would have to be augmented with additional redistributive devices to serve the poorest segments of society with piped water, electricity, paved roads, and other basic services. It may also be necessary to develop public subsidies that ensure the cheap availability of goods normally purchased by the poor and thereby raise their effective real earnings above the poverty level. Obvious targets for such a sub- sidized program of poverty alleviation are food, housing, health services, education, and transport. In summary, Korea will be a fairly high-income, heavily industrial- ized, export-dependent economy with a rapidly dwindling reliance on agriculture in 1990. The population in that year will be about 47 mil- lion, or nearly one-third more than in 1976. If it is assumed that average GNP growth will be 8 to 10 percent a year, per capita income in 1975 prices will have increased to between $1,200 and $1,800. Manufactur- ing is likely to account for about 40 percent of cNP, compared with about 30 percent in 1976; agriculture will contribute only about 10 percent, compared with more than 20 percent in 1976. If a rate of eco- nomic growth of more than 9 percent a year can be sustained and if special redistributive measures are adopted, Korea can look forward to full employment and the elimination of absolute poverty by 1990. If these changes occur, as indeed seems probable, Korea will have ac- complished the transformation from a low-income developing economy to a developed economy in just thirty years. PART ONE The Main Report Chapter 1 D. C. Rao The Pattern of Economic Growth, 1961-76 KOREA is relatively small and densely populated.' Most of the country's total land area of about 99,000 square kilometers is mountainous: only about 23 percent of the land is cultivated; about 67 percent is forested mountain slopes; cities, industrial facilities, and roads occupy the re- mainder. The rugged topography and harsh winters limit the productive potential of agriculture. Mineral resources are also limited. The princi- pal mineral, anthracite coal, is available in great quantity but is poor in quality and unsuitable for coking. The pressures of population on these limited resources are intense. Korea's population density of 363 persons per square kilometer of land and 14.8 persons per hectare of farmland is among the world's highest.2 Rising levels of education and income and a strong effort to disseminate family planning services brought down the rate of population growth from about 3 percent a year in 1961 to 1.7 percent a year in 1976. Even so, the population will continue to grow rapidly because of the aftereffects of the baby boom that followed the Korean War. During the 1961-76 period Korea's economy grew at an average rate of about 10 percent a year, accelerating from 9 percent a year in 1963- 66 to 10.5 percent in 1967-71 and 11 percent in 1972-76, the period of Korea's third plan. Per capita income in current prices rose from less than $100 in 1961 to about $700 in 1976.3 In real terms the rise was 1. The Republic of Korea, also known as South Korea, is referred to in this book as Korea. 2. In 1975 the population density of Japan was 294 persons per square kilo- meter; that of the Netherlands 333. 3. All dollar figures in this book are U.S. dollars. The Korean currency unit is the won. At the time of writing Wl,000 were equivalent to $2.06; $1 was equivalent to 485 won. Large numbers in this book are in accord with American usage: I billion is equal to 1,000 million; 1 trillion to 1,000 billion. I5 I6 PART ONE: THE MAIN REPORT Table 1.1. Composition and Growth of Gross National Product, 1961-76 (percent) Composition of GNP Annual growth rate Sector 1961 1966 1971 1976 1961-66 1966-711971-76 Agriculture, forestry, and fishing 44.1 38.9 26.5 20.4 5.1 2.3 5.4 Mining and manufacturing 12.0 15.9 24.4 35.9 14.2 20.3 19.9 Social overhead, 6.1 9.0 12.9 13.7 16.9 18.7 12.2 Other' 37.8 36.2 36.2 30.0 6.7 10.4 6.9 Total 100.0 100.0 100.0 100.0 7.7 10.5 11.9 Note: All figures are in 1970 prices. Sources: Table D.7 in appendix D; Economic Planning Board (EPB), The Fourth Five-Year Economic Development Plan, 1977-81 (Seoul, 1976). a. Includes construction, transportation, storage, communication, electricity, water, and sanitary services. b. Includes trade, banking, insurance, real estate, ownership of dwellings, public administration, defense, other services, and net factor income from abroad. nearly threefold. The sustained high rate of expansion in incomes trans- formed Korea from one of the poorest developing countries, with heavy dependence on agriculture and a weak balance of payments financed almost entirely by foreign grants, to a semi-industrial, middle-income nation with an increasingly strong external payments position. Rapid growth transformed the structure of the economy (table 1.1). Despite average annual growth of value added in agriculture of about 4 percent, agriculture's share in GNP declined dramatically-from 44 percent in 1961 to an estimated 20 percent in 1976. The primary reason for this decline was the manufacturing sector's growth of nearly 19 percent a year, almost double the growth rate of aggregate GNP. The share of manufacturing in GNP rose from 12 percent in 1961 to 36 percent in 1976. Social overhead sectors, growing at about 15 percent a year, maintained their share in GNP, but the share of other services declined. Exports The leading edge of this extraordinary expansion was the growth of manufactured exports. In 1965 commodity exports were $175 million, THE PATTERN OF ECONOMIC GROWTH 17 of which $144 million consisted of manufactured goods.4 By 1976 ex- ports had risen to nearly $8 billion, having grown at an average rate of nearly 40 percent a year in current prices and 32 percent a year in constant prices. As a proportion of GNP, exports of goods and services rose in current prices from 8.5 percent in 1965 to 15 percent in 1970 and 36 percent in 1976. Although exports of primary products rose rapidly during the 1960s, the growth of manufactured exports over- shadowed their performance. Even in 1960 manufactures constituted 65 percent of commodity exports. In 1965 their share had risen to 82 percent; in 1976 their share was about 90 percent. The composition of manufactured exports also shifted. Despite the continuing expansion of traditional light industry products, a trend emerged in the direction of sophisticated, skill-intensive products. The change was relatively slow, not because of the neglect of new exports, but because of the continued buoyancy of such traditional exports as textile yams, fabrics, and clothing, which maintained at least a 35 per- cent share in total commodity exports after 1970. Several new products were developed even in the light industry sector: travel goods, hand- bags, footwear, and plastic articles. More important, exports of other manufactures were growing rapidly. Exports of electrical machinerv, mostly electronics, made their mark as early as 1970. Subsequently an entire range of commodities demanding more sophisticated production techniques gained in prominence. Among these were transport equip- ment (including ships), industrial machinery, metal products, chemi- cals, and precision instruments. By 1976 Korea had developed a sizable export volume in a fairly wide range of manufactures. The United States and Japan took the bulk of Korean exports during the 1961-76 period. The United States is by far the largest single market for the manufactured exports of developing countries, having taken 30 percent of the total in 1975. Japan, because of geographic proximity and historical ties, is equally important for Korean exports and even more so for Korean imports. In recent years, however, Korea has made vigorous efforts to develop other markets. Europe's share of Korean ex- ports expanded from 9 percent in 1970 to 17.5 percent in 1976. The Middle East also emerged as a significant market, having taken 7 percent of Korea's commodity exports in 1976 and a rapidly growing volume of 4. The term, manufactured exports, encompasses all exports other than those of food and live animals, beverages and tobacco, mineral fuels, lubricants and related materials, and animal and vegetable oils and fats. This definition is con- sistent with that used by Korean planners. I8 PART ONE: TIE MAWN REPORT construction activity by Korean firms using Korean labor. Until 1974 the combined share of the United States and Japan was about 70 per- cent, but in 1976 it declined to 56 percent. Although it is obvious that manufactured exports acted as the engine of growth, the question remains: What made the export-led industrial strategy so successful? The rapid growth of world trade between 1955 and 1975 clearly helped, but one striking feature of Korea's export per- formance was the ability to exploit market opportunities to the fullest extent. Korea outstripped many of its competitors and raised its share in developing countries' exports of manufactured goods from 2 percent in 1965 to about 16 percent in 1976. Responding to the boom in indus- trialized countries, Korea increased its export volume by 61 percent in 1973 and nearly doubled its export earnings. After the world currency realignments and the shifts in Japan's trade policy, Japanese imports increased by a record 44 percent in 1973. Korean exports to Japan in that year increased by more than 200 percent. Commodity exports to the Middle East increased from negligible amounts to more than $300 mil- lion in 1975, and Korea won construction contracts, mostly in the Middle East, worth about $2.5 billion in 1976. Although Korea's prin- cipal exports-textiles and clothing-were subjected to quota restrictions in recent years, exporters adapted by upgrading quality and switching to categories not subjected to quota provisions. It is difficult, however, to explain the phenomenal success of Korea's export expansion only by market signals and entrepreneurial drive. Government policies provided the appropriate impetus and framework for exports in line with Korea's comparative advantage. In contrast with other developing countries, Korea's incentive structure favored export sales over import substitution. A broad range of measures raised the profitability of export sales, but provided relatively little protection in the domestic market. Rebates of indirect taxes and duties on inputs ensured that exporters obtained material inputs and made overseas sales as if they operated in a free trade area. In addition, exporters were entitled to subsidized credit and electricity. They also were granted generous wastage allowances in determining the allocation of raw ma- terial imports free of duties and indirect taxes. This permitted them to use some of these imports to produce for sale on the domestic market at high rates of profit. Furthermore the so-called export-import link system entitled exporters to import certain popular items not otherwise approved for import. Although the general price level rose faster in Korea than among its principal trading partners, the level of incentives was varied over time THE PATTERN OP ECONOMIC GROWTH I9 so that, combined with changes in the nominal exchange rate, the real effective exchange rate for exports remained fairly stable. In addition to setting economic incentives in favor of exports, government exercised considerable powers of moral suasion to ensure that manufacturers maxi- mized exports. Detailed targets for exports were specified, usually in consultation with manufacturers' and traders' associations, and moni- tored at the highest levels in government. This dedication to export performance enabled Korea to increase its export volume by 20 percent under the adverse market conditions of 1975. In determining the level of industrial incentives, government was conscious of the implications for resource allocation. The finding that export profitability was unreasonably high in 1973 led to the abolition of the export-import link system and a reduction of wastage allowances. In order to remove an impediment to the development of an indigenous machinery sector, government abolished the tariff exemption for im- ported machinery. The nominal and effective rates of protection were low by international standards and relatively uniform among sectors. Consequently Korea's industrial growth and export expansion exploited its comparative advantage. Resource allocation along the lines of com- parative advantage meant two things: labor-intensive manufactured exports paid for capital-intensive imports; more important, the share of trade in total economic activity increased tremendously and led to rapid growth of employment and incomes. The beneficial impact of the orientation toward labor-intensive exports can be seen in the factor proportions and the efficiency of factor use in manufacturing. The sector is typified by low capital-output and capital- labor ratios and by high rates of productivity growth. The unlagged incremental capital-output ratio (ICOR) for the manufacturing sector during 1965-76 was 1.3 in 1970 prices.5 There was no perceptible in- crease in this ratio during 1970-76 despite the increased relative impor- tance of steel, petrochemicals, shipbuilding, and machinery. This per- formance reflects not only the heavy emphasis on labor-intensive goods, but low construction costs and high rates of capacity use which, in turn, reflect good management and organization. The average cost of job creation in manufacturing, after falling during 1960-66, was stable 5. The icoR is conventionally measured by dividing one year's investment by the change in output in that or the succeeding year. Periods longer than one year are used to calculate lagged IcoRs. The ICOR serves as an indicator of the capital cost of additional output and is inversely related to the productivity of capital. See chapter 8 for a more detailed discussion. 20 PART ONP: THE MAIN REPORT during 1966-76 at about $5,000 in 1975 prices. Labor productivity grew at an average rate of about 7 percent a year during 1966-76, reflecting changes in output mix and improvements in productivity in existing industries. The high and sharply increasing level of labor productivity was a fundamental strength of the Korean development process during the entire 1961-76 period. Real wages grew in line with increases in labor productivity, and wage costs remained internationally competitive. In 1976 manufacturing wage rates in Korea were lower than those in Hong Kong, Singapore, and the Republic of China (Taiwan), and higher than those in the Philippines, Malaysia, and Thailand. Although wage rates in Korea were about one-fifth of those in Japan, Korean productivity levels in the traditional industries probably did not lag too far behind Japanese productivity. The main explanation for this per- formance lies in the high levels of education and skills and in the strict labor discipline. The efficient use of capital and the steady improvement in labor productivity enabled the Korean economy to maintain a G-NP growth rate of nearly 11 percent a year, while investing only about 26 percent of GNP on average. Manufacturing Korea's remarkable export performance influenced the pace of in- dustrialization and the structure of the manufacturing sector. Between 1962 and 1975 value added in manufacturing rose in constant prices at an annual rate of 18.7 percent. During this period changes in the composition of the sector were extensive (table 1.2). The share of food- processing industries in output declined; the share of the heavv manu- facturing sector rose. The most striking increase was in the share of machinery and equipment after 1971. This sector includes electronics, one of the leading commodities in the growth of exports. Growing exports also sustained the shares of textiles and light manufactures. In the pace of its industrialization and the diversity of its manufac- turing activity, Korea was well ahead of other countries of comparable size, income, and other structural characteristics.6 The shares of the chemical and metal sectors in Korean manufacturing were relatively 6. This conclusion is based on a cross-country regression analysis described in chapter 8. THE PATTERN OF ECONOMIC GROWTH 21 Table 1.2. Composition of Gross Manufacturing Output, 1961-75 (percent) Commodity group 1961 1966 1971 1975 Textiles, clothing, and footwear 28.5 23.9 25.6 32.1 Food, beverages, and tobacco 33.8 27.6 25.0 17.0 Various light manufactures, 15.4 17.9 16.0 15.3 Machinery and equipmentb 7.9 10.2 9.3 14.6 Chemicals, coal, and petroleum 7.1 11.6 15.5 12.0 Basic metals and metal products 5.8 7.0 6.1 6.7 Miscellaneous 1.5 1.9 2.5 2.5 Total 100.0 100.0 100.0 100.0 Total manufacturing output (billions of won in 1970 prices) 393.5 789.3 2,041.7 4,290.9 Source: Bank of Korea (BOK), National Income in Korea, 1975, pp. 202-03. a. Includes wood and wood products, furniture, paper and paper products, printing, leather and rubber products, clay, glass, and stone and plastic products. b. Includes electronics and transport equipment. high, even given the dominance of exports. Thus, while the export orientation was the dominant influence on the structure and rate of expansion of the manufacturing sector relative to GNP, the basic inter- mediate sector also expanded in response to the growth of domestic demand and the selective substitution of imports by domestic produc- tion. As a result, aggregate dependence upon imports of intermediate inputs and investment goods was maintained at roughly the same level for the 1965-75 period. A breakdown of the growth of manufactured output by sources of direct demand shows that the direct contribution of exports increased from 5 percent in 1960 to 36 percent in 1973 and averaged 26.8 percent for the entire period (table 1.3). Import substitution, on the other hand, contributed little, and its low share is striking. Furthermore import substitution appears to have generated indirect demands on sectors having higher-than-average requirements for imported intermediate in- puts. Import substitution nevertheless played a significant role in some sectors in certain time periods. In the early 1960s domestic production substantiallv replaced impOTtS of light consumer goods-such as textiles, apparel and accessories, sewing machines, and paper products-and those of some products in the metal and chemical sectors. Between 1968 and 1973 import substitution was significant in synthetic fibers, fertilizer, iron and steel, finished metal products, transport equipment, nonelectri- cal machinery, and chemicals. Much of this import substitution resulted 22 PART ONE: THE MAIN REPORT Table 1.3. Direct Contributions to Growth in Manufacturing, 1960-73 (percent) Source of growth 1960-63 1963-66 1966-68 1968-70 1970-73 Export expansion 4.8 14.2 12.4 15.1 35.9 Import substitution -2.8 0.2 11.4 7.2 -4.0 Note: These estimates, in current domestic prices, are obtained from 38-sector level data for the production of all goods and services. Source: Larry E. Westphal and Kwang Suk Kim, "Industrial Policy and De- velopment in Korea," World Bank Staff Working Paper, no. 263 (Washington, D.C.: World Bank, 1977). from the production of basic steel products and petrochemical deriva- tives, which also figured prominently in the expansion of exports during this period. The pursuit of backward linkages, usually from exports, was the strategy underlying these changes in industrial structure. The most striking example is in textiles, where domestic activity moved through synthetic fibers to the production of basic petrochemicals. Other ex- amples are backward integration from television assembly to the pro- duction of basic components and from automobile assembly to the pro- duction of various components. With the possible exception of automo- bile assembly and petrochemicals, where investments probably were too small to capture economies of scale, import-substituting investmentS generally proved efficient. Nevertheless the scarcity of such raw ma- terials as petroleum, iron ore, and certain agricultural products placed limits on the extent to which industrial growth could rely on backward integration. Thus the driving force of the expansion of manufacturing in Korea continued to be export demand. But a number of sectors had further scope for expanding the share of domestic value added in output without violating the principle of comparative advantage. Part of the production of electronics exports continued to be largely an assembly operation with little infusion of sophisticated technology. In shipbuild- ing and machinery Korea had not realized the full advantages of having a local design capability. In metals the output concentrated on high- volume, standardized products, not on sophisticated specialty or precision products. As already pointed out, the direct contribution of export demand to the expansion of the manufacturing sector became increasingly sig- nificant. This measure nevertheless understates the contribution of exports to the manufacturing sector. Adding the indirect-output effect- THE PATTERN OF ECONOMIC GROWTH 23 that is, the demands placed on other sectors for the supply of interme- diate inputs-almost doubles the contribution of the export sector and commensurately reduces the contribution of domestic demand and im- port substitution. There are two additional beneficial effects from export expansion: one is the multiplier effect arising from increased consump- tion and investment of additional income, which shows up in the fore- going calculus as the contribution of domestic demand expansion; the other is the contribution of foreign exchange, which relaxes the import bottleneck and enables increased production. The second effect is evi- dent in the tremendous increase in the use of manufacturing capacity after the takeoff of exports in the mid-1960s.7 The strong orientation toward exports also had some indirect benefits, principally for economic efficiency. Exports in 1976 accounted for about one-quarter of Korea's gross manufactured output and, in such sectors as shipbuilding and electronics, as much as one-half. Consequently maintaining the international competitiveness of export products was a spur to productive efficiency. Although it cannot be conclusively dem- onstrated, this influence probably was instrumental in the doubling of total factor productivitv between 1960 and 1973 and the small decline in the capital-labor ratio. The resulting changes in the composition of output and the choice of technique at the microeconomic level can both be regarded as improvements in allocative efficiency. Given the limited size of the Korean market and the scarcity of natural resources, exports of labor-intensive manufactured products provided the best means for achieving faster economic growth and more employment. Agriculture Consumption of all foodgrains rose at an annual rate of 4 percent from 5.9 million tons in 1961 to 10.2 million tons in 1975.8 The propor- tion of this demand met from domestic production fluctuated between a high of 95 percent in 1966 and a low of 69 percent in 1973. The physical output of rice and barley, the principal grains in the Korean diet, remained fairly constant at 5.9 million tons between 1970 and 7. According to figures presented in chapter 8, the proportion of capacity use in all 8,760 hours of the year increased from 18 percent in 1962 to 32 percent in 1971. 8. Foodgrains, as defined in official Korean statistics, also include soybeans, potatoes, and corn. All tonnage figures in this book are metric tons equal to 1,000 kilograms. 24 PART ONE: THE MAIN REPORT 1974.9 In 1975 the output of rice and barley grew by about 9.2 percent; in 1976 by about 9.5 percent. These increases were only in part the result of favorable weather conditions. Other influences were the spread of high-yielding varieties of rice and the fruition of past investments in land development and water management. Although domestic produc- tion of rice and barley exceeded consumption in 1976, almost all wheat was imported. Wheat is not an economical crop in Korea because the long growing season of known varieties interferes with double-cropping. These gains in agricultural output were achieved despite considerable odds. As already noted, only 23 percent of the land is cultivated; paddy fields constitute 13 percent and upland fields 10 percent. On average about 11,000 hectares of cultivated land are lost each year to nonagricul- tural uses. The losses are offset by the conversion of upland to paddy and the reclamation of land from forests and tidal flats. As a result, the total area of cultivated land rose from 2 million hectares in 1960 to 2.24 million hectares in 1975; it reached a peak of 2.32 million hectares in 1968. With nearly half the land double-cropped, the area under crops was about 3.3 million hectares in 1975. Other indications of the intensity of land use in Korea are invest- ments in drainage, land consolidation, and irrigation. About 86 percent of all paddy land is irrigated; the average consumption of fertilizer reached 208 kilograms per hectare of cropped area in 1976; the use of agricultural chemicals was also widespread. Modem farming techniques enabled a steady increase in yields of rice and barley. In 1976 average rice yields exceeded 4.3 tons a hectare and vields of 6 tons a hectare were not uncommon. These yields were among the highest in Asia and are exceeded only in Japan. In recent years Korea has developed local adaptations of high-yielding rice varieties that are expected to increase yields by another 15 to 20 percent. The area under these varieties, known as Tongil and Yushin, reached 533,000 hectares, or 44 percent of the area under rice, in 1976. Infrastructure The rapid growth of industry and incomes was accompanied by the commensurate growth of infrastructure facilities. Per capita consump- 9. The growth of crops other than rice and barley and the shift from barley to rice accounted for the higher growth of value added in agriculture. THE PATTERN OF ECONOMIC GROWTH 25 tion of electricity increased from 87 kilowatt-hours in 1965 to 471 kilowatt-hours in 1975. Only 13 percent of generating capacity in 1975 was based on hydroelectric power; the remainder was thermal. Korea started to develop its nuclear generating capacity in the early 1970s, and the first nuclear plant was completed in 1977. The transport sector also grew rapidly. Freight and passenger traffic increased by about 12 to 13 percent a year in the 1965-75 period. Paved roads increased more than fivefold, and three-quarters of passenger traffic was by road. The share of the railways in freight and passenger traffic declined. Nevertheless about one-half of freight traffic, consisting mainly of long hauls of bulk commodities, still went by rail in 1975. Coastal shipping expanded rapidly between 1965 and 1970 and carried a stable volume of traffic after 1973. The cargo-handling capacity of well-developed ports, spaced at intervals of 200 kilometers or less along the coastline, rose from 15 million tons in 1967 to 31 million tons in 1975. With the growth of foreign trade, the gross tonnage of Korea's oceangoing fleet rose from 163,000 in 1965 to 2.2 million in 1975. The Korean educational system also contributed to industrialization during the 1961-75 period. Indeed the breadth and intensity of im- provements in educational attainment were major elements in Korea's development strategy. Investment in education had a significant impact on the quality of the labor force. The proportion of employed persons having some secondary or higher education doubled from 16 percent to 32 percent between 1960 and 1970. The ratio of those having no formal education declined from 45 percent to 24 percent in the same period and was even lower in 1976. Thus a well-trained, highly produc- tive labor force facilitated the expansion of manufacturing activities that intensively used labor, particularly skilled labor. This expansion resulted in rapid growth of labor productivity and incomes which, in turn, en- abled the wider distribution of the benefits of industrial and agricultural expansion. Urbanization The tremendous growth of industry relative to agriculture and the wide disparities in the quality of economic infrastructure and cultural amenities were the principal determinants of the pace and pattern of urban growth. In 1960 about 37 percent of the population lived in urban areas and 28 percent lived in cities; by 1975 the proportions bad respec- 26 PART ONE: THE MAIN REPORT Table 1.4. Urban Population, 1960-75 (millions of persons) Urban area 1960 1966 1970 1975 Seoul 2.4 3.8 5.5 6.9 Busan 1.2 1.4 1.9 2.5 Daegu 0.7 0.8 1.1 1.3 Other cities 2.7 3.7 4.4 6.1 Towns 2.3 2.6 2.9 3.7 Total 9.3 12.5 15.8 20.5 Sources: EPB, Korea Statistical Yearbook, 1975; EPa, Preliminary Release of the Census of Popula- tion, 1975 (Seoul, 1976). tively increased to 59 percent and 48 percent.10 The absolute number of rural dwellers began to decline in 1966. To a remarkable degree, urbanization was associated with an increase in the concentration of population. In 1960 the population of the two largest cities, Seoul and Busan, was 3.6 million, or 14 percent of the national population (table 1.4). In 1975 these cities accounted for 27 percent of the national popu- lation. Seoul had 7 million inhabitants, or roughly 20 percent of the national population. If surrounding urban settlements are included, this proportion was about 30 percent. The population concentration in Seoul is seen by government to in- crease Korea's vulnerability to external aggression. Considerations of national security have therefore been among the main motivations of efforts to decentralize the population. Since about 1970, government has applied a variety of instruments to resolve the particular problem of Seoul's size in the broader context of population distribution. The most effective direct instrument appears to have been the rigid enforce- ment of zoning regulations that restrict the growth of manufacturing activities in major cities and establish greenbelts around cities. Although industrial estates near metropolitan areas flourished, others far removed from metropolitan areas were generally unsuccessful in attracting in- dustry. The burgeoning industrial estates located in the Seoul-Incheon and Busan-Masan areas already had the highest urban and industrial concentration. The pattern of migration changed in the early 1970s. Metropolitan 10. Urban areas are defined as cities with a population of more than 50,000 and towns with a population of more than 20,000. THE PATTERN OF ECONOMIC GROWTH 27 areas began to grow less rapidly than before. The most rapid growth was found in cities vith populations of 100,000 to 200,000 persons and lying outside the greenbelt near metropolitan areas. Cities on the Seoul-Busan axis, such as Daegu and Daejeon, also exhibited rapid growth. Even though these trends were encouraging, the population of Seoul continued to grow at about 4 percent a year. This posed a serious question: How could the tendencies toward concentration generated by rapid economic growth be reconciled with considerations of inter- regional equity and national security? Investment and Financing Korea's expanding export volume generated rising demands for invest- ment, and the growth of investment was rapid. Starting at about 11 percent of GNP in 1960, fixed investment rose in current prices to more than 20 percent of GNP in 1966 and subsequently was generally between 23 and 26 percent of GNP (table 1.5). The expansion of investment was Table 1.5. Composition of Investment, 1962-76 (percent) Annual average Sector 1962-66 1967-71 1972-76 Agriculture, forestry, and fishing 10.2 7.1 9.0 Mining and manufacturing 22.9 19.7 22.1 Social overhead 28.9 36.7 32.5 Other fixed investment 25.6 29.4 30.2 Stocks 12.4 7.1 6.2 Total 100.0 100.0 100.0 Total investment (billions of won) 209.6 609.0 1,008.6 (percentage of GNP) 14.5 25.9 26.2 Note: All figures are in 1970 prices. Sources: BoK, Monthly Economic Statistics, September 1975 and Sep tember 1976; Boic, Economic Statistics Yearbook, 1976, pp. 267-77, 280-81. balanced, and the shares of various sectors in total investment remained fairly stable during the entire period. Investment in social overhead capital, such as electricity, transport, and construction, was relatively 28 PART ONE: THE MAIN REPORT high during 1967-71 and subsequently declined. Nevertheless invest- ment in infrastructure was consistently much higher than that in the manufacturing sector. With the exception of housing, the supply of infrastructure was adequate. Production did not suffer from shortages of transport, power, and communication facilities. Reflecting the emphasis on achieving self-sufficiency in grain production, investment in agricul- ture more than doubled between 1967-71 and 1972-76, from W43 billion to W94 billion a year. For corresponding periods the investment in mining and manufacturing grew from W120 billion to W217 billion a year. Growth in the second period was held back by the recession in export demand during 1974 and 1975. In an effort to sustain domestic economic activity and employment, government facilitated the buildup of manufacturing inventories in 1974. The increase in stocks, including grain stocks, was 4 percent of GNP in 1974 and an additional 1.1 percent of GNP in 1975. Financing this growing volume of investment posed a major chal- lenge."1 The basic problem was that Korea entered the period of rapid growth with a low rate of national savings compared with other countries at similar income or development levels. During Korea's first five-year plan for 1962-66, national savings financed only 42 percent of invest- ment (table 1.6). An equal amount was financed by net transfers from abroad. The sharp decline in net transfers from abroad, from 8.8 percent of GNP in 1962 to 1.2 percent of GNP in 1975, reflected a U.S. policy shift and Korea's growing self-reliance. National savings rose substan- tially, but not enough to compensate for the decline in transfer receipts or to finance the rise in investment. Consequentlv increases in external borrowing were large. Improved fiscal performance was instrumental in mobilizing domestic resources for investment. Current revenue increased from an average of 10.5 percent of GNP during 1962-66 to 17 percent in 1975. Central government tax revenue, derived largely from higher indirect taxes and improved tax administration, accounted for almost all of the increase. Direct tax schedules were modified in 1972 and 1974 to provide finan- cial relief to businesses and compensate for the impact of inflation on income tax liabilities of individuals, particularly the lower income groups. These measures weakened the tax effort during 1972-74, but 11. For a detailed discussion see Parvez Hasan, Korea: Problems and Issues in a Rapidly Growing Economy (Baltimore and London: Johns Hopkins University Press, 1976), pp. 37-56. THE PATTERN OF ECONOMIC GROWTH 29 Table 1.6. Financing of Investment, 1962-75 (percentage of GNP) Item 1962-66 1967-71 1972-75 Investment 16.6 26.3 26.4 Savings 15.6 25.2 27.2 National 6.9 14.8 18.6 Private 6.3 8.9 14.8 Governnment 0.6 5.9 3.8 Net current transfers from abroad 6.9 3.2 1.4 Foreign' 1.8 7.2 7.2 Statistical discrepancy 1.0 1.1 -0.8 Note: All figures are in current prices. By averaging current price ra- tios, the figures fail to take into account the changes in relative prices of investment and consumption. Nevertheless the magnitudes are broadly correct and avoid some arbitrary assumptions involved in adjusting sec- toral savings flows for changes in prices and terms of trade. Sources: BOK, Monthly Economic Statistics, September 1975 and Sep- tember 1976; BOIa, Economic Statistics Yearbook, 1976, pp. 300-01. a. Defined as equal to the deficit in the current account of the bal- ance of payments. were compensated for by the growth of the tax base, partly as a result of inflation and increased revenue from government monopolies. The introduction of a global system of income tax in 1975 also appeared to have yielded additional revenue. One other major change in tax rates was the defense surcharge introduced in July 1975. It was expected to yield revenue equivalent to 2 percent of GNP in 1976. As a result, the ratio of tax to GNP increased to nearly 18.3 percent in 1976. Budgetary expenditure, growing at about the same rate as GNP during the 1965-75 period, was held in check. General expenditure other than defense averaged about 7 percent of GNP, and a large part of it was directed to education. Defense expenditure averaged slightly more than 4 percent of GNP Up to 1974, but began to increase more rapidly than GNP thereafter. The budget allocation for defense in 1976 was 6.1 per- cent of GNP, and the increases after 1974 were equal to the revenue from the defense surcharge. The decline of the public savings ratio from a peak of 7 percent of CNP in 1970 to 4.6 percent of GNP in 1973 was largely the result of the reduction in the tax burden. The sudden emergence of large subsidies for grain and fertilizer sales contributed to the continuing low ratios of 30 PART ONE: THE MAIN REPORT Table 1.7. Extrabudgetary Operating Deficits, 1973-76 1973 1974 1975 1976 Per- Per- Per- Per- Bil- cent- Bil- cent- Bil- cent- Bil- cent- Extra- lions age lions age lions age lions age budgetary of of of of of of of of account Gon GNP won GNP won GNP won GNP Grain Man- agement Fund 25.4 0.5 125.0 1.8 93.6 1.0 50.3 0.4 Fertilizer Fund 15.0 0.3 32.7 0.5 70.0 0.8 -2.0 0.0 Total 40.4 0.8 157.7 2.3 163.6 1.8 48.3 0.4 Source: Data supplied by EPB. 3 percent of GNP in 1974 and 5.3 percent in 1975.12 Three principal factors were responsible for the increase in deficits of the Grain Manage- ment Fund and the Fertilizer Fund: the increase in the cost of imports in 1973 and 1974; the desire to cushion the impact of inflation on urban consumers; and the need to maintain farmers' profit ratios so as to pro- gress toward self-sufficiency in rice and barley. Deficits in these two accounts reduced public savings by about 2 percent of GNP in 1974 and 1975 (table 1.7). Govemment subsequently moved to contain these deficits, and the actions taken were sufficient to eliminate operating deficits in the Fertilizer Fund in 1976. The decline in the international price of wheat enabled the elimination of the subsidy on wheat.'3 Domestic procurement and selling prices of rice and barley were also modified to reduce operating deficits in these accounts. As a result of these measures, public savings in 1976 improved to 6.7 percent of GNP. There is no evidence that increases in government revenue and public savings seriously affected the rate of private saving. From rather low levels in the early 1960s, the ratio of private savings to GNP increased significantly (see table 1.6 for average ratios and table D. II in appendix 12. Govemment purchases and sales of grain and fertilizer are channeled through the Grain Management Fund and the Fertilizer Fund, neither of which is included in the budget. To meet most of their financing needs, the two funds borrow from the Bank of Korea. Their operating losses are included in the determination of public savings in table 1.6. 13. In current prices there actually was a surplus, which was paid into a wheat price stabilization fund. THE PATTERN OF ECONOMIC GROWTH 31 Table 1.8. Debt Outstanding and Debt Service, 1966-76 (millions of dollars) Item 1966 1971 1975 1976 Medium- and long-term debt outstanding and disbursed- 326 2,401 5,958 7,243 Debt service 19 325 678 940 Debt service ratio' (percent) 4.7 22.0 11.9 10.3 Source: Data supplied by the Debtor Reporting System of the World Bank. a. Includes debt not officially guaranteed. b. The ratio of debt service payments to exports of goods and nonfactor services. D for annual ratios). This ratio displayed much volatility from year to year, but the underlying determinants were not obvious. Rising incomes probably were a major influence. Some analyses suggest that the interest rate reform in 1965 had a discernible effect on real savings ratios and on savings in the form of financial assets. More than one-half of private savings comes from the corporate sector, and this suggests that the rate of profit retention probably influences corporate savings significantly. Analysis of corporate financial statements shows, however, that the rate of profit retention is low and that most corporate investment is financed by borrowing, not from internally generated funds. Although private savings improved substantially during 1965-75, reliance on external borrowing continued to be high. The primary rea- sons for this pattern were large increases in investment and the substi- tution of borrowing for grant assistance formerly provided by the United States (see table 1.6). Direct foreign private investment was not a significant source of investment resources.14 Annual net inflows averaged $54 million from 1966 to 1975. During most of this period, external borrowing averaged about 7 percent of CNP and financed more than one-quarter of total investment. As a direct result, Korea's outstanding external obligations increased substantially (table 1.8). In 1971 out- standing and disbursed debt reached 27 percent of GNP; the ratio of debt service to exports of goods and nonfactor services rose to the ex- tremely high level of 22 percent. Although outstanding debt continued to rise faster than GNP because of the balance-of-payments crisis in 1974 14. Except in the initial stages of exporting electronics goods, multinational firms also appear to have played only a small role in promoting exports. 32 PART ONE: THIE MAIN REPORT and 1975, the debt service ratio sharply declined because of the rapid increase in export earnings. Servicing external debt thus did not con- stitute an excessive burden on Korea's balance of payments. Employment and Equity Even with a rapidly growing employable population, the rate of un- employment in Korea declined. The rate of population growth declined from 3 percent a year during 1955-60 to about 1.7 percent during 1970- 75. Despite this decline the employable population continued to grow at higher rates.15 Until 1969 the annual growth rate of the employable population was about 2.7 percent. It then accelerated to about 4 percent in 1971 and 1972 and decelerated to about 3.3 percent in 1975. The main reason for this pattern was the gradual maturing of young Koreans born during the baby boom after the Korean War. Although the expan- sion of school enrollments accounted for an increasing proportion of the employable population, the labor force grew faster than the employable population (table 1.9). Because of an increase in the proportion of the employable population seeking jobs, the labor force grew at an average rate of 3.4 percent a year from 1963 to 1975.16 Nevertheless the rate of unemployment declined from 8.2 percent of the labor force to 4.1 percent because employment was growing at the extraordinarily high rate of 3.7 percent a year. The sectoral composition of employment changed markedly in re- sponse to shifts in the structure of production. The number employed in each of the three major sectors tended to expand, except for declines in agriculture in some years. But the shift in composition toward manu- facturing, away from agriculture and in recent years away from services, was significant (table 1.10). Despite its small base in 1963, the mining and manufacturing sector absorbed 38 percent of incremental employ- ment during 1963-75 and 47 percent during 1971-75. The contribution of the manufacturing sector to employment was crucial and can be directly traced to the strategy of emphasizing labor-intensive export ex- 15. In accord with Korean practice, the employable population is defined as including all persons aged 14 years and older, but excluding military personnel, prisoners, and other institutionalized persons. 16. Labor force participation rates varied from year to year in response to the demand for labor. THE PATTERN OF ECONOMIC GROWTH 33 Table 1.9. Manpower Use, 1963-75 1963 1975a Annual Millions Percent- Millions Percent- growth of age of of age of rate Category persons total persons total (percent) Employable populationb 15,085 100 21,833 100 3.2 Attending school 1,097 7 2,822 13 8.2 Labor force 8,343 55 12,340 57 3.4 Employed 7,662 51 11,830 54 3.7 Unemployed 681 5 510 2 - Ratio of unemployed to labor force (percent) - 8 - 4 - - Not applicable. Source: Epa, Annual Report of the Economically Active Population, 1974 (Seoul, 1975), pp. 39, 45. a. Estimates. b. Defined as the population aged 14 years and over, excluding military, pris- oners, and other institutionalized persons. Table 1.10. Composition of Employment, by Sector, 1963-75 (percent) Sector 1963 1966 1971 1975 Agriculture, forestry, and fishing 63.1 57.9 48.4 45.9 Mining and manufacturing 8.7 10.8 14.2 19.1 Social overhead and other 28.2 31.3 37.4 35.0 Source: BOK, Economic Statistics Yearbook, 1976, pp. 254-55. pansion. In addition to its direct absorption of employment, the manu- facturing sector increased the demand for services and stimulated pro- ductivity growth in both services and agriculture. Although the rural population declined during 1965-75, employment in the agricultural sector increased.17 Almost all the increase of about 600,000 persons took place after 1971. Before 1971 the number em- ployed in agriculture fluctuated between 4.8 and 4.9 million. This phenomenon is not easily explained, but a number of hypotheses are 17. This increase was primarily the result of an increase in labor force partici- pation by women in rural areas. 34 PART ONE: THE MAIN REPORT plausible. The rising employment in agriculture could have been the result of the growing importance of crops that require a larger input of labor, such as vegetables; of technological changes that are not labor saving, such as high-yielding seeds, improved water management, and even power tillers; or of the tendency of agriculture to absorb the residue when manufacturing and services did not absorb the full increment in the labor force, as happened in 1970-71 and 1974-75 when demand for labor in manufacturing was slack. Some empirical evidence can be found to support each of these possibilities. The basic facts nevertheless are that labor productivity in the agricultural sector was still low in comparison with other sectors and was rising less fast than in other sectors of the economy. In 1975 value added per worker in agriculture (including forestry and fishing) was $876, compared with $2,407 in mining and manufacturing and $1,586 for the economy as a whole. The growth of labor productivity in agriculture was 4.3 percent a year during 1963-71, but slowed down during 1971-75 because of the increase in the number emploved in this sector. It is generally accepted that some underemployment still exists in agriculture. If some means were found to cope with the sharp seasonal peaks in labor demand in the sector, the number of those employed in agriculture could decline without any significant loss of agricultural output. Rapid growth in employment was accompanied by impressive rates of productivity growth in all sectors (table 1.11). In the agricultural Table 1.11. Employment and Productivity, by Sector, 1963-75 (percent) Annual growth rate Value added Sector Employment per wvorkera Agriculture, forestry, and fishing 1.0 3.5 Mining and manufacturing 10.7 6.5 Social overhead and other 5.6 4.3 All sectors 3.7 6.1 Note. All figures are in 1970 prices. Sources: EPB, Economically Active Population, 1974, p. 32; EPB, Monthly Statistics of Korea, January 1976, p. 137. a. Value added per worker is only a crude measure of changes in labor productivity. It should be noted that sectoral differences also reflect different rates of capital deepening. The estimates are pre- sented here simply to make the point that productivity has grown rapidly in all sectors. THE PATTERN OF ECONOMIC GROWTH 35 sector the productivity gains were derived from the increased use of modern inputs, water management, and the larger share of fishing and high-value crops in total output."8 In manufacturing and services, changes in the output mix probably were the most important factor in productivity improvements. A study of capital-labor ratios up to 1973 suggests that there was no secular drift toward capital-deepening in the manufacturing sector. Both capital stock and employment increased by a little more than 100 percent between 1966 and 1973 (see chapter 8). This pattern suggests that the gain in labor productivity was the result of technological improvements or of the more efficient allocation of labor to sectors having higher productivity, not of the greater capital intensity of production. Although the data are inadequate to enable analysis of the sources of productivity growth in the service sectors of the economy, capital- deepening and changes in output mix probably were important. During 1965-75 these sectors absorbed about one-third of total employment, but more than three-fifths of total investment. The contribution to GNP of construction, electricity, and transport increased from 9 percent in 1966 to 13.7 percent in 1975. These sectors absorbed more than 30 percent of total investment and would be expected to have relatively high labor productivity. On the other band, investment in public administration, housing, and such other services as education and health declined as a proportion of GNP-from 19.2 percent in 1965 to 12.9 percent in 1975. For the economy as a whole, the average real wage increased at the rate of nearly 7 percent a year during 1963-75. The share of wages in national income increased slightly in the mid-1960s and was about 38 percent after 1967. In aggregate, worker compensation rose faster than worker output during 1966-70, but then slowed considerably during 1970-74.19 Real wages in agriculture rose by about 5 percent a year during 1965-75, except for a sudden, absolute decline in 1974. There are no data on how wage rates changed in the services, and data cited above on wages in manufacturing and agriculture are not always com- 18. Because these productivity trends were calculated in constant 1970 prices, they do not include the impact of changes in the terms of trade between agri- culture and other sectors. See table 1.12. 19. It is not entirely clear why real wages slowed down in the second period. In part this slowdown probably represented lags in the response of nominal wages to the acceleration of the rate of inflation. This could be the mechanism by which wage eamers were made to share the loss in national income caused by the adverse shift in the intemational terms of trade. 36 PART ONE: THE MAIN REPORT parable. Because of the close relation of wage trends in the manufactur- ing sector to those in the rest of the economy, it would appear that wages in services continued to grow rapidly, even in sectors where productivity growth might not have been so fast. Size distribxtion of income Data on the distribution of income in Korea, although better than in many countries, are still fragmentarv and generally deficient. By com- bining information from various sources, some studies have attempted to formulate acceptable generalizations.20 First, the distribution of in- come in Korea is among the best in the developing world. A survev in 1970 showed that the bottom 40 percent of Koreans received 18 percent of total income and the top 20 percent received 45 percent of total in- come. Information on forty-five developing countries presented in a recent study shows that these shares are more egalitarian than in most developing countries.21 Second, despite the increase in inequality ex- pected on the basis of cross-sectional evidence, it would appear that there was no significant deterioration in the distribution of income until at least 1970. Korea's performance in distributing the benefits of growth was partly a result of the spread of education and the elimination of great dispari- ties in asset ownership in the period before rapid industrial growth.22 The partition of Korea in 1945 and the Korean War destroved what little industrial wealth existed in the South, and sweeping land reform radically altered the distribution of agricultural wealth. Before the land 20. Irma Adelman and Sherman Robinson, Income Distribution Policy in Developing Countries: A Case Study of Korea (Stanford: Stanford University Press; Oxford: Oxford University Press, 1978); Bertrand Renaud, "Economic Growth and Income Inequality in Korea," World Bank Staff Working Paper, no. 240 (Washington, D.C.: World Bank, February 1976); H. C. Choo, "Some Sources of Relative Equity in Korean Income Distribution: A Historical Perspec- tive," in Income Distribution, Employment and Economic Development in South- east and East Asia: papers and proceedings of the conference sponsored jointly by the Japan Economic Research Center and the Council for Asian Manpower Studies, December 16-20, 1974, 2 vols. (Tokyo: Japan Economic Research Center, 1975). 21. Montek S. Ahluwalia, "Inequality, Poverty and Development," Journal of Development Economics, vol. 3 (1976). 22. This discussion draws heavily from Adelman and Robinson, Income Dis- tribution and Policy and Choo, "Sources of Equity." THE PATTERN OF ECONOMIC GROWTH 37 reform, it is estimated that about 94 percent of farmers were tenants either in part or full. Redistribution of land previously held by Japanese in 1947 and further land reform in 1949 virtually eliminated tenancy, put a ceiling of three bectares on paddy holdings, and established a structure of small owner-operated farms. The structure of landholdings in agriculture in 1974 shows that almost 94 percent of holdings were smaller than three hectares and accounted for 93 percent of all agricul- tural land.23 Although tenancy arrangements retumed on a small scale, the broad characteristics of land distribution did not change. The even distribution of land contributed to the equitable gains of farmers from the growth of farm incomes. In addition, the early spread of education was instrumental in the extensive modernization of agri- culture and enabled a wide segment of the population to participate in the rapidly expanding modern manufacturing sector. Primary education was nearly universal by the early 1960s, when manufactured exports began to take off. Laborers already had some basic education, and this must have made it easier to train them for work in modern factories and to achieve the sustained growth of labor productivity. Because national data on the size distribution of income are not avail- able, the impact of recent developments cannot be assessed with full confidence. Nevertheless examination of the development strategy and policies adopted by government tends to support the view that equity objectives were not ignored. The distribution of the benefits of growth was more the result of the growth strategy than of fiscal intervention in the form of taxes and subsidies. Choosing to rely on exports because of the poor endowment of natural resources, Korea was driven by the forces of comparative advantage to expand its labor-intensive manufac- turing activities. Pursuing this goal with dedication generated rapid employment growth which, in turn, directly benefited those at the lower end of the income scale. The marked improvement in employment opportunities and the forces that prevented earnings differences from widening undoubtedly were the most important influences favoring equity in Korea. Even though the policy focus generally was on maintaining economic growth, government actively promoted policies having a favorable in- fluence on equity. The attempt to raise rural incomes by improving the rural-urban terms of trade through changes in the price of grain is one case in point. The attempt to curb the accumulation of urban real 23. See table 7.1 in chapter 7. 38 PART ONE: rHE MAIN REPORT estate by imposing progressive property taxes and taxing capital gains is another. These fiscal measures acted as a brake on the growth of real estate values, but the magnitude of their impact on the distribution of wealth is difficult to assess in the absence of data. It can be conjectured, however, that ownership of urban wealth became more skewed than ownership of farmland and that aggregate urban wealth was growing much faster than rural wealth. Thus, although continued enforcement of land reform ceilings prevented large rural landholdings, the distribu- tion of national wealth probably was less equal in 1976 than in the early 1960s. But its effect on the distribution of income could not have been overwhelming. In relation to national income, income from property (including interest income) rose from 10.8 percent in 1965 to more than 13 percent in 1970, but then declined to 11.4 percent in 1974. Income from unincorporated enterprises outside agriculture increased from 16 percent in 1970 to 23 percent in 1974. These increases were not large. It is also relevant that a significant number of deposit-holders and unincorporated enterprises came from the middle of the income scale. As noted earlier, government did not rely on direct fiscal intervention to improve the size distribution of income.24 The payment of subsidies from the central government budget, excluding grain and fertilizer subsidies but including transfers to local governments for education, amounted only to 1.7 percent of GNP in 1975. Incentives to the private sector tended, moreover, to encourage capital intensity, and interest subsidies and accelerated depreciation allowances were used rather than wage subsidies. The redistributive impact of government operations has not been analyzed, but it can be conjectured that this impact was not significant because the government enterprise sector was not large in relation to the rest of the economy. Nor did the tax system appear to be especially progressive. Although income tax rates were very progres- sive, they were diluted by a variety of exemptions. Indirect tax rates were high for luxury commodities, but their net effect on income cannot be assessed in the absence of incidence studies. Thus only the expendi- ture on grain subsidies was specifically directed to the alleviation of poverty. True, educational expenditure, at about 3 percent of GNP, was relatively large and clearly had a beneficial effect on the distribution 24. Intervention to influence the rural-urban distribution of incomes was sub- stantial, as discussed below. Because average rural incomes are below average urban incomes, this intervention also had an impact on overall size distribution of income. THE PATTERN OF ECONOMIC GROWTH 39 of income. But it was the dedication to economic growth and exports that substantially improved the living standards of Koreans by generat- ing opportunities for employment and higher incomes. Rural-urban balance With more than one-half of the labor force in agriculture, the sub- stantially faster increase in productivity in the manufacturing sector accentuated the problem of disparities in household incomes between rural and urban sectors. Because of the social importance of reducing rural-urban income disparities, government used a variety of instruments to bolster rural incomes. The most important of these was regulation of the grain price, which affects the rural-urban terms of trade. After Teaching a peak in 1963 the farmers' terms of trade deteriorated rapidly up to 1966 and reached a trough in 1968 (table 1.12). Between 1968 Table 1.12. Farmers' Terms of Trade, 1963-76 (1975 = 100) Item 1963 1968 1973 1976 Prices received 14.9 27.6 62.2 124.0 Grain prices received 15.6 23.9 55.9 120.2 Prices paid 14.8 33.9 61.6 124.9 Terms of trade 100.7 81.4 101.0 99.3 Sources: EPB, Monthly Statistics of Korea, various issues. and 1973 the prices paid by farmers for their purchases increased by nearly 9 percent a year, but wholesale prices increased even faster, principally because of the rate of increase in grain prices. Applying an index of 100 for 1970, the farmers' terms of trade reached a peak of 114.7 in 1973. There was a small decline after 1973 because of increases in their cost of living during 1973-76 of 27 percent a year, compared with 21 percent a year for the urban consumer price index. To some extent the improvement of the farmers' terms of trade was achieved by subsidizing inputs. The selling price of fertilizer was held below cost by an increasing margin after 1971, and in 1975 the unit cost exceeded unit revenue by as much as 88 percent.25 Government support 25. Fertilizer prices were raised in December 1975 to eliminate the subsidy. 40 PART ONE: THE MAI REPORT Table 1.13. Grain Price Support by Government, 1968-76 (percent) Ratio of government release prices to purchase prices, Grain 1968 1971 1973 1975 1976 Rice 119 117 99 85 86 Barley 90 88 86 75 71 Sources: Data supplied by EPB and the Ministry of Agri- culture and Fisheries (MAF). a. Excludes handling costs and wastage. for higher grain prices, intended to provide higher rural incomes, was more significant (table 1.13). After 1968 government's release price of barley was consistently below its purchase price, and the difference continued to widen. For rice the purchase prices were below release prices until 1971, when the situation reversed. As the difference widened, the defi- cits for the government's grain operations increased. Without govern- ment support of grain prices, however, the deterioration of the farmers' terms of trade could have been much worse than the 2 percent deteriora- tion between 1973 and 1975. Another instrument for reducing rural-urban disparities, heavily sup- ported by government, has been the New Community movement (Saemaul Undong).26 The broad goal of the movement has been to galvanize villages into improving their living environment and produc- tivity. Elected village committees select projects that are implemented principally by volunteer labor, but with some assistance from central government in cash and building materials. The nature of projects depends on the level of organizational ability in the village. Starting with such projects as roof improvements and drains, villages progress to infrastructure projects (roads, bridges, and irrigation) and schemes to supplement incomes (livestock raising, silk production, specialty crops, and cottage industries). Although the physical infrastructure con- structed under the movement has been impressive, it should be stressed that the major objectives of the movement are to develop self-reliance, 26. The organization, objectives, and achievements of the New Community movement are described in Hasan, Korea: Problems and Issues, pp. 159-64. THE PATTERN OF ECONOMIC GROWTH 41 Table 1.14. Rural and Urban Income Levels, 1963-76 (percent) Ratio of rural incomes to urban irncomes' Type of income 1963 1968 1973 1975 1976 Household 116 63 87 102 100 Per capita 107 57 80 93 91 Per worker 43 27 41 48 47 Sources: EPB, Annual Report on the Family Income and Expenditure Survey (Seoul, 1974); MAF, Report on the Re- sults of Farm Household Economy Survey (Seoul, 1975); EPB, Monthly Statistics of Korea, January 1976. a. These ratios are of the average monthly incomes of farm households to those of urban wage and salary earners. leadership, and a sense of community and thus to release latent develop- ment energies throughout Korea.27 Official statistics show a striking improvement in the distribution of incomes between rural and urban households in recent years (table 1.14). Income disparities were greatest in 1967, when the manufactur- ing sector was doing well and agricultural output declined because of poor weather conditions. After 1967 the relative improvement in rural incomes was sustained, and the gap was virtually closed in 1975. Com- pared with average urban households, rural households were slightly larger (5.6 against 5.2) and had more workers (2.9 against 1.3). Thus official statistics show a wide disparity in per worker incomes and a small disparity in per capita incomes between rural and urban sectors. But the statistics underlying table 1.14 have some serious limitations. First, the urban survey was confined to wage and salary earners, and it excluded those who earned more than a specified ceiling. The rural survey apparently defined income as including changes in the book value of inventories and thus overstated rural incomes in periods of rising inventories and grain prices. Second, because the comparison was based on nominal incomes, no account was taken of the difference in rural and urban rates of inflation. Consequently the comparison over- states the increase in rural incomes in relation to urban incomes. Such a bias is confirmed by studies using national accounts data and compar- 27. See table D.31 in appendix D. 42 PART ONE: THE MAIN REPORT ing consumption, not income.28 These studies lead to different conclu- sions: per capita consumption in 1975 was about twice as high in urban areas as in rural areas; if differences in inflation rates are allowed for, the relative position of urban and rural households did not change much in recent years. Nevertheless it is remarkable that average per capita consumption in rural areas-a measure not subject to some of the errors that beset the estimation of rural incomes-grew at about 4 percent in real terms between 1965 and 1975. Preventing income disparities from widening was a major task because labor productivity in manufacturing was growing at nearly 7 percent a year and because agriculture suffers from severe land and climatic constraints. The basic problem, however, was the slow growth of nonagricultural income in the rural sector. Having risen to nearly 24 percent of farm household incomes for a brief period during 1968 to 1970, it remained at about 20 percent of farm household income for the next five years. As noted, government policies attempted to change this by establishing industrial estates outside existing indus- trial concentrations and encouraging the new community factories in rural areas. The heavy reliance on the manipulation of grain prices to remedy rural-urban disparities, in addition to its significant effects on public savings, had important effects on equity. In the urban sector, low cereal prices benefited the lower income groups. The share of cereals in total expenditure ranged from 30 percent for urban families with lowest total expenditure to 11 percent for those with highest total expenditure. Furthermore lower income groups consumed relatively more barley than rice, and barley subsidies were sustained and proportionately higher. The impact on the distribution of rural income was less favor- able. The benefits from supported grain prices were distributed among farmers according to the proportion of output marketed. Families with smaller landholdings tended to consume a greater proportion of their cereal production and to derive a smaller proportion of their incomes from agricultural sources. Consequently the proportionate impact on income of an increase in rice support prices is estimated to have been five times higher for farmers holding more than two hectares than for farmers holding less than one-half hectare. Although there are grounds for doubting whether rural-urban income disparities were reduced and whether the internal terms of trade were 28. See Renaud, 'Economic Growth and Income Inequality." THf PATTERN OF ECONOMIC GROWTH 43 the best instrument for achieving its objective, it should be noted that Korean villagers enjoyed an unusually high level of basic services in 1976. There were 2.8 million rural households in 45,000 villages, of which about 18,600 were legally constituted and had 100 or more house- holds. All villages had easy access to primary schools, and most villages to middle schools. Family planning material was widely disseminated, and rural fertility rates were declining about as fast as urban rates. About 91 percent of all households had access to electricity. More than one-half of the legal villages had community telephones. Almost all villages were fairly well connected by roads, although there was much room for improvement in the quality of rural access roads. Access to piped water was still relatively low. On the whole, the breadth of dis- tribution of most services probably was better than that in other coun- tries at comparable levels of development. Chapter 2 D. C. Rao Korea's Fourth Plan, 1977-81 IN ORDER To UNDERSTAND BEI7ER the objectives and strategy of Korea's fourth five-year economic development plan for the 1977-81 period, it is useful to examine Korea's third plan for 1972-76 in the context of recent economic trends. During the first half of the 1970s the external economic environment was subjected to unexpected, sharp, and sudden changes which affected the level of economic activity in Korea and its external trade and payments (table 2.1).1 The boom in world trade led to increases in Korean export volume of 46 percent in 1972 and 54 percent in 1973. But in 1974, with the quadrupling of oil prices and a sharp rise in the prices of foodgrains and capital goods, the terms of trade deteriorated by 19 percent. Concurrently the deep recession in the industrialized countries led to a slowdown in the growth of export volume to 9 percent. The purchasing power of exports actually declined by nearly 11 percent in 1974. Export volume in 1975, assisted by a de- valuation of the won in December 1974 and a vigorous drive to expand and diversifv markets, grew by an astonishing 20 percent. Nevertheless part of this gain was offset by a further deterioration of 10 percent in the terms of trade. In 1976, with the recovery in the industrialized countries, the purchasing power of exports increased by about 60 per- cent, reflecting a gain of about 40 percent in export volume and an improvement of about 14 percent in the terms of trade. Large swings in the terms of trade and fluctuations in the rate of export growth caused dramatic shifts in the current account of the balance of payments. The current account deficit, which averaged $340 million in 1972-73 and was equal to about 3 percent of GNP, jumped 1. For a detailed discussion of these changes see Parvez Hasan, Korea: Problems and Issues in a Rapidly Growing Economy (Baltimore and London: Johns Hopkins University Press, 1976), pp. 7, 67-78. 45 46 PART ONE: THE MAIN REPORT Table 2.1. Balance of Payments, Current Account, 1973-76 (millions of dollars) Item 1973 1974 1975 1976 Exports 3,271 4,515 5,003 7,815 Imports 3,837 6,452 6,674 8,400 Trade balance -566 -1,937 -1,671 -585 Services (net) 67 -308 -442 -77 Transfers (net) 190 222 226 349 Current account balance -309 -2,023 -1,887 -314 Memo items Terms of trade (1970 = 100) 93.6 76.2 68.6 78.2 Annual GNP growth rate- (percent) 16.1 8.6 7.9 15.5 Ratio of investment to GNPl (percent) 26.8 29.5 27.3 27.1 Ratio of national savings to GNP' (percent) 22.1 19.2 18.1 21.4 Sources. Bank of Korea (noR), Economic Statistics Yearbook, 1976 and earlier years; BOB, Monthly Economic Statistics, various issues. a. In 1975 prices. to an average of nearly $2 billion in 1974-75, or 11 percent of cNP. The deficit in 1976 was about $315 million, or slightly more than 1 percent of GNP. The large deficits during 1974-75 were virtually un- avoidable. Any attempt to adjust to a large shift in the terms of trade without incurring sizable balance-of-payments deficits would un- doubtedly have led to large-scale unemployment and a loss of the economy's momentum. Such an outcome would have had severe con- sequences for Korea's long-term development. The essence of govem- ment policy during 1974 and early 1975 was to smooth the process of adjustment and make steady progress toward a long-term improve- ment in the balance of payments. In support of Korea's long-term goals, it was necessary to maintain the momentum of the economy and remain poised to take rapid advantage of the anticipated recovery in export demand. This policy was predicated on the belief that the world reces- sion would be relatively shallow and short-lived, a belief shared at the time by many economists and policymakers throughout the world. The balance-of-payments deficits recorded in 1974 and 1975 were larger than expected. A measure of this is the fact that more than 40 percent of the deficit for these two years was financed through short- term borrowings and a reduction in net external reserves. The unex- pectedly sizable improvement in the balance of payments in 1976 never- KOREA'S FOUTRTH PLAN 47 theless alleviated immediate concem about the payments position. But because more than one-half of this improvement was apparently attrib- utable to the significant gain in the terms of trade, it was difficult to predict how lasting it would prove. In many ways 1976 was an excep- tional year. Consumption and investment lagged behind the unex- pectedly rapid increase in export earnings. Agricultural outputy also was very high. But even allowing for these factors, the underlying current account deficit in 1976 still compared favorably with the deficit of about 7 percent of GNP in 1970-71. Considering the unfavorable external environment and the net fall in the terms of trade of about 22 percent between 1971 and 1976, this achievement was remarkable. Although the increases in the outstanding external debt were large during the third plan period, the real burden of external debt in fact declined (see table 1.8 in chapter 1). The ratio of debt service payments to exports of goods and nonfactor services declined from an average of about 21 percent in the early 1970s to an estimated 11 percent in 1976. Total outstanding debt at the end of 1976 was estimated at about $7 billion for medium- and long-term maturities and an additional $2 billion for maturities less than one year. Debt service payments in 1976 were nearly $1 billion, an amount well within Korea's repayment capacity. Confidence in Korea's ability to meet its external debt service obliga- tions is based on the continuation of rapid export growth. Korea's export performance in recent years demonstrated that its competitive position continues to be strong. Between 1970 and 1974 Korean manufactured exports in current prices increased more than fivefold, from $0.7 billion to $3.8 billion. This growth pushed up Korea's share in world trade in manufactures from 0.4 percent to 0.9 percent. During this period ex- ports of manufactured goods from all developing countries were rising relatively rapidly-at 19.5 percent a year during 1972-74, compared with 14 percent a year during 1966-71. Although there was little growth in the world trade in manufactures during 1975 and 1976, Korea's export performance was outstanding. The 20 percent growth in Korea's export volume in 1975 was in part achieved by allowing a decline in unit values. The losses associated with this decline were more than recouped in 1976, when export volume rose by about 40 percent. Even in the recessionary years 1974 and 1975 Korea's GNP grew by more than 8 percent a year.2 In 1976 growth of GNP was 15.5 percent, 2. Because of losses in the terms of trade, gross national income grew at an average rate of 3.3 percent a year. Gross national income is GNP adjusted for the income effects of changes in the terms of trade. 48 PART ONE: THE MAIN REPORT primarily because of buoyant exports. Although part of this increase in exports may have been met by the depletion of inventories, it could not have been achieved without substantial additions to productive capacity. One notable feature of Korea's economic management during the crisis in 1974 and 1975 was that fixed investment continued to grow in con- stant prices at an average of 9.4 percent a year: its ratio to GNP rose from 24.2 percent in 1973 to 24.8 percent in 1975. One additional reason for the relatively rapid increase in GNP in 1976 is the extraordinarily good growth in agriculture. Value added in the agricultural sector grew by more than 7 percent a year in 1974 and 1975. Although weather con- ditions were exceptionally favorable in 1975, they were less favorable in 1976. This suggests that yields improved as a result of better tech- nology and greater inputs. With the slowing down of imported inflation and the introduction of considerable monetary restraint, the strong inflationary pressure of the 1973-75 period was largely brought under control. Wholesale prices, which increased by nearly 80 percent in 1974 and 1975, increased by only about 9 percent in 1976. Inflation during 1974 and 1975 was for the most part the result of the deliberate policy of allowing international price increases to be reflected in domestic costs and putting the burden of adjustment on the domestic consumer. Import costs, which more than doubled between 1972 and 1974, apparently increased little during 1975 and 1976. At the same time, growth in domestic credit is estimated to have been limited to 21 percent in 1976, compared with increases of 54 percent in 1974 and 32 percent in 1975. Underlying the slower rate of monetary expansion was the improvement in the fiscal position of government. Public savings, which declined from 6 percent of GNP in 1971 to 3 percent in 1974 because of sluggish tax revenue and sizable subsidies on food and fertilizer, recovered to about 5 percent of GNP in 1976. The national saving rate also improved substantially over the third plan period, even allowing for the fact that part of the improve- ment in savings during 1976 may prove to be short-lived. Despite the balance-of-payments difficulties in 1974-75 and an inter- national environment less favorable than was expected when the third plan was formulated, Korea achieved most of the objectives set for the 1972-76 period. The rate of GNP growth during the plan period was about 11 percent a year, compared with the original target of 8.6 percent. Considerable progress was achieved in expanding the manufacturing sector, particularly in metals and chemicals, and in developing capa- bilities for shipbuilding and manufacturing machinery. The progress in restraining inflation and reducing the reliance on foreign savings never- KOREA'S FOURTH PLAN 49 theless fell short of original expectations because of the unfavorable external environment. Furthermore, even with employment growing at 4 percent a year, there was little improvement in the employment situa- tion. More than 2 million jobs were created in the third plan period, but they were barely sufficient to keep pace with the growth in labor force. The rate of open unemployment in urban areas continued to be about 7 percent. Korea thus entered the fourth plan period with a strong base, but without having overcome the basic problems of resource scarcity and job creation. In addition, further progress was needed in achieving changes in the industrial structure to reflect Korea's compara- tive advantage and in developing a viable strategy for increasing exports in the long run. These are the problems policymakers addressed in the fourth plan. Objectives of the Fourth Plan The major goals of the fourth plan are to strengthen the balance of payments, to achieve structural shifts in the manufacturing sector in order to ensure the long-term viability of Korea's export-oriented devel- opment strategy, and to distribute broadly the benefits of growth by maintaining rapid employment growth and increasing the availability of essential services in rural areas. The plan projects elimination of the deficit in the current account of the balance of payments by 1979 and a surplus of $1.2 billion in 1981. Even with the projected improvement in the balance of payments, the plan estimates that direct foreign investment and gross disbursements of medium- and long-term capital will have to reach about $12 billion in current prices during 1977-81, compared with about $6 billion during 1972-76. Strengthening the balance of payments and reducing the in- flows of foreign savings will require a commensurate increase in the mobilization of domestic savings. National savings, exclusive of trans- fers, are projected to rise to 26.1 percent of GNP by 1981; the marginal saving rate assumed for the plan period is 35.5 percent. In addition to significant improvements in the savings of the govemment and corporate sectors, household savings are projected to rise from 5.2 percent of GNP in 1976 to 6.7 percent in 1981. The second major goal-structural shifts in manufacturing-should be seen in the context of shifts in Korea's comparative advantage. The plan is based on the expectation that Korea's export markets for textiles, clothing, and other light manufactures have limited growth potential 50 PART ONE: THE MAIN REPORT and are subject to encroachment by other developing countries. There- fore Korea's comparative advantage will increasingly lie in such indus- tries as machinery, electronics, and shipbuilding, which require more skilled labor and greater industrial sophistication. The output of these industries is projected to increase at 22 percent a year, which would boost their share of manufacturing output from 13 percent in 1975 to nearly 21 percent in 1981. Rising employment has been the most important instrument for achieving an equitable distribution of the benefits of growth in Korea. Despite the success in reducing the rate of growth of population to about 1.7 percent a vear, the labor force continues to grow at more than 3 percent a year because of the baby boom after the Korean War.3 One major objective of the fourth plan is to increase employment opportuni- ties at about the same rate as labor force growth and prevent an increase in unemployment. Given the proposed continuation of the labor- intensive strategy, the principal determinant of the rate of job creation is the rate of GNP growth. Agriculture and services cannot absorb large additions to the labor force if, at the same time, a reasonable growth in per capita incomes is assumed. Consequently meeting the employ- ment objective requires rapid growth of the manufacturing sector and continued emphasis on exports. The plan proposes to supplement rising employment with other measures to improve equity. The expansion of vocational training and middle and higher education will contribute to improved productivity and eamings. Health outlays will increase as a proportion of total investment, and the plan proposes special stress on preventive care and an innovative system of health delivery to serve lowincome families. The family planning program will also be strength- ened, and basic infrastructure facilities will be further improved in rural areas. To achieve these goals, the plan proposes to continue the basic strategy of rapid, export-led growth followed so successfully since the mid-1960s. Fueled by growth of 16.5 percent a year in the volume of manufactured exports, GNP is expected to rise at 9.2 percent a year. The manufacturing sector, which now contributes 32 percent of GNP, is expected to increase at 14.2 percent a year and contribute 54 percent of incremental CNP during the plan period (table 2.2). About 41 percent of the growth in manufacturing can be attributed to the increase in exports; the propor- 3. See chapter 6 for a discussion of six published projections of Korea's labor force and two additional projections generated by World Bank staff. KOREA'S FOURTH PLAN 5 1 Table 2.2 Composition and Growth of Gross National Product, 1971-81 (percent) 1976-81 annual Sector 1971 1976 1981 growth rate Agriculture, forestry, and fishing 31.0 23.6 18.5 4.0 Mining and manufacturing 25.5 32.7 40.9 14.2 Other 43.5 43.7 40.6 7.6 Note: All figures are in 1975 prices. Sources: aox, Economic Statistics Yearbook, 1976, pp. 164-67; Economic Plan- ning Board (EPB),The Fourth Five-Year Economic Development Plan, 1977-81 (Seoul, 1976). tion is even higher in such subsectors as electronics, shipbuilding, tex- tiles, and miscellaneous light manufactures. The share of manufactures in total exports is expected to rise to about 92 percent during the plan period. As in the past, the expansion of manufactured exports is ex- pected to be the primary source of improvement in the balance of payments. The manufacturing sector is also expected to be the principal source of employment growth. Although it absorbed only about 20 per- cent of total employment during 1972-76, it is to provide more than one-half of the additional jobs during 1977-81. Selective import substitution will continue in some sectors. In metals and chemicals the contribution of import substitution to growth is expected to rival that of exports; it is also expected to be significant in machinery. Nevertheless the net contribution of import substitution is expected to be small. Despite considerable efforts to increase the domes- tic supply of foodgrains and other agricultural products, the ratio of imports to total demand is projected to remain the same because of the limitations of Korea's natural resources. Demand for rice and barley, in which Korea attained self-sufficiency in 1975, is projected to be met entirely from domestic production. The development strategy of the fourth plan is basically sound, par- ticularly the emphasis on export growth and the proposed shifts in the structure of the manufacturing sector. But three basic issues impinge on the capacity of the Korean economy to realize the goals set out in the plan: the feasibility of projected growth in exports; the realism of the plan's assessment of investment requirements; and the availability of adequate resources for financing investment. These basic issues, along 52 PART ONE: THE MAIN REPORT with certain key social goals, are the subject of the remainder of this chapter. Exports Export growth will continue to play a preeminent role in Korea's development strategy in the fourth plan. From 1976 to 1981 exports of goods and services are projected to grow at 16.8 percent a year; com- modity exports at 16 percent a year. The projected level of merchandise exports in 1981 is about $20 billion. Adjusting for an assumed increase of 5 percent a year in the unit values of exports, this level would be equivalent to $14 billion in 1975 prices. Coming on top of the excep- tionally high growth of exports in 1976, these targets are a challenge. The projected growth in volume of Korea's merchandise exports is considerably slower than past rates. As noted in chapter 1, Korea has consistently registered exceptional growth rates in exports. Although growth in the early years was from a small base, Korea now is one of the largest exporters among developing countries. Because exports con- sist predominantly of manufactured goods, the economic activity and import demand in the industrialized countries are major determinants of Korea's export performance. Current projections of growvth of indus- trialized countries are uniformly higher than growth registered during 1971-75, when Korea's export volume increased by an average of 33 percent a year. For countries constituting the Organization for Economic Cooperation and Development (OECD), GNP growth is expected to im- prove from 2.7 percent a year during 1971-75 to about 4.5 percent during 1976-80. The increase is expected to be most marked in the United States, the biggest single market for Korea. Japan, the other major market for Korea, is expected to grow at 6.4 percent a year during 1976-80, compared with 5.1 percent a year during 1971-75. Unless the growth in OECD countries, particularly the United States and Japan, falls considerably short of these expectations, it would appear that the demand for Korean exports will be buoyant enough to enable achieve- ment of the plan's export targets. A comparison with the projected growth of exports by all developing countries reinforces confidence in the feasibility of the plan's export projections. During 1966-75 world trade in manufactures grew at 11 percent a year; manufactured exports of developing countries at 15.8 percent a year. Studies of LDC export prospects conclude that their man- ufactured exports will grow at more than 12 percent a year during KOREA S FOURTH PLAN 53 1976-81, and many consider this conclusion to be conservative. In the past, LDCs have increased their share of industrialized countries' imports of manufactured goods; their share, less than 7 percent in 1973, still is quite small. The small size of their share provides great scope for further increases in line with their comparative advantage. The plan's export projections imply a faster rate of growth than is generally projected for LDCs as a whole. Korea's share in manufactured exports by developing countries increased from about 2 percent in 1965 to nearly 13 percent in 1975 and 16 percent in 1976. The plan's pro- jections imply that this share will reach nearly 20 percent by 1981, and this expectation is realistic. Korea's principal competitors are Singapore, the Republic of China (Taiwan), and Hong Kong, countries that are likely to move away from Korea's basket of exports as their production costs, particularly wage costs, rise. Countries such as Turkey, Colombia, Thailand, the Philippines, and Malaysia are expected to increase their exports relatively rapidly, but their aggregate volume still is not large. With few exceptions, other LDCs are not yet sufficiently export-oriented or well organized in their export production to cut into Korea's export markets. Even allowing for stiff competition from Brazil and Mexico in the U.S. market and possibly from the People's Republic of China in the Japanese market, it is likely that Korea's competitive position will continue to be strong. One of the grounds for optimism about Korea's exports is the impres- sive market penetration by Korea in Europe and the Middle East. As already noted, 18 percent of Korea's exports, or more than $900 million in current prices, went to Europe in 1975. The Korean government and business community have been making a concerted effort to increase their contacts with Europe, and the indications are that trading rela- tions are closer than before. The same is true for Korea's trading relations with the Middle East. Korea's construction industry has performed out- standingly well in a number of Middle Eastern countries and has gained an international reputation for efficiency. The repatriation of profits and remittances by Korean workers now are a major source of Korea's foreign exchange earnings. In the fourth plan period Korea's exports to Europe and the Middle East are likely to expand quite fast. The prospects for growth in these countries are good, and Korea's share in their imports is still quite small. For example, even in the major categories of clothing and footwear, Korean exports are only 2 percent of the imports of the European Economic Community (EEC). To some extent caution about Korea's export prospects stems from the quantitative restrictions placed by industrialized countries on trade. 54 PART ONE: THE MAIN REPORT It also stems from the view that the dominant share of Korean exports in some markets makes further expansion difficult. These concerns are not ill founded, but they tend to be exaggerated. In the last few years quotas, import licensing, and restraints have affected many of Korea's traditional exports. A number of countries restrict textile yarns, fabrics, and garments. Sweden has restrictions on rubber boots; Denmark on flatware; France on radio receivers and chemicals. Although Korean exports would have been greater in the absence of restrictions, Korean exporters have done well to maintain exports in the face of these handi- caps. To circumvent quantitative restrictions, which are usually specified in physical units, Korean exporters have improved product quality and shifted to items having higher unit values, such as camping and ski equipment. There has also been a drive to enter new commodity markets not subject to severe quantitative restrictions. In addition, there is the prospect that international trade will gradually be liberalized, particu- larly into Japan. The extent to which Korean exports will face market limitations in specific products is not easy to assess. An analysis in appendix B, based on fairly broad trade categories, concludes that the problem is not se- rious. As already noted, Korean shares in Europe are small. Korean shares are relatively large only for plywood in the United States and clothing and footwear in Japan. If data were analyzed on a very dis- aggregated basis, it would probably be found that Korea's share is large for some very specific products in particular markets. This finding would not be important. In the past Korean exporters have shown admirable flexibility in adapting product composition to suit market conditions. The volatility of the shares of particular products or countries in total exports confirms that Korea has not become locked into tight product or country markets that might restrict its potential for growth. In summary, the plan's targets for exports are challenging. Given the extraordinary drive of Korean exporters, the continuation of export in- centives now in force, and the assumed growth of OECD economies at an average rate of about 4.5 percent a year during 1977-81, the plan's targets are also attainable. What, then, of the issues of export dependence and vulnerability? The ratio of exports to GNP, the index generally used to assess export dependence, was 26.7 percent in 1975; according to the fourth plan's projections, it will rise to 42.3 percent in 1981. Some argue that these levels are excessive. Korea's continued export orientation has also been criticized for making the economy vulnerable to external shocks that lie entirely outside its control. The sudden, massive rise in import prices in KOREA S FOURTH PLAN 55 1973-74 and the deep impact of the recession in export demand in 1974-75 are cited as illustrations. These problems do not appear to be grave in Korea for three reasons. First, the role of exports in expanding demand is vital. Because of Korea's limited domestic market and resources, export expansion alone can generate the increased production and employment needed during the fourth plan period. Second, the exposure to international fluctua- tions implies that it is possible to benefit from booms as well as suffer from recessions. Korea was quick to exploit the booms in 1969 and 1973 and able to borrow its way through the 1974-75 recession without sacrificing much growth. Despite heavy borrowing, the debt service burden did not become excessive and is now declining. Assuming that no arbitrary constraints will be placed on the availability of capital in years of international recession, the indications are that Korea should be able to weather future recessions as well, particularly because of the increased diversification of the composition and markets of Korean ex- ports. Third, comparing gross exports with GNP, which is a measure of value added, is not strictly correct. The ratio of value added in manu- factured exports to value added in manufacturing would be a better indicator of the impact of export fluctuations on domestic activity. Al- though a direct estimate of this ratio is not available, it can reasonably be assumed to be the same as the ratio of gross exports to gross manu- factured output. In Korea the ratio of gross exports to gross manufac- tured output is estimated to have been about 23 percent in 1975 and is projected in the fourth plan to rise to nearly 30 percent in 1981. This level of export dependence does not seem excessive. The high import content of exports is another source of vulnerability. Unless downturns in export demand can be perfectly forecast, there will be a lag between a decline in export earnings and the cutback in imports of intermediate inputs for the production of exports. This lao would place a direct strain on the balance of payments, perhaps to the extent of one or two month's supply of raw materials imported for export production. At 1975 import levels, the value of this supply wopld be between $200 million and $400 million. The impact would bemuch greater if it were decided that production should be continued to main- tain employment and build up stocks of finished goods. The implications are that Korea should maintain a cushion of additional foreign exchange reserves or be prepared at short notice to resort to short-term borrowing. The financial cost of holding the necessary additional reserves-esti- mated as the difference between the earnings on fairly liquid reserve assets and the interest paid on its marginal borrowing-is unlikely to be 56 PART ONE: THE MAIN REPORT more than $25 million a year. This would seem to be a small price to pay for the benefits of continuing export growth. Although the fourth plan's aggregate targets for exports are feasible and strategically appropriate, there is some doubt whether the projected change in the structure of Korean exports can be achieved as rapidly as projected (see table 2.3). The sharpest increases over the plan period Table 2.3. Export Targets, by Commodity Group, 1975-81 1975 1981 Millions Percent- Millions Percent- Percent- of age of of age of age of Commodity group dollars total dollars total increase Steel and metal manufactures 367 7.2 1,040 7.3 7.4 Machinery 289 5.7 1,415 10.0 12.4 Electronics 409 8.0 1,940 13.7 16.9 Ships 138 2.7 910 6.4 8.5 Chemicals 188 3.8 930 6.6 8.2 Textiles and clothing 1,817 35.8 3,740 26.4 21.2 Footwear 191 3.8 650 4.6 5.1 Plywood 243 4.8 500 3.5 2.8 Other manufactures 669 13.1 1,910 13.5 13.7 Primary products and processed food 770 15.1 1,130 8.0 4.0 Total 5,081 100.0 14,165 100.0 100.0 Note: All figures are in 1975 prices. Source: EPB, Fourth Plan. are projected in the shares of machinery, ships, and chemicals. The real growth of exports of these products is to be about 30 percent a year from 1976 to 1981. Electronics, the next fastest growing group, is to grow at 22 percent a year. The growth of light manufactured exports is projected to average about 11 percent a year. The share of textiles and clothing is projected to decline sharply from 35.8 percent to 26.4 percent of total exports. The general direction of the shift in export strategy is sound. In the long run Korea's exports of textiles and clothing will face increasingly stiff competition from other developing countries. Korea is attempting to move into exports that require higher levels of technology and trained labor-exports that are becoming uneconomical in industrialized coun- tries because of the labor content and that still lie beyond the grasp of most developing countries. Machinery, electronics, and ships fit this objective well. Although the danger to Korea's traditional exports does KOREA'S FOURTH PLAN 57 not appear to be immediate, it is necessary to put an early emphasis on structural shifts because of the relatively long periods necessary for the development of sophisticated product groups. Technological innovation, quality improvement, and market develop- ment will be important factors in determining whether these ambitious structural changes can in fact be achieved. In shipbuilding, for example, the development of a domestic design capability and the extension of credits to purchasers will be crucial. There is provision in the plan for ship sales on credit. But the question is open whether Korea can effec- tively compete with established shipyards in industrialized countries. These shipyards are hungry for business and get full support from their governments for orders and credit facilities. The problems are likely to be greater, however, for industrial machinery and electronics. Although a beginning has been made in electronics-with plans for setting up an electronics institute near Gumi, establishing a product development fund of about $20 million, and importing telecommunications technol- ogy-the precise steps to be followed in a sustained program of develop- ment have yet to be outlined. Similarly, in industrial machinery, it is not yet clear whether the package of incentives is adequate for en- couraging the growth of efficient manufacturing operations capable of exporting high-quality products. Prudence requires some provision for the possibility that exports of machinery, ships, and electronics will fall short of plan targets. If short- falls occur, they can probably be more than offset by the traditional light manufacturing categories. The main constraint might well turn out to be supply. The plan's investment program and the projected restraints on credit expansion and external borrowing imply that the unplanned light-industry sectors might not be able to raise enough investment capital for the required expansion of capacity. Without adequate capac- ity, Koreans exporters wvill not be able to exploit fully the market opportunities that arise; their ability to do so in the past has been the key to their success. If this problem can be solved, light industrial exports should continue to contribute substantially to export growth in the plan period. In addition, foundations can be laid for a radical change in the structure of Korean exports in later years. Investment Program The fourth plan projects annual growth in GNP of 9.2 percent a year. The gross fixed investment required to sustain this growth is projected to increase by 7.8 percent a year and total W18 trillion in 1975 prices. 58 PART ONE: THE MAIN REPORT In relation to GNP this investment is slightly lower than investment during the third plan period-26.2 percent of GNP, compared with 26.9 percent. The largest increases in the shares of investment are for elec- tricity and social development (table 2.4). The share of manufacturing investment in total investment is projected to increase by only 2 percent. The share of investment in services, which include construction, trade, banking, public administration, and other services, has been reduced from 18 percent of total investment to 8 percent, an absolute cutback of about 25 percent. Analyses of major sectors detailed in subsequent chapters indicate that the fourth plan underestimates Korea's investment needs. The allo- cation for manufacturing appears to be low. The allocation for miscel- laneous services, when compared with that in the previous plan period, has been unrealistically curtailed. The allocation for housing, about 13.4 percent of the total, provides for only a slight alleviation of the growing demand caused by rapid urbanization. The reserve margins proposed for the power sector should be reevaluated, particularly because of likely delays in the construction of nuclear power plants. At the same time, reductions in investment could be justified in some sectors. For transport a number of investments might well turn out to be uneco- Table 2.4. Allocation of Investment, by Sector, 1972-81 1972-76 1977-81 Billions Percent- Billions Percent- Percent- of age of of age of age of Sector won total -von total increase Agriculture, forestry, and fishing 1,341 11.8 1,979 10.4 8.4 Mining 139 1.2 302 1.6 2.1 Manufacturing 2,805 24.6 5,088 26.7 30.0 Electricity 770 6.8 2,479 13.0 22.4 Transportation 1,856 16.3 2,784 14.6 12.2 Communication 346 3.0 750 3.9 5.3 Science and technology 100 0.9 203 1.1 1.4 Other services 2,028 17.8 1,525 8.0 -6.6 Social development 2,023 17.7 3,918 20.6 24.9 Total 11,048 100.0 19,028 100.0 100.0 Note: All figures are in 1975 prices. Source: upB, Fourth Plan. a. Investment allocations in this section are presented on an outlay basis and include interest charges during construction, land purchases, research and devel- opment, and purchases of second-hand equipment. KOREA'S FOURTH PLAN 59 nomic if subjected to detailed cost-benefit analysis. It also is evident that the proven organizational ability of Korean enterprise can squeeze more use out of the existing transport infrastructure if operating im- provements are made. For agriculture some reduction in investment in land development and livestock could probably be justified on economic grounds. The magnitude of possible reductions in the investment pro- gram in specific sectors can only be estimated after undertaking more detailed analysis. Some savings are contingent upon a considerable administrative effort and, to some extent, on a reconsideration of the self-sufficiency goals set for grains other than rice.4 Consequently it is doubtful whether savings in these sectors can be realized during the fourth plan period. This conclusion makes it prudent to consider in- creasing the size of the total investment program. There is, of course, a considerable margin of error in estimating invest- ment requirements. Using the crude criterion of unlagged ICORs, it is found that the fourth plan projects an ICOR of 3.1, which is about the same as that realized over the 1965-75 period. Because of the large increase in the proportion of investment allocated to the power sector, some increase in the IOoR should have been assumed. The power sector is capital intensive, and the prices of capital goods for power installations have risen sharply in recent years. Based on separate ICORs assumed for each sector, the macroeconomic model described in appendix A esti- mates that the aggregate ICOR for the 1977-81 period will be about 3.2 and that the total investment required is about 7.5 percent more than the fourth plan's projections. Furthermore the adequacy of the fourth plan's investment program cannot be judged solely in the context of the plan period. Because of lags between investment and output, investment in the later part of the plan period should be related to the rate of economic growth projected for the 1980s. Here again it is found that the fourth plan is in danger of underestimating investment requirements. Analysis of Korea's long-term development strategy suggests that continuing high rates of growth are necessary to prevent a worsening of unemployment and poverty in the 1980s. From this perspective the sharp decline in investment proposed for heavy and chemical industries in 1980 and 1981 might prove to be unwise.5 The foregoing analysis suggests that the fourth plan underestimates 4. See chapter 7 for further discussion of this point and, more generally, of issues and constraints facing the agricultural sector in Korea. 5. Annual investment in these sectors is projected to decline from an average of W670 billion in 1977-79 to W468 billion in 1980-81. 6o PARr ONE: THE MAIN REPORT the investment required to achieve its export and output targets. The extent of the underestimation, given the unavoidable margins of error inherent in such estimation, probably is not large. It should be noted, however, that the risks associated with underestimating and overestimat- ing resource requirements are not symmetrical. Underestimation can lead to bottlenecks, possibly to a loss of export opportunities if capacity is un- available, and in the extreme to a serious slowdown in the momentum of growth. Overestimation is less harmful for three reasons. Planned major investments can be delayed if they tum out to be unnecessary. Invest- ments in the unplanned sectors will take place onlv in response to de- mand conditions. And excess capacity is quickly absorbed in an economy that is growing as fast as Korea's. Based on these considerations, a reasonable investment allocation for 1977-81 would be about 5 to 10 percent more than the fourth plan specifies. This increased allocation would impose the burden of mobiliz- ing an additional WI to W2 trillion in 1975 prices during the plan period. In aggregate this increase is small. Its impact would be signifi- cant, however, if the additional resources were allocated to the manu- facturing and services sectors, for which the planned investment alloca- tions probably are too low. Domestic Resource Mobilization The fourth plan's targets for mobilizing domestic savings are ambi- tious. Although the investment ratio is projected to decline from 27.7 percent of GNP in 1976 to 26 percent in 1981, the requirements for domestic savings rise because the plan projects the elimination of foreign savings by 1979. National savings excluding transfers are projected to increase from 18 percent of GNP in 1975 to 26 percent in 1981 (table 2.5). The plan assumes a national savings ratio of 21 percent in 1975 prices for the base year 1976 and a marginal saving rate of 35.3 percent for the plan period. But the realized savings ratio of 21 percent in 1976 was artificially high because of the exceptional circumstances noted earlier. With the acceleration of the GNP growth rate and the recovery of the international terms of trade, gross national income in 1975 prices grew by more than 20 percent in 1976, compared with only 5 percent in 1975. Allowing for these factors, particularly the normal lag between consumption and income, the savings ratio for the base year should probably be adjusted downward to about 20 percent. This ratio would still represent an excellent improvement over the savings ratio of 18 percent for 1975. KEOREA'S FOURTH PLAN 61 Table 2.5. Investment and Savings, 1975-81 Percentage of GNP 1977-81 Percent- Billions age of Item 1975 1976 1981 of won GNP Fixed capital formation 25.7 25.7 24.0 16,633 24.2 Increase in stocks 1.6 2.0 2.0 1,375 2.0 Total 27.3 27.7 26.0 18,008 26.2 National savings' 18.0 21.0 26.1 16,645 24.2 Net transfers 1.2 1.0 0.5 464 0.7 Foreign savingsb 10.1 5.7 -0.6 899 1.3 Statistical discrepancy -2.0 ... ... ... Zero or negligible. Source: EPB, Fourth Plan. a. Excludes transfers. b. Defined as the deficit in the current account of the balance of payments. The marginal saving rate of 35.3 percent assumed for the plan period is much higher than recent rates, which have averaged about 25 percent. Bearing in mind the tremendous importance attached to achieving an im- provement in savings performance, a marginal saving rate of around 30 percent would be more realistic. If this rate is assumed, aggregate national savings excluding foreign transfers are likely to reach about W15.4 trillion for the plan period, or about W1.2 trillion less than projected. When compared with the past, this performance would still imply a significant increase in the average ratio of savings to income (table 2.6). Because the fourth plan's projections are presented in 1975 prices, the historical data in table 2.6 have been converted to 1975 prices and ad- justed for changes in the terms of trade. For 1976 it has been assumed that the savings ratio was 20 percent. If a ratio of 21 percent is assumed as the base, it is estimated that available savings for 1977-81 would rise by W500 billion to W15.9 trillion. Even this performance would fall short of the projection in the plan by about W700 billion. The foregoing assessment of projections of aggregate domestic savings is supported by more detailed examination of projections for public, corporate, and household savings in chapters 11 and 12. These chapters also discuss policy measures devised to bring about necessary improvements in sav- ings. The conclusions of this examination are: targets for public savings are feasible, although some hard decisions remain to be made and there may be some undesirable fallout effects in relation to prices, income 62 PART ONE: THE MAIN REPORT Table 2.6. Gross National Savings and Income, 1967-81 (billions of won in 1975 prices) 1977-81 Fourth World Bank Item 1967-71 1972-76 plan estimate Gross national savings' 3,699 8,202 16,645 15,400 Gross national income 27,403 44,001 68,744 68,744 Savings ratio (percent) 13.5 18.6 24.2 22.4 Note: Figures have been adjusted for changes in the international terms of trade, using 1975 as the base year. Sources: BOK, National Income in Korea, 1975, pp. 141-43; Box, Economic Sta- tistics Yearbook, 1976, pp. 260-63; EPB, Fourth Plan; World Bank estimates. a. Excludes transfers. distribution, and household savings; targets for corporate and household savings probably are high (table 2.7). External Resource Requirements Korea's requirements for extemal resources during the fourth plan period must be assessed in the context of the extraordinary volatility of the balance of payments during the later part of the third plan period. Table 2.7. Savings, by Sector, 1975-81 (percentages of GNP in 1975 prices) Sector 1975 1976 1981 Public savings 5.3 5.6 6.1 Private savings 12.7 15.4 20.0 Corporate net surplus 2.0 2.9 3.9 Depreciation 7.8 7.9 9.4 Households 2.9 4.6 6.7 Total national savings' 18.0 21.0 26.1 Note: The figures for 1975 are actual; those for 1976 estimated; and those for 1981 planned. Source: EPB, Fourth Plan. a. Excludes transfers. KOIRA'S FtOURTH PLAN 63 Table 2.8. Unit Value and Volume of External Trade, 1972-76 (percent) Change from previous year Item 1972 1973 1974 1975 1976 Commodity exports Unit value 1.1 26.6 26.6 -7.4 11.7 Volume 46.4 54.0 8.9 19.7 39.4 Commodity imports Unit value 1.7 33.5 55.5 2.9 -1.7 Volume 3.5 25.9 3.9 3.2 26.1 Sources: Tables D.13 and D.41 in appendix D. That volatility arose from large and sudden changes in the unit prices and quantities of Korea's exports and imports (table 2.8). Those changes, in turn, were to a large extent imposed by external circum- stances. The higher import bill for 1974-75 was the result of the inter- national scarcity of foodgrains and the increase in the price of oil. Fluctuations in export prices were caused by the buoyancy of export demand in 1973 and early 1974 and by the protracted recession in industrialized countries after mid-1974, when developing countries be- gan to engage in stiff price competition in an attempt to retain their respective shares of manufacturing exports in a sluggish market. It obviously is impossible to foresee the changes in the terms of trade or the conditions of export demand that might suddenly befall Korea during the fourth plan period. But it is reasonable to believe that changes are unlikely to be as massive as those of the preceding few years. The plan assumes that prices of exports and imports will rise at 5 percent a year and that the terms of trade will be constant at 5 percent more than their rather low level in 1975 and slightly less than that in 1976 (table 2.9). More recent estimates of the terms of trade show that the gain in 1976 was about 14 percent, not 8 percent as assumed in the fourth plan. Thus the prospect for the terms of trade has consider- ably improved, even though they are still below the levels prevailing before the increase in oil prices. The principal issues are projections of the volume of trade and the availability of external capital. Export prospects were discussed earlier in this chapter: it was concluded that the plan's export projections rep- 64 PART ONE: THE MAIN REPORT Table 2.9. Terms of Trade, 1974-81 (1975 = 100) Item 1974 1975 1976 1977 1981 Export price 108.0 100.0 112.0 117.6 142.9 Import price 97.2 100.0 104.0 112.3 136.5 Terms of trade 111.1 100.0 107.7 104.7 104.7 Source: EPB, Fourth Plan. resent a challange, but can be met. It appears, however, that the import projections, particularly of crude oil and capital goods, are unrealis- tically low (table 2.10). The plan's estimates of petroleum demand are derived from projections of the total demand for energy and the extent to which this energy can be supplied by coal and nuclear fuel. These projections have an optimistic bias that understates the 1981 require- ments for crude oil imports by a large margin, possibly as much as 14 percent (see chapter 9). Because oil imports are projected to be about 17 percent of total commodity imports in 1981, the addition to total import requirements would be about 2.4 percent. Imports of capital goods are projected to grow at 11.2 percent a year. This rate of growth probably involves a substantial effort at import substitution in machinery, electronics, and shipbuilding. The major concern in both machinery and electronics is the ability of Korean Table 2.10. Composition of Commodity Imports, by Group, 1975-81 1975 1976 1981 Mil- Mil- Mil- 1976-81 lions Per- lions Per- lions Per- annual of cent- of cent- of cent- growth Commodity dol- age of dol- age of dol- age of rate group lars total lars total lars total (percent) Capital goods 1,909 26.2 2,237 22.2 3,804 25.5 11.2 Crudeoil 1,271 17.5 1,490 18.5 2,523 16.9 11.1 Grain 689 9.5 514 6.0 556 3.7 1.6 Raw materials for exports 2,180 30.0 2,944 31.2 5,670 38.0 14.0 Other imports 1,225 16.8 1,265 22.1 2,362 15.8 13.3 Total 7,274 100.0 8,450 100.0 14,915 100.0 12.0 Note: All figures are in 1975 prices. Source: EPB, Fourth Plan. KOREA'S FOURTH PLAN 65 entrepreneurs to absorb modern technology and maintain necessary standards of precision and quality. There are considerable doubts about whether the domestic machinery industry will develop fast enough to achieve the requisite degree of import substitution, in such a short time, without recourse to import quotas on machinery. If restraints on im- ports become excessive, the effect would be to reduce the competitiveness of Korean goods in foreign markets and jeopardize the efficiency of users of domestically produced machinery.6 Because of the remarkable improvements in domestic grain production and the increase in stocks in recent years, the plan's projections for imports of grain are reasonable. The projections for imports of raw materials for exports and for other imports might be low. Analysis with an input-output model reveals that the import intensity of exports is likely to increase over the plan period because of the changing composition of exports. It also seems that the restraint imposed on the growth of miscellaneous imports in 1974 and 1975 will be difficult to sustain. Even if it is possible to conserve on imports to the extent projected in the plan, it will be necessary to run a balance-of-payments deficit in order to finance the investment program. The reason for this is the inadequacy of national savings. For the entire plan period, foreign savings are projected at only 1.3 percent of GNP (see table 2.5). As concluded earlier, the projected increases in national savings are high, and the likelihood of a shortfall should be recognized. It was also con- cluded that the fourth plan already underestimates its investment re- quirements. Combining the estimated additional investment require- ment of at least Wl trillion and the estimated shortfall in national savings of about WI trillion, the additional requirement for external financing is at least W2 trillion ($4 billion) in 1975 prices for the 1977-81 period. If this assessment is correct, the deficit in the current account of the balance of payments during the plan period would aver- age about 4 percent of GNP in 1975 prices. These estimates do not neces- sarily imply increases in the requirements for medium- and long-term capital. To some extent the required finance can be found within the scope of the plan's projections of capital availability. The plan projects a substantial net repayment of short-term borrowings, a process that could easily be delayed if necessary. But if the expansion of the volume of 6. These considerations also apply more generally to the entire range of heavy and chemical sectors, which will be the target of increasing import substitution both during and beyond the fourth plan period. 66 PART ONE: THE MAIN REPORT trade continues, the scope for short-term financing will increase, not diminish. It should nevertheless be stressed that, because of the improvement in the terms of trade, there is a large difference between the resource gap figured in constant 1975 prices and that figured in current prices. It should also be noted that the relative prices of investment goods significantly declined from 1975 to 1976. As mentioned earlier, the fourth plan projects the international terms of trade at a level 5 percent higher than that in 1975. Given the 14 percent improvement in the terms of trade in 1976, this assumption might turn out to be conserva- tive. The resource gap in current prices might become very small or even disappear.7 The difference could be substantial: if the index of the terms of trade during 1977-81 is 109 (1975 = 100) and not 105 as assumed in the fourth plan, the financing gap in current prices would be reduced by nearly $3 billion over the plan period. Thus, unless there is a setback from the 1976 level of the terms of trade, the balance-of- payments deficits in current prices will be only marginally higher than projected. Nevertheless uncertainty about the prospects for OECD econ- omies compounds the difficulty of making firm projections of Korea's export earnings and financing requirements in current prices. Because of these uncertainties and the Korean economy's exposure to interna- tional fluctuations, it is important that an adequate flowv of new long- term commitments be maintained. The main task in establishing such commitments is to achieve the plan's targeted inflow of medium- and long-term capital at an average rate of at least $2.4 billion a year (table 2.11).s The magnitude of direct private investment continues to be fairly small. Nearly half of the gross disbursements of medium- and long-term capital will be absorbed by amortization; another quarter by medium- and long-term export credits. To generate the disbursements as shown in table 2.11, the plan projects new commitments of medium- and long-term debt at an average of $2.5 billion a year (in current prices), with about $1.1 billion from official sources and $1.4 billion from commercial sources. Multilateral agencies, such as the World Bank and the Asian Development Bank, will prob- ably have to provide about half the new official commitments. Total medium- and long-term outstanding debt, excluding undisbursed com- 7. In fact, that is precisely what occurred. 8. The fourth plan defines medium- and long-term capital as having maturities of more than three years. The inflow of capital with maturities between one and three years is small. KOREA'S FOURTH PLAN 67 Table 2.11. Sources and Uses of External Capital, 1977-81 (millions of dollars in current prices) 1977-81 annual Item 1977 1981 average Sources Direct private investment 84 162 123 Medium- and long-term capital 1,998 2,472 2,359 Short-term capital and other 46 -1,546 -873 Uses 2,129 1,091 1,611 Current account deficit 634 -1,172 -243 Amortization 784 1,445 1,144 Addition to reserves 711 818 710 Source: EPB, Fourth Plan. mitments, is estimated to increase from about $7 billion at year-end 1976 to $13 billion at year-end 1981. The service of medium- and long-term debt is estimated to rise from about $0.9 billion in 1976 to almost $2 billion in 1981. Export earnings are projected to increase even faster, however, and the debt service ratio is to decline from about 13 percent in 1975 to about 11 percent in 1981.9 Although the tremendous improvement in the balance of payments in 1976 increased the confidence of private lenders in Korea's long-term strength, it still is important that a reasonable proportion of Korea's new borrowing continue to be met from official sources. Commercial lending to Korea still is on rather onerous terms and almost always is at maturi- ties less than ten years. As noted earlier, the bunching of major invest- ments is substantial in the plan's early years, when the financing re- quirement is greatest."' Excessive reliance on medium-term loans from commercial sources would result in a bunching of amortization require- ments before Korea has fully recovered from the consequences of the balance-of-payments crisis of 1974 and 1975. Korea would then become critically dependent on its ability to roll over its short- and medium-term 9. The debt service ratio is defined as the ratio of interest and amortization of debt with maturities exceeding one year to the exports of goods and nonfactor services. 10. It is estimated that major projects in the manufacturing sector alone will require capital-goods imports of $1.6 billion in the first three years of the plan. 68 PART ONE: THE MAIN REIPORT commercial borrowing-a condition which greatly increases the magni- tude of gross inflows and exposes the economy to greater risk. This con- sideration is of particular importance in Korea because the economy is exceptionally sensitive to fluctuations in export earnings and because the import content of exports is large. Allowing for the fact that net export earnings are only about 60 percent of their gross levels, the debt service ratio would be nearly twice as high as that measured in the conventional manner. That is, the conventional ratio of 10 percent in 1981 translates to nearly 17 percent on a net basis. It should also be noted that debt service will remain at more than 4 percent of GNP during the plan period, indicating the scale of its burden on the economy. Employment and Wages One major objective of the fourth plan is to prevent a deterioration of the employment situation. Given the projected growth in the labor force, more than 2 million jobs will have to be created during the plan period to accomplish this objective (table 2.12). The consensus among seveial projections of Korea's labor supply is that it will grow at slightly more than 3 percent a year (see chapter 6). This growth corresponds to the assumption in the fourth plan. The basic reason for such a high growth rate is the delayed result of the baby boom that followed the Korean War. Persons born during the baby boom entered the working- Table 2.12. Labor Force, 1976-81 (millions of persons) Annual growth rate Item 1976 1981 (percent) Total population' 35.9 38.8 1.6 Employable populationb 23.8 26.1 2.7 Labor force' 13.2 15.4 3.1 Source: EpB, Fourth Plan. a. Chapters 4 and 6 argue for slightly higher rates of growth of population and labor force. b. Defined as the population aged 14 and over, excluding military, prisoners, and other institutional persons. c. Referred to in the fourth plan as the economically active popu- lation. KOREA'S FOURTH PLAN 69 Table 2.13. Employment, by Sector, Projections for 1976-81 (millions of persons:) Annual growth rate Sector 1976 1981 (percent) Agriculture, forestry, and fishing 5.5 5.6 0.5 Mining and manufacturing 2.5 3.3 5.7 Services 4.2 5.4 5.1 Total 12.2 14.3 3.3 Source: Projections of the macroeconomic model described in ap- pendix A. age population in the late 1960s and early 1970s; the growth rate of the employment-eligible population peaked at 4 percent a year in 1971-72. The growth rate of the labor force nevertheless is likely to remain high for many years because an increasing proportion of the population is entering age groups that have displayed relatively high rates of labor force participation in the past. The plan's projections of growth in employment are supported by findings of the macroeconomic model described in appendix A. The model assumes that GNP will grow at 9.1 percent a year during 1977-81 and projects that employment will grow at 3.3 percent a year (table 2.13). This employment growth would keep the unemployment rate at about 4 percent. Given the underemployment in the agriculture sec- tor, it is not surprising that about 38 percent of new jobs are to be in mining and manufacturing. Employment in services is projected to grow slightly faster than that in manufacturing. The aggregate employ- ment elasticity implied by these projections is 0.36, which is close to the historical average of 0.37 for the 1963-75 period. Whether the historical relation between employment and output continues to hold will depend on the consequences of projected changes in the structure of the economy, particularly the manufacturing sector. As is described in chapter 8, the share of textiles and other light industries in manu- facturing value added is projected to decline sharply during the plan period from 58 percent to 41 percent; that of machinery, electronics, and shipbuilding is to rise sharply from 15 percent to 24 percent; that of basic metals is to rise from 2.6 percent to 3.7 percent; that of chemicals is to decline from 25 percent to 22 percent. The net effect of these structural changes on the rate at which manufacturing growth 70 PART ONE: THE MAIN REPORT will create new jobs will depend on the labor content in machinery, electronics, and shipbuilding relative to that in light industries."' Shipbuilding is more labor intensive in Korea than in other countries. The proposed shift in production from large tankers to smaller vessels will tend to increase the labor intensity of shipbuilding during the plan period. The elasticity of employment in machinery and electronics is difficult to predict with confidence. Currently electronics is quite labor intensive because of the concentration on assembly operations. But the continued growth of this sector will depend on shifting to sophisticated product categories which are more intensive in technology and capital. The machinery sector is extremely heterogeneous, and the extent of its employment generation will depend on the way the sector develops from its current embryonic state. Major investments in this sector-for ex- ample, the large investments in the machine tool industry-are likely to be for products that are relatively less labor intensive than those in light industry. It is possible, however, that some other subsectors likely to expand rapidly in Korea, such as foundries and metal fabrication, might well be more labor intensive. The precise direction of the machinery sector's development and the extent to which growth will take place in smaller, labor-intensive establishments are still unclear. It would be surprising if the elasticity of employment in the new sectors exceeds that in light manufacturing. But because the light manufacturing sectors are expected to expand faster than projected, it is probable that the employment projections of the fourth plan will indeed be met. Nominal wages rose by 36 percent in 1974 and 26 percent in 1975, compared with increases of 25 percent each year in the consumer price index. In 1976 the consumer price index increased by about 12 percent in nominal terms. The deceleration in the rate of inflation should be reflected in a corresponding deceleration in the rate of growth of nomi- nal wages. Failing this, real wages will probably be growing faster than labor productivity which, in turn, would result in rising wage costs in the manufacturing sector. Korea's ability to keep unit wage costs under control is of vital importance in maintaining the intemational competi- tiveness of export industries. Given the extraordinary influence that govemment agencies can exert on wage settlements, this would normally be easier to accomplish in Korea than elsewhere. The strength of gov- ernment influence in this regard is nevertheless being eroded by the relatively large salary increases granted to civil servants since 1974 and 11. The major light industries in Korea are textiles, garments, accessories, leather and wood products, wigs, and other miscellaneous manufactures. KORfEA'S FOURTH PLAN 71 explicitly provided for in the fourth plan. There is wide agreement that government salaries are undesirably low in Korea. Government's objec- tive is gradually to raise the salaries of civil servants to the level of salaries paid in government corporations. To this end, government salaries were raised 150 percent during 1974-76, compared with an increase of 71 percent in manufacturing earnings. They are expected to rise at a nominal rate of slightly more than 25 percent a year during the plan period.12 In the interest of maintaining the degree of wage restraint necessary in the Korean economy, some moderation of these increases appears to be desirable. Although wage restraint is demanded by macroeconomic considera- tions, the problems persist of having large numbers of unskilled laborers work long hours for low wages. Their condition has led to suggestions for minimum-wage legislation. As already noted, Korea will have con- siderable difficulty absorbing labor at a satisfactory rate during the plan period. Its comparative advantage continues to depend on labor costs that are low relative to labor skills. If a minimum wage were set at a level that affected many workers, labor costs would rise substantially and affect technical choices and the absorption of labor. To avoid this out- come, minimum wages could be set at a level that would prevent the exploitation of isolated pockets of workers, but not affect the wages of most workers. Typically, however, the opposite occurs. Large firms, which already pay relatively well, obey the law, raise wages, and slow down their rate of absorbing labor. Small firms ignore the law with im- punity because it is not easily enforced. Thus minimum-wage legislation probably is inadvisable at Korea's present stage of development. Tough legislation for industrial safety, broadly enforced, would be more appro- priate. Even if measures to improve work conditions were to add slightly to industrial costs, they should have less impact on employment than minimum wages because the costs associated with such measures usu- ally do not increase in step with the number of workers affected. Social Development The fourth plan proposes a considerable expansion of social develop- ment activities. Except for housing, the shares of all social development 12. The increases are not planned to be granted across the board, and the opportunity will be taken to rationalize the structure of pay scales as well. Given the scope and complexity of these problems, it would seem appropriate to have a thorough, independent study of the proposed reform of salaries of civil servants. 72 PART ONE: THE MAIN REPORT Table 2.14. Investment Program for Social Development, by Service, 1972-81 1972-76 1977-81 Percentage Percentage Billions of total Billions of total Service of won investment of won investment Education and manpower development 254 2.2 749 3.9 Health 38 0.3 211 1.1 Housing 1,670 14.6 2,641 13.9 Water supply and sewerage 61 0.5 182 1.0 Total 2,023 17.7 3,783 19.9 Note: All figures are in 1975 prices. This table follows the classification of ex- penditure in the fourth plan. See the notes to table 5.1 for details about items included and excluded under this classification. Source: EPB, Fourth Plan. sectors in total expenditure will be greater in 1977-81 than in 1972-76 (table 2.14). Public expenditure for housing will nevertheless increase as a proportion of the total. Recurrent expenditure on education is not regarded as part of the plan's investment outlays, but it will rise sharply despite the fact that there will be little growth in the school-age population. The plan proposes to expand enrollment at all levels above primary education and to improve the quality of vocational and scientific train- ing. One objective is to achieve 90 percent enrollment in middle secon- dary schools. This target is ambitious, but it may be appropriate in Korea given the apparent contribution of high levels of basic education to development. The plan also projects that the advancement rate from middle school to high school will be nearly 80 percent in 1981, when it is expected that one-half of 18-year-olds will have graduated from high school.13 The growth planned for institutions of tertiary education will also be fast. The most rapidly growing institutions will continue to be the junior technical colleges. It is expected that the intake of two- year colleges will nearly double over the plan period. Four-year institu- tions, in contrast, will grow by only about 4 percent a year. 13. This proportion is very high by international standards. The proportion of the population aged 17 and 18 in school has already reached the levels found in such developed countries as England and New Zealand. KORE.A'S IFOURTH PLAN 73 For skill formation the plan forecasts potential surpluses at most training levels, and much heavier reliance will be placed on enterprise training. It nevertheless is difficult to make such predictions accurately, especially when the economy is undergoing a rapid structural transfor- mation. In view of anxieties frequently expressed on this score, it is important to stress that the current stock of trained manpower-crafts- men, technicians, engineers, and scientists-already constitutes a high proportion of total employment in manufacturing (see chapter 6). Not only is the current stock high, the growth of demand is projected to be rapid, and the technical skills of Korean workers will be even more advanced in 1981. The fourth plan provides for only a slight alleviation in the scarcity of housing. Between 1960 and 1975 the ratio of housing units to house- holds fell from 82 percent to 75 percent; it is expected to rise to 80 per- cent in 1981. Of the total increase of 1.3 million housing units, the public sector will be providing 500,000, or a higher proportion than in the third plan period. The average unit cost of urban public housing is estimated to be W1.5 million ($3,100), compared with W2.5 million ($5,200) in the private sector. Even at this low cost, houses will be out of the reach of families in the lowest deciles of the distribution of income. Government has, on several occasions, postponed implementation of the National Welfare Pension Law of 1974. Financing such a large- scale pension scheme would cause considerable problems if it resulted, as seems likely, in an increase in the unit cost of labor. It would raise domestic prices, increase the relative cost of labor, and probably affect the growth of employment, which would be highly undesirable at Korea's present stage of development. Furthermore the benefits, con- fined for administrative reasons to larger firms, would not go to the workers most in need of assistance. Consequently government has focused on the health component of the social insurance scheme. The intention is to improve access to medical facilities. For this effort to be effective, it is necessary to develop an effective delivery system for health services. These considerations appear to be well recognized in Korea. A limited medical insurance scheme for employees of large private com- panies has been enacted, and special emphasis is being placed on the development of an innovative health delivery system to serve the needs of low-income groups. Despite substantial improvements in health conditions in recent years, the incidence of unnecessary morbidity apparently is substantial. Im- munization and low-cost screening and treatment could eradicate much 74 PART ONE. TH1E MAIN REPORT underlying pathology at a cost Korea can afford, as could improved water supplies and sanitation. The plan seems well directed to take care of major priorities. About 20 percent of the planned outlays for health will be used to improve water supplies, mostly in rural and small urban centers; 9 percent will be used for sewerage, mostly in large cities. Ex- panded preventive measures against disease, especially tuberculosis, are also a feature of the fourth plan. This program is consistent with govern- ment's general objective of having public investment in social develop- ment make the maximum contribution to the general well-being of Koreans. Chapter 3 Parvez Hasan and D. C. Rao Devdlopment Issues in the 1980s WILL KOREA BE ABLE during the 1980s to sustain a GNP growth rate of about 10 percent a year, the level achieved during the 1960s and the level likely to be achieved during the 1970s? Will export expansion con- tinue to be crucial for maintaining a high level of economic activity? If it does, what broad structural shifts will be needed to keep the pattern of exports in line with Korea's shifting comparative advantage? Will re- source constraints continue to be a major policy concern in the 1980s? Or will the focus increasingly shift to issues, so often dominant in the later stages of industrialization, of technological improvement and upgrading labor skills? Such issues concerning the feasibility of continuing high rates of growth cannot, and should not, be isolated from the general issue of the social desirability of such a course. Is rapid growth indeed worth while? At least two aspects of this question deserve special attention. First, the implications of high rates of growth for employment and poverty need to be compared with the implications of lower rates of growth. Will high rates of growth automatically reduce unemployment and bring down the number of absolute poor? If not, what combination of policies will be necessary to achieve the desired distribution of the benefits of growth? Second, even if the beneficial effects of high rates of growth on the incomes of the poor can be taken for granted, these might need to be matched against problems which often accompany a faster pace of growth. What is to be done, for example, about deterio- ration of the environment, rapid urbanization, and lags in the provision of public services? Some of these and related questions are examined in this chapter. Major structural changes will be more or less inevitable under a variety of developmental paths. For example, given any reasonable range of assumptions about future growth, Korea will be a fairly high-income, 75 76 PART ONE: THE MAIN REPORT heavily industrialized, export-dependent economy with a rapidly dwin- dling reliance on agriculture in 1990. The Korean population in that year will be about 47 million, nearly one-third more than in 1976. Assuming average growth of 8 to 10 percent a year, per capita income in 1975 prices will have increased to a range between $1,200 and $1,800. Man- ufacturing is likely to account for about 40 percent of GNP, compared with more than 30 percent in 1976; agriculture will contribute only about 10 percent, compared with more than 20 percent in 1976. Exports will rise sharply as a proportion of GNP and could conceivably outstrip GNP. But the proportion of manufacturing exports in gross manufactur- ing output, a more appropriate indicator of export dependence, is likely to be about 40 percent in 1990, compared with 26 percent in 1976. Export Prospects Given the dominant position of manufacturing for export in the economy, it is natural to expect, even in the long term, that interna- tional demand for Korean exports will be fundamental in determining the feasibility of continuing high growth. High rates of export growth help to overcome the constraints of the domestic market and favorably affect savings and aggregate capital-output ratios. The basic reasons for this are that export activities have tended to be less capital intensive than production for the home market and that the growth of export demand appears to have been an important influence on corporate profitability. In the discussion of the fourth plan's export targets in chapter 2, the underlying strength of Korea's export prospects was asserted. Because the subsidization of exports has declined in recent years, the strong com- petitive position of Korea's export production has become more evident, providing some basis for optimism regarding export prospects. One important question, however, is whether Korean exports can, in fact, continue to expand in the 1980s at 16 percent a vear, the growth rate assumed for the next several years. This expansion will depend critically on trends in world trade in manufactured goods. Since the 1950s world trade in manufactured goods has grown at a pace considerably faster than world production of manufactures. Even though much of the absolute expansion in trade has been among the industrial countries, the rate of growth of manufactured exports from developing countries, estimated at about 16 percent during 1965-74, has been quite high. DEVELOPMENT ISSUES IN THIE I980S 77 Barring major international crises, the encouraging trend witnessed during the last twenty-five years could continue for a long time. But it is worth noting that in no other period during the last 100 years has world trade in manufactures grown faster than world output of manu- factures. Between 1876-80 and 1911-13 the volume of trade in manu- factures increased by more than 3 percent a year; production grew by an estimated 4 percent a year. Between 1911-13 and 1948-50 world trade in manufactures grew by less than I percent a year because of disruptions caused by world wars and the great depression.' Assuming that world trade in manufactures will grow at 8.5 percent a year, compared with an average of 9.5 percent a year during 1960-73, Korea's share in total world trade in manufactured goods will have to rise from nearly I percent in 1975 to nearly 3 percent by 1990 in order to achieve growth of 16 percent a year in Korean exports. Given Korea's excellent export perfornance and continued emphasis on exports, sub- stantial further growth in Korea's share in world manufactured exports seems feasible. During 1955-72 Japan increased its share in world com- modity exports from 2 percent to nearly 7 percent; the Federal Republic of Germany from 7 percent to 11 percent.2 Thus a further substantial rise in Korea's share of world trade in manufactures is not without precedent. The continued high growth rate of Japanese exports, averaging 15 percent a year during 1955-70, and the marked increase in its share in world manufactured exports were made possible bv major shifts in the structure of exports. There were three major growth sectors in Japanese exports, especially during the last decade: automobiles, ships, and con- sumer electronics. The share of automobiles and trucks in total exports rose from 0.5 percent in 1955 to 3 percent in 1965 and 11 percent in 1973. Exports of consumer electronics were negligible in 1955, but rose to 6 percent of the total by 1973. The share of motorcycles and ships rose from 5 percent in 1955 to 14 percent in 1973, as Japan emerged as the world's largest shipbuilder. The share of textiles and cloth- ing, which were Japan's dominant exports in the mid-1950s, dropped 1. Alfred Maizels, Growth and Trade (London: Cambridge University Press, 1970), p. 80: an abbreviated version of Industrial Growth and World Trade (1963). 2. Hugh Patrick and Henry Rosovsky (eds.), "Japan's Economic Performance: An Overview," in Asia's New Giant (Washington: D.C.: Brookings Institution, 1976), p. 57. 78 PART ONE: TH1 MAIN REPORT sharply from 36 percent in 1955 to 16 percent in 1965 and 6 percent in 1973.3 Changes in Industrial Structure In the short or medium run there may sometimes be a tradeoff be- tween the rate of growth of exports and the extent of change in the structure of industrial exports. In the long run, however, a changing pattern of exports, reflecting and responding to shifting comparative advantage, is essential for sustained export expansion. Korean planners are fully conscious of the need to bring about structural changes in exports and industrial output. They are depending on electronics, ma- chinery, and ships to achieve this. It is planned that the share of these groups in total Korean exports will increase from 16 percent in 1975 to 31 percent in 1981. The discussion in chapter 2 suggested that, although the general direction of the shift in export strategy is sound, the per- formance of these groups of exports is likely to fall short of the fourth plan's targets. Thus, during the plan period, there will be a relatively slow reduction in the share of textiles and clothing, which stood at 36 percent in 1975. Although the threat to Korea's international posi- tion in textiles and clothing is not immediate and the base of products and markets is fairly secure and diversified, the rate of growth of this group will come down sharply in the 1980s. Imports of textiles and clothing into developed countries grew in current prices at an average annual rate of 27 percent during 1967-73; clothing alone grew at 36 percent a year. This expansion was the direct result of two factors: the demand of industrial countries for imports increased at a much faster rate than their domestic consumption; the share of developing countries in total imports of industrial countries also increased. At the end of 1974, however, developed countries still supplied 67 percent of the world's exports of textiles and clothing, which were estimated at $42 billion and included $15 billion in clothing. De- veloping countries supplied 24 percent; centrally planned economies 9 percent. In clothing, the share of developing countries was approaching 50 percent. Yet imports of textiles and clothing from developing coun- tries accounted for only 4 percent of apparent consumption in major 3. Lawrence B. Krause and Sueo Sekiguchi, "Japan and the World Economy," in Asia's New Giant, pp. 398-99, 409. DEVELOPMENT ISSUES IN THE r g8os 79 industrial countries in 1973.4 Even assuming, say, a doubling of this ratio in response to shifting comparative advantage and trade liberaliza- tion, the demand of industrial countries for textiles and clothing from developing countries will increase by 9 to 10 percent a year on average over the 1976-90 period. The income elasticity of demand for clothing and footwear probably does not exceed unity. During the fourth plan period the growth will probably be higher, especially for clothing. This will benefit Korea, which in 1975 had two-thirds of its textile and clothing exports in clothing, compared with one-half for developing countries as a whole. Korean textile and clothing exports might grow in current prices from an estimated $2.5 billion in 1976 to $6 billion in 1981. Indeed this growth will be necessary if the overall export target of $20 billion in current prices is to be reached. In the early 1980s the intemational demand for textiles and clothing might well begin to slow down. At the same time, competition among such major exporting countries as Hong Kong, Singapore, the Republic of China, and Korea will be in- creasing. For purposes of long-term planning, therefore, it is not realistic to rely on growth in textile and clothing exports of more than 7 or 8 percent a year during the 1980s. Given this growth, achieving an overall growth rate of 16 percent a year over the 1981-90 period will require a sustained growth of 17 percent a year for exports of all other manu- factured goods exports; as a result the share of textiles and clothing in total Korean exports would come down to 10 or 11 percent by 1990. This conclusion is not intended to throw doubt on the possibility of sustaining high rates of export growth, but merely to illustrate the shifts in the pattern of exports that will be required to do so. The foregoing discussion again demonstrates that government's con- cem with industrial deepening and the shift toward heavy industry exports is justified not because of an immediate weakening of export prospects, but because of a potential slowing down of traditional exports in the 1980s. Unless new exports are vigorously developed, Korea can- not hope to capture an increasing share of the world market of manu- factured goods. Because govemment's rather ambitious plans for ship- building have run into the problem of uncertain international demand, electronics and machinery will be particularly crucial in export diversifi- cation and growth. 4. Don Keesing and others, "Draft Report on Textiles and Clothing," World Bank internal paper (Washington, D.C., 1976; processed). 8o PART ON1: THE MAIN REPORT Through these shifts in the structure of the manufacturing sector, Korea will be able to sustain rapid growth of productivity and incomes. A growth rate of 9 percent a year in GNP implies that manufacturing value added will grow at about 12 percent a year which, in turn, implies that value added per worker will have to grow at about 7 percent a year. Productivity in Korea's leading industries, notably in textiles and cloth- ing, already is quite high. As a result, sustaining productivity growth of 7 percent a year, which implies a doubling of labor productivity over a ten-year period, will be difficult unless significant shifts are made toward industries having higher productivity. Export Dependence The long-term prospects for Korean exports have thus far been dis- cussed mainly in relation to intemational demand, under the assump- tion that problems of comparative advantage would not arise. A related question is whether a high and rising dependence on exports is desirable. This question was briefly treated in the discussion of the fourth plan's targets in chapter 2. It was concluded that the prospective increase in export dependence during the next few years is not excessive. If exports continue to grow at about 16 percent a year, a rate considerably faster than output growth, it is likely that the ratio of gross manufactured exports to gross manufactured output will reach about 40 percent in 1990, compared with about 26 percent in 1976.5 This growth implies that about 45 percent of incremental output in the manufacturing sector will be exported. Are these ratios too high? So long as Korea is trading in accord with its comparative advantage, the share of exports in GNP or in the output of the major exporting sector cannot be called too high or too low. Government may nevertheless be concerned about the economy's ex- cessive dependence on external markets. In this context it is useful to compare the export ratios in Korea with those in other countries having a similar trading pattern or a similar level of development (table 3.1). The most striking feature of the table is the wide variation in export ratios, even among trade-oriented countries. Of course, international comparisons of such ratios must take into account the differences in the 5. The proportion of exports to GNP might reach about 90 percent by 1990; as noted earlier, this is not a valid measure of Korea's export dependence. DEVELOPMENT ISSUES IN THIE 1980s 81 Table 3.1. Export Ratios, Korea and Selected Countries, 1960-70 (percent) Manufactured exports to Exports to gross gross manufacturing domestic producta output' Country 1960 1970 1960 1970 Korea 3.4 14.8 n.a. 15.0 Federal Republic of Germany 19.9 22.0 13.2 22.0 Hong Kong 79.3 n.a. n.a. 79.0 Italy 14.7 19.2 n.a. 22.0 Japan 11.1 11.2 8.3 9.0 Netherlands 50.2 47.3 18.3 26.0 Norway 41.1 40.6 18.6 26.0 Republic of China (Taiwan) 18.4' 30.6 n.a. 21.0d Singapore n.a. n.a. n.a. 32.0 Spain 11.2 15.0 4.2 7.0 United Kingdom 20.3 22.6 n.a. 16.0 n.a. Not available. Sources: Exports and GDP from World Bank, World Tables 1976 (Baltimore and London: Johns Hopkins University Press, 1976), series I and series III, table 8; manufactured exports from United Nations, Yearbook of International Trade Statistics, various years; gross manufacturing output from United Nations, Statis- tical Yearbook, various years; and United Nations, The Gro-wth of World Indus- try (New York, 1973). a. Exports include goods and services. b. Manufactured exports are defined as groups 5 to 9 of the Standard Indus- trial Trade Classification (SITC); gross manufacturing output is defined as group 3 of the International Standard Industrial Classification (isic). Because manu- factured exports exclude several commodities included in gross manufacturing out- put, such as processed primary products, export ratios are systematically understated for all countries. Nevertheless, because exports are calculated on an FOB basis and manufacturing output is calculated on an ex factory basis, the ratios also are subject to an upward bias. c. Figure for 1965. d. Figure for 1969. degree of vertical integration in the manufacturing sector and the rela- tive import intensities of manufactured output and exports. Another striking feature is that manufactured exports between 1960 and 1970 grew faster than output in all the selected countries and faster than GDP in most of them. For Korea the ratio of manufactured exports to manufacturing out- put in 1970 was lower than that for all the countries listed in table 3.1 except Japan and Spain. But between 1970 and 1976 there appears to 82 PART ONE: THE MAIN REPORT have been a remarkable increase in this ratio for Korea. The ratio ap- parently increased for the Republic of China and Singapore as well, but probably not by as much as in Korea. The overall impression is that, allowing for such factors as value added in the services sector and the ratio of value added to gross manufacturing output, Korea's depen- dence on manufactured goods exports still is considerably lower than Hong Kong's, somewhat lower than Singapore's and the Republic of China's, but much higher than Japan's. Because world trade in manu- factured goods is projected to continue to grow faster than manufacturing output, the ratio of manufactured exports to manufacturing output will tend to rise globally. Still, if high rates of export growth continue in Korea, it seems likely that it will have a more export-oriented manufactur- ing sector by 1990 than any country except Hong Kong and Singapore. That Korea will be more trade oriented than any other major country need not in itself be a matter of serious concern. But managing the economy in the face of fluctuations in world trade and economic activity will become increasingly important. In the late 1980s Korea will have to balance the economic gains from additional trade against the increas- ing exposure to international fluctuations. It must also be stressed that, as the proportion of domestic demand for manufactures met from im- ports progressively rises, Korea's comparative advantage will be shifting. In other words, meeting home demand will increasingly offer attractive opportunities in a situation in which import reliance has already become high. In summary, the likelv prospect for Korea is that exports will continue to grow faster than output and that an increasing share of manufactured output will be exported. After a while this trend may be automatically moderated as domestic demand picks up. Or, if it is felt that exporting about one-half of total manufactured output involves an unacceptable degree of exposure to international fluctuations, the growth of exports might need to be administratively moderated and the rate of growth of the economy might consequently be reduced. This should not happen before 1985. Policy Implications Despite the challenges associated with maintaining high rates of export growth during 1976-90, the conditions outlined in the foregoing discussion give cause for optimism. If the environment for world trade remains favorable and Korea's effectiveness in making use of trading DEVELOPMENT ISSUES IN THE 1980S 83 opportunities remains unchanged, export growth averaging 16 percent a year over the entire period cannot be considered out of reach. Even on a relatively conservative assumption, Korean exports could grow by at least 12 percent a year during 1982-90, after growth of 16 percent a year during 1976-81. The critical importance of flexibility in long-term planning is brought out by the wide difference in export projections for 1990, which in current prices amounts to perhaps as much as $30 billion. For many key export industries the difference between the low and high rates of export growth could mean the difference between doubling and tripling output in a five-year period.6 For heavy indus- tries, where the gestation period between initiating projects and begin- ning production may easily be five years, long-term contingency plan- ning will increasingly be necessary. In the past, most of the unexpected increase in export demand was in light manufactures, where the supply response can be much faster because the gestation period is much shorter. In the 1980s it will be increasingly difficult to achieve rapid export growth by taking advantage of unexpected spurts in demand unless some excess capacity is available. Indeed, if there is reasonable confidence about future prospects of the world economy, there might be a case for deliberately erring on the generous side in planning the size of industrial projects, particularly those in the basic intermediates sector. In order to benefit from the greater flexibility that should be built into the industrial plans, Korea should increase its monitoring and anal- ysis of trends in world trade. Commercial intelligence about emerging international trade flows apparently is quite well organized in Korea. But it needs to be supplemented by greater attention to macroeconomic trends and trade policies, especially those in OECD countries. The general importance of adequate investment allocations for the manufacturing sector also cannot be overstressed. As discussed earlier, the fourth plan constitutes a departure from earlier plans in that the investment require- ments for the manufacturing sector have been conservatively estimated. Because Korea is heavily dependent on external financing for major industrial projects, the danger is that the tightness of the allocation for industrial investment will act as a brake on overall growth. Thus there 6. The two assumptions of the macroeconomic model described in appendix A lead to export projections in 1975 prices of $40 billion and $54 billion in 1990. Corresponding projections in current prices assumed inflation rates of about 5 to 6 percent a year. The low and high growth rates assumed for key industries respectively were 15 percent and 25 percent a year. 84 PART ONE: THE MAIN REPORT is merit in reverting to the earlier practice of making liberal allocations for industry. The considerable scope for further expansion of exports also points to the need for strengthening the institutional arrangements for trading. In Japan general trading companies have contributed to the expansion of intemational trade. In the 1963-72 period, for example, the ten largest trading firms in Japan bandled as much as 50 percent of exports and 60 percent of imports.7 In recent years trading companies in Korea have, with active support from government, begun to contribute increas- ingly to external trade. Although large trading houses offer an oppor- tunity to exploit economies of scale in marketing, Korea must be careful to avoid the harmful effect of competition that might result from the emerging oligopolistic organization of the market. The continuing ex- pansion of exports also carries an obligation to liberalize imports in the 1980s. This obligation is implied by the compulsions of the international political environment and by the welfare costs of excessive import sub- stitution. Care will have to be exercised to ensure that Korea's industrial structure evolves in line with its comparative advantage. Resource Requirements The dominant constraint on Korea's economic growth up to 1976 was the availability of resources. Although the marginal saving rate was quite high at about 25 percent, the low level of national savings in the early 1960s meant continued heavy reliance on foreign savings to finance rapidly growing investment outlays. As discussed earlier, the fourth plan projects a sharp expansion in national savings during 1977- 81. Korean planners may be too optimistic in this regard, but it should be possible to reduce the savings-investment gap during the fourth plan period to about .1 percent of GNP, compared with the average of about 6 percent of GNP in current prices during the third plan period. Thus substantial further progress in overcoming the resource problem is likely to be made during the next few years. Nevertheless the mobilization of resources will continue to require attention during the 1980s if a GNP growth rate of 9 to 10 percent is to be sustained. The estimates made with the use of the macroeconomic model de- scribed in appendix A suggest that gross investment as a proportion of 7. Krause and Sekiguchi, "Japan and the World Economy," p. 392-93. DEVELOPMENT ISSUES IN THE 198os 85 GNP must rise from 26 percent in 1976 to 27.5 percent in 1981 and 31 percent by 1990 in order to sustain an overall growth rate of about 9 percent during 1976-90. This investment will require an average marginal saving rate over the period of 31 percent, assuming that the sur- plus in the current account of the balance of payments will be equal to 1 percent of GNP by 1990. The investment requirements appear to be understated in the model. First, the sectoral capital coefficients are pro- jected to remain at the 1976 level. This assumption might not be reason- able in the long term for one important reason: the tightening of the labor market will probably cause some substitution of labor by capital. Second, the investment in social sectors, which are very capital intensive, is probably understated because of inadequate allowance for shifts in demand. Furthermore, if infrastructure investments are reduced in the fourth plan period to accommodate the scarcity of resources, compen- sating increases will be needed in the future. Finally, it is difficult to forecast investment requirements associated with emerging problems of urban congestion and pollution. Thus the actual investment rate might have to rise to 33 to 35 percent of CNP by 1990, requiring a marginal saving rate of about 35 percent and assuming little net reliance on foreign savings during the 1980s. Problems of domestic resource mobili- zation therefore are not likely to disappear. Fortunately, with the pros- pect of a strong balance of payments, any shortfall in domestic savings could be met by running moderate deficits in the current account of the balance of payments. Thus the growth rate need not suffer. It should nevertheless be stressed that, even with quite small current account deficits in the 1980s, gross borrowing will remain sizable because of requirements to meet amortization payments, extend export credits, and build up foreign exchange reserves. Compulsions for Growth Although Korea can grow at about 9 percent a year in the 1980s, the issue is whether this rate of growth is socially desirable. A slower rate of growth would give Korea time to deal more effectively With the emerging problems of urban congestion and industrial pollution and to consolidate the progress made in achieving structural changes that are vital in establishing the long-term viability of export growth. An implicit assumption in arguments for moderating the rate of growth of the Korean economy is this: basic goals of eliminating unemployment and absolute poverty can be achieved during the 1980s even at growth rates 86 PART ONrE: THE MAIN REPORT Table 3.2. Population of Korea, Projections for 1975-2000 Annual rate Population of increase for (millions of five-year period Year persons) (percent) 1975 35.3 1.76 1980 38.5 1.89 1985 42.4 1.86 1990 46.5 1.64 1995 50.5 1.38 2000 54.1 Source: Lee-Jay Cho, "Projections of the Popu- lation of the Republic of Korea, 1975-2000" (Honolulu: East-West Population Institute, 1976; processed), assumption II. substantially less than 9 percent a year, the rate projected for the fourth plan. Analysis of projections of population changes and economic growth rates suggests that this assumption is mistaken. The labor force is pro- jected to grow at nearly 2.7 percent a year in the 1980s. And despite the creditable record of the Korean economy in distributing the benefits of growth, as many as 3 million Koreans, or 8 percent of the population, were below the minimum income level in 1975. Significant progress toward basic equity goals can be attained only if employment demand grows at more than 3 percent a year, which would require GNP growth of at least 9 percent a year in the 1980s. Demographic trends Because of the increase in the proportion of women of childbearing age, the rate of growth of Korea's population will accelerate slightly until the mid-1980s despite a continuing decline in total fertility rates (table 3.2) 8 The rate of population growth, which declined to an esti- 8. The total fertility rate is the sum of age-specific birth rates over all ages of childbearing and represents the average number of children born to each woman. DEVELOPMENT ISSUES m TIHE 1980S 87 Table 3.3. Age Composition of the Population, 1960-75 and Projections for 1975-2000 (percent) Age group 1960 1966 1975 1990 2000 Under 15 40.6 43.5 38.3 30.9 28.1 15-64 55.7 53.2 58.2 64.3 65.5 Over 64 3.7 3.3 3.9 4.7 6.4 Sources: Tables 4.2 and 4.10. mated 1.7 percent a year in 1976, is expected to rise again to 1.9 percent a year during 1980-85 and then decline steadily to about 1.3 percent a year by the end of the century.9 These projections assume that the total fertility rate will decline from 3.6 in 1976 to about 2.2 in 1990. The decline in total fertility rates is offset, however, by the fact that women born during the baby boom that peaked in 1958 are now enter- ing age groups for which the rates of marriage and fecundity are typi- cally high (see chapter 4). One additional factor might tend to increase the fertility rate. In the past the relatively advanced age at marriage was significant in reducing fertility. But recent figures show that the age at first marriage has begun to decline, possibly because of greater pros- perity. Consequently the rate of population growth could be even higher than is projected in table 3.2 unless the proportion of women practicing contraception or abortion substantially increases. Aware of the implica- tions of these trends, government is strengthening its commitment to the family planning program and attempting to use additional policy instruments to achieve targets for lower fertility. The momentum generated by the baby boom will continue to have a strong influence on the age structure of the population for many decades (see table 3.3). The principal consequence is that the labor force will continue to grow faster than total population, probably until the end of this century. Assuming that the current structure of labor force par- ticipation rates does not change, annual growth of the labor force is estimated at about 2.7 percent a year during 1981-90, compared with the population growth rate of 1.9 percent a year.10 The total labor force 9. Given the assumptions about future fertility patterns discussed in chapter 4, the growth rate of the Korean population is expected to reach zero in about 2075. 10. This projection does not provide for an increase in female labor force participation rates, which currently are relatively low. 88 PART ONE: THE MAIN REPORT in 1990 is projected at 18.9 million persons, compared with about 12.7 million in 1976 and 14.9 million in 1981. As a result of the change in age structure and the decline in fertility rates, the rate of growth of the population under 15 years is projected to be only 0.5 percent a year between 1975 and 2000. This rate has obvious implications for planning the educational system, particularly for training teachers. At the other end of the spectrum, declining mor- tality rates will raise the proportion of the population over 64 years to 6.4 percent in 2000. This prospect raises issues concerning social insur- ance. With the bulk of the population living in urban areas by then, it can be presumed that traditional family ties will have weakened and that a greater proportion of the responsibility for the care of the aged will need to be borne by public agencies. How such a system is best designed and financed is a question of prime social importance. It was noted in chapter 1 that urbanization is well advanced in Korea and that about 60 percent of the population lives in urban areas. Whether the pace of urbanization will continue at the historical rate will primarily depend on the rate of job creation in urban areas and on the job opportunities available in rural areas. Over the 1965-75 period the shift of population from rural to urban areas exceeded the shift in labor deployment, reflecting the facts that participation rates were lower and dependency ratios were higher in urban than in rural areas. This pattern is likely to continue. By 1990 about 80 percent of the population will probably live in urban areas, even though about 30 percent of the labor force will continue to work in the agriculture sector. The continued pressure of urbanization will make it difficult to reduce the population concentration in Seoul and Busan. Even assum- ing that the rate of growth of Seoul's population will decline to 2.5 percent a year-the bulk of the growth being the result of natural in- crease, not of net immigration-the population of Seoul will be about 8.7 million in 1985, compared with 6.9 million in 1975. The shares of Daegu and Daejeon, which lie between Seoul and Busan, and Gwangju in the southwest will probably continue to increase significantly, as will those of smaller cities in the Seoul metropolitan region. On the whole, there is unlikely to be much change in the pattern of population distribution. Two problems of economic management associated with the distribu- tion of population are basic: developing attractive industrial locations outside the present concentrations around Seoul and Busan; finding means of increasing per capita incomes in the southwestern and north- DEVELOPMENT ISSUES IN TlE I 980S 89 eastern regions. These problems are already being seriously tackled, but it is important that a regional development plan be prepared in a long- term framework after adequate analysis of alternatives. The most prom- ising areas for further industrial development are the corridor between Seoul and Busan and the southern coast, areas which enjoy relatively good access to markets and labor. Careful study of the development needs of these two areas, focusing on but not confined to the required growth of infrastructure, is a high priority. A policy framework also needs to be devised for facilitating the growth of small cities; it should cover their administrative and financial structure as well as their indus- trial potential. If measures to manage the urbanization process are ade- quate, it would appear that rapid economic growth should be maintained in order to enable the growing labor force to engage in productive employment. Gro-wth and equity Korea has been able to prevent a significant deterioration in the dis- tribution of income largely because jobs were created at a rate of about 360,000 a year over the 1965-75 period. The same close relation of employment to equity is likely to prevail in the future. Analysis of alter- native simulations of a macroeconomic model confirms that continuance of the current strategy of rapid growth is most likely to achieve signifi- cant improvements in the distribution of income and the reduction of poverty. The macroeconomic model is fully described in appendix A. The model uses an input-output table, national accounting relations, and information on the distribution of income to construct a detailed simula- tion of the Korean economy. The model is primarily long run in its structure; investment is savings-constrained; exports are exogenously specified. There is no optimization process because sectoral investment shares are exogenously specified. Consumption elasticities are functions of per capita income and occupation classes; wages are determined by productivity, the cost of living, and unemployment; prices adjust to changes in the cost of production; nonwage incomes of workers are derived as a residual by deducting the cost of inputs, including wages and indirect taxes, from production value. Aggregate household income for each sector is adjusted-by the number of workers (determined bv sectoral employment elasticities), the number of dependents (deter- 90 PART ONE: THE MAIN REPORT mined by demographic factors and unemployment), and the incidence of taxes-to derive mean per capita disposable income in each of thirty- one classes.-' It is assumed that total income in each class is distributed according to a lognormal distribution with parameters estimated from base period data. That is, it is assumed that the distribution of income within each of the thirty-one income classes remains unchanged over time. Nevertheless the distribution of household income for the entire economy changes because of differences in the rates of growth of aver- age per capita incomes in different sectors. In addition to the rate of unemployment and such measures of rela- tive income inequality as decile income shares and the Gini coefficient, the model derives the number of persons whose per capita income is less than a minimum, W67,000 a year ($138) in 1975 prices.'2 The number below this minimum level is another important indicator of equity. It should be emphasized, however, that the estimate of this number is not exact. The estimation of the minimum income level is based on a number of assumptions, and the derivation of household incomes does not take into account the transfers from rich to poor house- holds, which would tend to ameliorate poverty. Furthermore the esti- mate is naturally sensitive to the employment situation and assumptions about sectors that carry the burden of caring for the unemployed. Therefore the absolute magnitude of the number of those below the minimum income level must be interpreted with caution. The measure nevertheless serves as a useful indicator of the relative effects of different strategies on poverty and of the likely changes in poverty over time. The implications of alternative growth paths for equity can be illus- trated by comparing two simulations: for the high growth path the average rate of GNP growth is 9.4 percent a year during the 1976-90 period; for the low growth path it is 7.8 percent a year (table 3.4). The growth rates associated with these two paths are lower than those achieved by Korea during 1965-75 and higher than those achieved by most other developing countries. Both paths are feasible. They neverthe- less have significantly different implications for equity. As might be 11. The income classes distinguish wage and nonwage sources of income in each of seventeen production sectors, except for two agricultural sectors and the mining sector. 12. This income level was estimated as the requirement to enable the daily intake of 2,210 calories by each Korean using prevailing dietary patterns. DEVELOPMENT ISSUES IN THE I980s 9I Table 3.4. Indicators of Growth and Equity, Projections for 1976-90 Low growth High growth path' path' Indicator 1976 1981 1990 1981 1990 Per capita GrNPb (1975 dollars) 594 814 1,311 840 1,609 Employment (millions of persons) 12.2 14.2 18.2 14.3 18.9 Agriculture 5.5 5.6 5.8 5.6 5.8 Mining and manufacturing 2.5 3.3 4.6 3.3 5.2 Services 4.2 5.3 7.8 5.4 7.9 Unemployment rate (percent) 3.7 4.8 3.7 3.6 Persons below minimum income' (millions) 2.7 2.3 5.1 2.1 1.2 Share of bottom 40 percent 18.4 17.0 13.4 17.0 16.0 Share of bottom 20 percent 6.9 6.3 3.6 6.2 5.6 Share of top 20 percent 46.2 47.9 50.2 48.1 48.8 Ratio of top 20 percent to bottom 20 percent 6.7 7.6 13.9 7.8 8.7 Gini coefficient 0.383 0.406 0.456 0.407 0.419 ... Zero or negligible. Source: Simulations of the macroeconomic model described in appendix A. a. Under the low growth path GNP is projected to grow at 8.4 percent a year during 1976-81 and 7.4 percent a year during 1981-90. Under the high growth path GNP is projected to grow at 9.1 percent a year during 1976-81 and 9.5 per- cent a year during 1981-90. b. The base population for 1976 is estimated to be 35.9 million persons. The population is projected to be 39.3 million in 1981 and 46.5 million in 1990. c. Minimum income is defined as per capita income of W67,000 a year in 1975 prices. It should be noted that the magnitude in each year is sensitive to the assumed minimum income level. expected, equity indicators are significantly better in 1990 under the high growth path than the low growth path, primarily because of the rapid elimination of unemployment. Under the low growth path un- employment rises until about 1985, when the employment situation begins to improve because of the slowdown in the rate of labor force growth from 2.9 percent a year during 1981-86 to 2.3 percent a year during 1986-90. Under the high growth path full employment is reached in 1988, when GNP growth begins to be reflected more in higher 92 PART ONE: THE MAIN REPORT average labor productivity than in higher employment.'3 The increase in labor productivity occurs primarily because of the shift of labor from agriculture and services, where it is underemployed and has lower pro- ductivity, to the manufacturing sector, where the demand for employ- ment is growing fastest. The improvement in the employment situation under the high growth path has a dramatic effect on measures of relative inequality and on the number of persons below the minimum income level. By 1990 the ratio of incomes of the top 20 percent to the bottom 20 percent is projected to rise to 14:1 under the low growth path and 9:1 under the high growth path. Comparison of Gini coefficients and the shares of the different income classes shows that the high growth path leads to a more egalitarian distribution of income. This holds true at corresponding time periods and at equivalent levels of per capita income. The difference in relative income inequality is particularly noticeable in the compari- son of shares of the bottom 20 percent shown in table 3.4. Under the low growth path, even with average cNP growth of 7.8 percent a year, the absolute and relative numbers belovw the minimum income level can be expected to increase substantially. Under the high growth path, on the other hand, the absolute number below the minimum income level declines and their proportion of the total population falls from 8 percent in 1976 to 3 percent in 1990. In general it should be recognized that the relation of output growth to employment growth is not fixed, but is influenced bv the investment strategy followed. In making long-term projections of Korea's growth, however, the model assumes the continuance of 1976 levels of capital intensitv, which are rather low by international standards. If anything, future growth might generate employment demand more slowly than in the past because of increasing capital intensity in the manufacturing sector. It would be simple to assume that agriculture and services will 13. Although full employment is a useful construct, its artificialitv should be recognized. For example, there has been no allowance for frictional unemployment: if an allowance of 2 percent of the labor force is made and if it is assumed that this is met entirely by reducing underemployment in agriculture, agricultural employment would be reduced from about 5.7 million in 1986 to about 5.3 million in 1990, compared with 5.5 million in 1976. It should also be noted that no attempt has been made to estimate the current degree of underemployment. Nor has the response of labor force participation to the employment situation been considered. Briefly, then, the model's projection of full employment might be used simply as an indicator of tightness in the labor market. DEVELOPMENT ISSUES IN THE Ig8os 93 absorb employment faster than the model projects for the 1980s, but relying on these sectors would not contribute much to the growth of per capita incomes because productivity in these sectors is relatively low. Consequently the employment situation would not improve. The per capita incomes of the bottom 40 percent of the population rise at only 2.3 percent a year under the low growth path, compared with 4.5 percent a year under the high growth path. Under both growth paths the indexes of relative inequality tend to rise and are higher in 1990 than in 1976. One reason for this rise is the continuing divergence of average incomes among various sectors of the economy because of different growth rates of productivity. Another reason is the shift in economic activity from the rural sector, where incomes are relatively evenly distributed, to the urban sector where income inequalities are much greater.14 Once full employment is reached, however, the disequalizing effects of this structural shift will be offset by the more rapid growth of productivity and incomes in the low-income.urban occupations because of the reduction of underemploy- ment. Under the high growth path, relative inequality begins to decline in the mid-1980s, whether measured by Gini coefficients or income shares. The Gini coefficient reaches a peak of about 0.45 when per capita income is nearly $1,300 and then declines. At about this stage, the income shares of the bottom 20 percent and 40 percent begin to rise; that of the top 20 percent begins to fall. Under the low growth path, however, the various indexes of relative inequality continue to rise at least until 1990.15 Even if the higher degree of relative inequality could be tolerated, the large increase in the number below the minimum income level is a compelling argument against following the low growth path. To raise their average incomes above the minimum level of W67,000 a year in 1990, the required income supplement would be more than W200 billion. This supplement would be equal to about 0.8 percent of GNP and 4.2 percent of government expenditure in that year. Although these ratios do not appear large, allowance must be made for the immense difficulty of targeting welfare expenditure so that it reaches the poor. The outlays actually needed to achieve the required income supplement .14. For sectoral incomes in 1976 the share of the bottom 40 percent is estimated at 24 percent in agriculture and 17 percent in nonagriculture. 15. If the simulation experiment were carried beyond 1990, it is possible that relative inequality would begin to decline at some point on the low growth path as well. This was not examined. 94 PART ONE: THE MAIN REPORT would probably be some multiple of the figure cited and would be large enough to affect public savings and future growth adversely. The solu- tion of the poverty problem under the low growth path thus would probably be beyond the scope of fiscal transfers. Although the high growth path is distinctly superior on equity grounds, some deficiencies are still apparent in the indicators presented in table 3.4. The first area of concern is the growing relative inequality before the turning point in the mid-1980s. Avoiding this deterioration completely, without adopting drastic measures that would jeopardize the momentum of growth, might be impossible. Because the per capita income of the bottom 40 percent is projected to grow at 4.8 percent a year, such a temporary decline in their share of aggregate income might be acceptable. Because the model assumes that there is no benefit from the tightening of the labor market until full employment is reached, the share of the low-income groups tends to decline further and recover more sharply than would be likely. The reduction in relative inequality would probably start earlier and more gradually than is projected in the model. As the labor market tightens, labor would gradually move away from the low-productivity sectors into manufacturing, thus causing an earlier reduction in relative inequality. This process would begin in the early 1980s under the high growth path, but probably not until nearly 1990 under the low growth path. A second area of concern is this: even under the high growth path, the number of Koreans below the minimum income level might be sizable in 1990. Hence some alternative form of income supplement would be needed. The amount of direct transfers needed to bring them above the minimum income level is not that great-0.1 percent of GNP or 0.5 percent of government expenditure in 1990. But the number of workers affected-more than 2 percent of the urban work force-is significant.16 Thus, while the fiscal burden would be appreciably lower under the high growth path than under the low growth path, it might still be difficult to eliminate the residual problem of poverty by relying only on direct fiscal transfers. The use of additional redistributive devices would probably be necessary. One measure would be to ensure that the poorest segments of society are adequately served with piped water, electricity, paved roads, and other basic services. In addition, public subsidies could ensure the cheap 16. The terms, urban and rural, are loosely used. The model distinguishes only between agricultural and nonagricultural activity. DEVELOPMENT ISSUES IN THE 1980S 95 availability of goods purchased by the poor and possibly raise the effec- tive real eamings of the poor above the poverty level. Food, housing, health services, education, and transport are possible targets for such a subsidized program of poverty alleviation. In order to minimize the leakage of benefits to other income classes, it might be useful to provide a broad variety of goods and services. Lower quality goods could be subsidized to make them affordable to the poor; higher quality goods preferred by the middle and upper income classes would be concurrently provided, but not subsidized. For example, barley, bus transport, and low-income dwellings would be subsidized under such a program; rice, common-use taxi transport, and apartment houses would not be subsi- dized, but their supply would be increased. The main policy conclusion from this discussion of the relation of growth to equity is that the rate of GNP growth will have to be sustained at more than 9 percent a year in order to avoid increases in the number of Koreans below the minimum income level and in the relative in- equality of incomes. This rate of growth would prevent these problems from getting too large. It would also generate additional resources that could be used for selective measures to raise the real incomes of the poor and improve the quality of life. If a growth rate of more than 9 percent a year can be sustained and if special redistributive measures are taken, Korea can look forward to full employment and the elimina- tion of absolute poverty before 1990. The Development of Agriculture The agricultural sector, which includes forestry and fishing, now con- tributes less than one-quarter of GNP and is allocated less than 1 1 percent of total investment. In the long run these ratios are likely to fall even more. The contribution of agriculture to aggregate CNP will probably be about 10 percent in 1990. The sector nevertheless has, and will con- tinue to have, considerable social importance. Currently about 45 per- cent of the labor force is employed in agriculture; the ratio is likely to be about 30 percent even in 1990, assuming the economy grows at more than 9 percent a year. There are four principal issues in the agricultural sector: the extent to which domestic production can meet the growing demand for food- grains; the economic efficiency of investment required to achieve further increases in grain production; the availability of labor for farm activi- ties; and the likely trends of productivity and per capita incomes in 96 PART ONE: THE MAIN REPORT Table 3.5. Food Demand, Projections for 1975-90 Per capita Total consumption consumption Annual 1975 1990 1975 1990 growth rate Food item (kilograms) (millions of tons) (percent) Rice 113 116 4.3 5.9 2.1 Barley 53 43 2.1 2.3 0.6 Wheat 48 54 1.7 2.5 2.6 Fruit and vegetables 90 122 3.7 6.6 3.9 Livestock products 44 81 1.7 4.1 6.0 Note: Per capita consumption is presented in net weight to focus on the as- sumed change in food habits. Total consumption is presented in gross tons, includ- ing allowances for seed and wastage, to enable comparison with projected production. For more details see chapter 7. Source: World Bank calculations. agriculture, particularly in relation to the growth of urban incomes. The technical efficiency of agricultural practices is not a major issue in Korea. As noted in chapter 1, rice yields already are among the highest in Asia; with the introduction of new high-yielding varieties, average yields are continuing to increase. In general the research and extension network performs well, and Korean farmers make good use of modern inputs and know-how. In addition to the natural growth in demand arising from the growth of population and incomes, a significant shift is likely in the structure of demand, reflecting changing food habits and the demographic struc- ture. On the basis of various studies of the likely evolution of patterns of food consumption, including a comparison with trends over the past twenty years in Japan, the probable magnitude and composition of food demand in 1990 have been projected. A summary of the principal com- ponents of demand is presented in table 3.5; details are available in chapter 7.17 Per capita rice consumption is projected to rise until the late 1970s, mostly because of the relaxation of administrative restraints on rice consumption, and then decline as demand shifts to fruits, vege- 17. These projections are primarily based on Thodey's study of food and nutrition and subsequent work by the Korean Agricultural Sector Study (KAss) team. Alan R. Thodey, "Food and Nutrition in Korea, 1965-74," SAss Special Report 11 (Seoul: National Agricultural Economics Research Institute; Last Lansing: Michigan State University; n.d.). DEVELOPMENT ISSUES IN THE I980s 97 tables, and livestock products. This shift would significantly increase the intake of high-quality protein. Taking population growth and the changes in demographic composition into account, total consumption is projected to change by 1990 in the following manner: compared with 1975, rice demand is projected to increase by 36 percent, barley by 10 percent, wheat by 48 percent, beef by 130 percent, milk and eggs by 150 percent, and fruit and vegetables by 78 percent. The demand for fruit and vegetables can be met from domestic pro- duction. The amount of land required for vegetables is not so great and for fruit is generally not competitive with grain production. The con- straints on grain production, discussed in chapter 2, are more severe. To recapitulate, less than one-quarter of Korea's land area is cultivable; of the present land base of about 2.2 million hectares, the net loss each year to nonagricultural uses is about 13,000 hectares, or 0.6 percent of the total; the severity of winters restricts multiple-cropping, and the extent of double-cropping has recently been declining, primarily because of the high cost of labor during the extraordinarily sharp seasonal peaks in labor demand. Despite these constraints, it would be feasible and economically justi- fiable to maintain self-sufficiency in rice. The import of wheat and probably of barley will continue to be necessary. If reasonable assump- tions are applied, the average rice yield can be expected to reach nearly 5 tons a hectare by 1990. If the land area under rice is not significantly reduced, self-sufficiency in rice can be maintained. The availability of labor is unlikely to pose a constraint. As noted, the number of persons available for employment in agriculture is unlikely to decline signifi- cantly before 1985; the projected level for 1990 is only 3 to 4 percent lower than the 1976 level, even assuming a relatively high rate of eco- nomic growth (see footnote 13). Beyond 1990, however, the availability of labor might well pose a constraint. That is sufficiently far in the future to allow a reassessment of the potential for mechanization and other agricultural investment at the end of the fourth plan period. As subsequently discussed, labor scarcity might nevertheless affect the pro- duction of barley and wheat. The analysis supporting the judgment on rice self-sufficiency is given in chapter 7 and briefly summarized here. In essence, examination of the costs and yield benefits of recent investment in land development in Korea suggests that new investment starts of slightly more than $100 million a year in 1975 prices, in addition to the investment necessary for the completion of ongoing projects, would be economically justified. This amount is considerably less than the allocation for land develop- 98 PART ONE: THE MAIN REPORT Table 3.6. Cultivable Land under Two Levels of Investment, Projections for 1975-90 (thousands of hectares) 1990 No $120 million additional additional Type of land 1975 investmenta investmentb Paddy land 1,277 1,282 1,309 Upland 963 975 1,023 Total 2,240 2,257 2,332 Source: World Bank calculations. a. Except for the completion of projects already under way. b. Annual investment in addition to the investment necessary for the completion of projects already under way. ment in the fourth plan-W600 billion, or an average of $248 million a year-even allowing for the fact that the allocation includes provision for the completion of ongoing projects and for large-scale integrated development schemes. The investments would cover the conversion of slope land, the reclamation of tidal land, the improvement of irrigation, and the consolidation of paddy holdings to improve drainage and facili- tate mechanization. If this level of investment were maintained until 1984, the additional land improved would be about 237,000 hectares over and above the area to be improved under ongoing investments alone. Allowing for the loss of land to nonagricultural uses, the net area available for cultivation would decline slightly (table 3.6). The projections of production and the implications for import re- quirements in 1990 are shown in table 3.7. Given the improvement of rice yields to an estimated 4.3 tons a hectare in 1976, it would seem that the high-yield assumption of 5 tons a hectare in 1990 is not opti- mistic. Hence, there might even be a small surplus in rice production over demand. The situation is different for the upland crops. Because of possibilities for their substitution in both production and consump- tion, estimates for wheat and barley are shown jointly. A substantial volume of these grains will continue to be imported.18 The magnitude of the deficit will depend on whether double-cropping stays at 40 per- cent, continues its downward trend to 20 percent in 1990, or can be 18. For 1973-75 annual average imports of wheat were 1.5 million tons; those of barley 509,000 tons. DEVELOPMENT ISSUES IN THE I 980S 99 Table 3.7. Foodgrain Production, Demand, and Import Requirements, Projections for 1990 (millions of tons) Import' Item Production Demand requirement Rice - 5.9 - High yield assumption 6.3 - 0.4 Low yield assumption 5.7 - -0.2 Wheat and barley - 4.8 - Twenty percent double-cropping 0.8 - 4.0 Forty percent double-cropping 1.7 - 3.1 Sixty percent double-cropping 2.9 - 2.0 - Not applicable. Note: Production figures assume new investment of $120 million a year in addition to the investment necessary for the completion of projects already under way. The high yield assumption for 1990 is 5 tons of rice a hectare; the low yield assumption 4.5 tons a hectare. Assumptions about demand are described in the note to table 7.7. Source: World Bank calculations. a. Imports of feedgrains are not included. They are projected to reach 3 to 4 million tons by 1990. raised. Even assuming that imports will have to be around 4 million tons in 1990, the foreign exchange cost is estimated at about $800 million in 1975 prices, compared with wheat imports of $294 million in 1975. Compared with the total import bill in 1990, which is estimated at about $40 billion, the direct foreign exchange requirement for the higher level of foodgrain imports is small. Nevertheless the degree of import dependence will increase to 38 percent, compared with about 25 percent in 1976. In assessing the possible level of imports of wheat and barley, two additional factors should be considered. First, the demand for barley might drop faster than has been projected. In Japan the per capita consumption of barley dropped from 17 kilograms in 1954 to 1 kilogram in 1972. Under conditions of rapidly growing incomes, the changes in food habits are difficult to predict. If the supply of rice is increased, as is projected in table 3.7, people might shift from barley to rice. Second, the current research effort to develop suitable strains of wheat might succeed in the longer run, although the extent of success pre- dicted for the fourth plan period probably is unrealistic. If local strains of winter wheat can be developed that do not interfere with the rice 0oo PART ONE: THE MAIN REPORT crop and have higher yields than can now be attained with barley, the supply of these grains will markedly change. Cropping intensity could increase and, as demand and production shift more rapidly from barley to wheat, rural incomes could also benefit. Although this analysis of foodgrain production and demand has been based on rough assumptions, it strongly suggests that the level of invest- ment for land development projected in the fourth plan is substantially higher than is either required for assuring rice self-sufficiency or justified bv considerations of economic costs and benefits. Therefore the major requirements are to spell out the long-term objectives of the agricultural sector and to prepare a development plan to meet them. Because of the long gestation period between investment in land improvement and the achievement of higher yields, such a plan must have a perspective of 10 to 15 years. A long-term plan for agricultural development should address four important issues. First, to what extent can recent improvements in rice yields be taken as a firn basis for projections of rice production? With the help of supporting investments to consolidate these gains, and further extension efforts to reduce the gap between experimental and actual farm vields, continued improvement in rice yields is possible. This nevertheless will need persistent attention. Second, what should be done to stem the decline in double-cropping and, if possible, reverse the trend? The decline in double-cropping in recent years probably is the result of the high cost of labor during peak seasons and the low profit on winter crops. Given Korea's climate, farmers who plant winter crops risk reducing yields from the main summer crop if adequate labor is not available for harvesting and plant- ing. The summer crop of rice is much more profitable than the winter crop of barley, and farmers tend to be unwilling to run the risk of sacrificing the summer crop. Because double-cropping has such a large impact on foodgrain self-sufficiency, a variety of steps should be taken to improve this situation. Perhaps the most important step would be to increase the mechanization of crucial operations in the peak season. To do this would involve shifting the emphasis from power tillers to other farm machinery. Some gains in reducing the intensity of labor demand in peak seasons could also be achieved if current research efforts succeed in developing rice and barley strains that have a shorter growing season. Double-cropping would also increase if the winter crop were made more profitable, but the many implications of such profitability for price and subsidy policies should be fully explored. The procurement DfVELOPMENT ISSUES IN THE 1980S 101 price for barley already is high in relation to international grain prices. And the large-scale subsidization of grain can cause severe fiscal strains which might interfere with investment financing and with monetary and price stability. Third, how is the growing demand for livestock products to be met) Although Korea can easily expand the production of chicken and pork, expanding the production of dairy and beef cattle runs into economic difficulties. The evidence suggests that relying on imports of dried milk and beef products might be more economical than attempting to develop a dairy and livestock industry based on imported feeds. The alternatives should be analyzed, and a policy should be developed for the long-term growth of the domestic livestock industry. In doing so, the potential for poultry and pork should be examined separately, because it might be possible to exploit some opportunities for export. Fourth, what are the implications of the projected trends in agricul- ture for income differences between the rural and urban sectors? Al- though the connection is imprecise between intersectoral income differ- ences and broader issues of the distribution of income and the allevia- tion of poverty, this question generates great interest in Korea. A sig- nificant proportion of the population lives in rural areas, and average rural incomes are lower than average urban incomes. In addition, gov- ernment has considerable influence over the rural-urban terms of trade which, in turn, have a major impact on rural-urban income differences. As noted in chapter 1, these differences are significant, although prob- ably no more so than is common in developing countries, and govern- ment has made a strong effort to prevent their widening. These differ- ences, and consequently the efforts of government, will probably continue to be significant at least until the mid-1980s. Until underemployment in agriculture is greatly reduced, it will be difficult to create conditions that enable rural incomes to increase faster than, or even as fast as, urban incomes. Nevertheless agricultural incomes will probably increase faster than they have in the past. As the agricultural sector becomes smaller in relation to the urban sector, manufacturing wages will have a stronger impact on agricultural wages, forcing changes in technology or mechanization to increase labor pro- ductivity and incomes. Rural incomes will probably benefit from the shift in the composition of demand toward crops that have a higher value-added component-for example, from cereals to fruit and vege- tables and from barley to wheat and rice. In addition, exports of agri- cultural products could expand if greater efforts were made to strengthen 102 PART ONE: THE MAIN REPORT the organization of marketing, particularly by developing links with other countries. Some idea of likely trends in rural-urban income differences can be obtained from the projections of the macro model (table 3.8). Assuming that value added in the agricultural sector grows at an average rate of 3.4 percent a year during 1976.90, and applying the high growth path as stated in table 3.4, the model projects that the per capita disposable income of households will grow at 3.9 percent a year in agriculture, 5.2 percent a year in other sectors, and 5.6 percent a year for the economy as a whole.'9 Because the distribution of income is less skewed in agriculture than in other sectors, average incomes of the bottom 40 percent of income receivers are estimated to be only slightly lower in agriculture. The income gap does not increase significantly, even though the differences for the sectors as a whole are projected to widen fairly steadily. The widening of sectoral differences simply reflects different rates of productivity growth in the various sectors. These projections perhaps are unduly pessimistic. They assume that the terms of trade will become less favorable for agriculture because no allowance is made for explicit intervention in the rural-urban terms of trade.20 It should also be noted that as nonfarm incomes gain in importance in the rural sector, income differences will be less than those projected in table 3.8. Incomes in this table are based on agricultural incomes alone; they omit about 20 percent of rural household incomes on average. For these rea- sons the simulations suggest that the problem of rural-urban income differences is unlikely to be severe in the long run if steps continue to be taken to raise agricultural productivity and provide opportunities for off-farm income. The base of comparison could be broadened beyond the narrow con- cepts of income by including the availability of such essential services as health, piped water, electricity, and transport. The fourth plan's pro- visions for significant improvements in essential services should be sus- tained in the longer run as well. Because of these measures, intervention to improve the terms of trade significantly in favor of agriculture might become unnecessary. Such intervention could, moreover, raise the risk 19. As noted earlier, the macroeconomic model does not distinguish between rural and urban sectors, but only between agricultural and nonagricultural sectors. 20. The model specifies a cost-plus pricing mechanism in each sector. Computed as the ratio of the price of grains to the cost of rural living, the index of the terms of trade falls from 100 in 1976 to 97.6 in 1990. DEVELOPMENT ISSUES IN THE 1980S 103 Table 3.8. Per Capita Disposable Income of Households, by Sector, Projections for 1976-90 (thousands of won in 1975 prices) Bottom 40 percent Total population Annual Annual growth growth rate rate Sector 1976 1990 (percent) 1976 1990 (percent) Agriculture 93 159 3.9 139 237 3.9 Nonagriculture 100 181 4.4 240 486 5.2 All sectors 95 176 4.5 207 449 5.6 Ratio of non- agriculture to agriculture 1.1 1.1 - 1.7 2.0 - - Not applicable. Note: These figures correspond to the high growth path defined in table 3.4. Souirce: Simulations of the macroeconomic model described in appendix A. of establishing an inefficient agricultural sector which, in turn, might become a burden to the rest of the economy in the long run. Government's Changing Role in Investment Programming By 1990 investment in power, transport, and communication will probably be more than $2 billion a year in 1975 prices; investment in the metal and chemical industries, about $2.5 billion. Government will have either direct responsibility for these investments, which require detailed evaluation and planning in a long-term framework, or a strong influence on them. Yet the administrative machinery of the public sector is not now geared to this responsibility. Given the long gestation of investment in most of these sectors, the necessary groundwork is over- due. The most urgent need is the forecasting of demand, taking inter- industry linkages into account. Although it hardly matters whether the forecasts are generated by government or private agencies, it is likely that government will have to take the initiative by forecasting infra- structure demand and making the bases of the forecasts widely available to industry. Considerable improvement is needed in the techniques of demand forecasting in Korea. Most forecasts are still based on elementary ex- 104 PART ONE: THE MAIN REPORT trapolative techniques or on simple relations with macroeconomic aggre- gates. Relatively little attention is paid to implications of the changing structure of intermediate demand. In the preparation of the fourth plan, EiPB used a 53-sector input-output model to a limited extent to check the consistency of investment allocations. As the sectors of the economy become more interdependent and as individual investments get larger, it will be necessary to employ more sophisticated techniques of investment programming to determine the optimum scale, phasing, location, and production mix of proposed investments. Some of these complex matters have already become important in such sectors as energy and transport, which are still being analyzed with relatively elementary techniques. Monitoring the consistency of demand forecasts will be one important task of EPB's new unit for project appraisal. The purpose of this unit is to evaluate economic benefits and costs of invest- ment projects in which government is involved either directly or through the provision of incentives. The establishment of such a unit should improve the efficiency of allocating investment funds, particularly in infrastructure and basic intermediates where direct govemment partici- pation is intensive. In the past, government has played a rather important direct role in most manufacturing subsectors. In the planned subsectors, particularly steel, fertilizer, chemicals, and shipbuilding, government identified major projects and ensured their implementation, sometimes to the point of initiating them through the agency of a public enterprise. Even in the unplanned subsectors, such as textiles, there has been active govern- ment participation from time to time in establishing programs for modernization and capacity expansion at the enterprise level. Govern- ment currently is playing a major direct role in implementing the fourth plan's goals in the machinery and electronics industries. In the longer term the need for government promotion of specific projects will de- cline. Industrial growth would probably be healthier if private entre- preneurs sought out profitable opportunities themselves and took the attendant risks. Mechanical engineering, electronics, and light industry will be too diverse for government to have a useful direct role. Even in the basic intermediates, private entrepreneurs will have the necessary management ability and financial strength to take the lead. Govemment will nevertheless continue to have an important role in providing infrastructure and promoting structural change in the manu- facturing sector. Discharging the first role is essentially a matter of investment programming, a topic already touched upon. The second role is partly a matter of investment programming, insofar as govern- DEVELOPMENT ISSUES IN THIE 1980s 105 ment monitors the capacity of the basic intermediate sector. More im- portant, promoting structural change will require the development of new institutions and a somewhat different relationship between govern- ment and industry. This is most important for the transfer of technology. The growth sectors in Korean manufacturing-machinery and elec- tronics-are characterized by rapidly changing technology. As Korea's wage costs rise in relation to other developing countries, there will have to be a shift toward production requiring sophisticated technology that is competitive with the developed countries. Government will have to pioneer institutions that advise Korean entrepreneurs on the range of available technology and the most advantageous means of acquiring this technology through partnership or other ventures with foreign enter- prises. In two other important areas, manpower development and credit availability, government already is active. But some changes will prob- ably be needed. In the past and to a lesser extent during the fourth plan period, government has been acting as a financial intermediary between foreign lenders and domestic industry. Korea's investment has relied on foreign borrowing, the bulk of which is guaranteed by govem- ment either directly or through the Korea Exchange Bank. By control- ling access to foreign loans, government has been able to exercise direct influence on investment allocation in private industry. The National Investment Fund, established for domestic long-term credits, also is under direct government control. As Korea's balance of payments im- proves, the share of foreign finance in investment will decline. Korean industry should then be able to tap private sources of funds abroad without government guarantees. The development of the securities market would also diminish the need for government financial inter- mediation. Both these developments will reduce government's capability for intervention in credit allocation, which will have to be left more to the operation of market forces. A similar trend has already emerged in manpower development. Private employers are taking more responsi- bility for meeting specific training needs, while government provides basic education. The main point of this discussion is that market signals will inevitably become more important in the manufacturing sector. As the economy becomes more sophisticated, decisionmaking will have to be more de- centralized, more automatic, and largely at the enterprise level. Govern- ment has already recognized this development and is now reforming the regime of import tariffs. A revision of the level and structure of tariffs has been proposed which, if implemented, would provide pro- io6 PART ONE: THE MAIN REPORT tection at two levels-a low level for established production, and a higher level for import-substituting production. Government has also recog- nized that the effect of reforming the system of tariff protection will be limited unless the system of import controls is simultaneously liberal- ized. Thus the nature of government intervention is changing, and government's role is becoming that of an indicative planner and catalyst. Through improved sector analysis and economic evaluation of major projects, government will nevertheless retain some broad influence over the direction of investment. In addition, government will continue to have a role in some crucial areas affecting industrial development, particularly the transfer of technology. PART TWO Human Resources Chapter 4 Catherine S. Pierce Population Growth, Programs, and Policies As A RESULT OF MODERNIZATION and a vigorous family planning pro- gram, Korea was able to reduce its annual rate of population growth from approximately 3 percent in 1955-60 to about 1.7 percent in 1975. The crude birth rate fell from about forty-five births per thousand popu- lation in 1955-60 to about twenty-nine in 1970-75. The total fertility rate fell from 6.3 children in 1960 to 3.6 in 1975. Other demographic indicators also improved. The expectation of life rose from 53 years in 1960 to about 66 years in 1975. Infant mortality, according to official esti- mates, declined from about eighty-three deaths per thousand live births in 1957 to about thirty-eight in 1975. Compared with other developing countries, such as Mexico and Brazil, Korea made considerable progress in reducing population growth. There nevertheless has been increasing pressure on available land. The country's population rose from 21.5 million in 1955 to 35 million in 1975 (table 4.1). With a population density of 358 persons per square kilometer in 1975, Korea had become one of the most densely settled countries in Asia. Japan's population density in 1975 was 294 persons per square kilometer. Korea's population is projected to grow at an average annual rate of about 1.7 percent during the remainder of this century and reach 54 million in the year 2000. The reduction of the fertility rate has already contributed to improved economic conditions in Korea. Average per capita income rose faster than it would have under a higher rate of population growth; the pro- portion of the population under working age and dependent on society's productive members has declined. The reduction of the fertility rate probably helped to raise enrollment ratios as well. The added burden of school fees, which almost certainly would have been higher if a greater proportion of resources had been applied to primary education, would have deterred many young persons from enrolling in secondary and higher education. Smaller families may have changed the patterns o09 IIO PART TWO: HUMAN RESOURCES Table 4.1. Population of Korea, 1949-75 Population. Annual growth rate (thousands for intercensal Year of persons) period (percent) 1949 (May 1) 20,167 1.01 1955 (September 1) 21,502 2.86 1960 (December 1) 24,989 2.64 1966 (October 1) 29,160 1.87 1970 (October 1) 31,435 1.97 1975 (October 1) 34,681 Sources: Economic Planning Board (EPB), 1970 Population and Housing Census Report, Complete Enumeration (Seoul, 1973) and earlier reports of the census; EPB, Preliinary Re- lease of the Census of Population, 1975 (Seoul, 1976). a. Population figures are those given in the censuses; no adjustment has been made for underenumneration. Census fig- ures exclude foreigners. of household expenditure and saving. If it is assumed that food expendi- ture is sensitive to population size and that incremental food demand would have been met by imports, slower population growth relieved pressure on the balance of payments. Smaller families may also have made the housing shortage more tolerable. But the most important economic benefits from the decline in fertility will begin to take effect as the growth of the labor force slows down, easing the employment situation and facilitating a further rise in real wages. Demographic Trends The proportion of population under 15 years of age rose from 41.2 percent in 1955 to its peak of 43.5 percent in 1966 and then fell to 38.3 percent in 1975 (table 4.2). The fall after 1966 was caused by a decline in fertility which can be attributed to a number of conditions: among them, the government's vigorous family planning program, the rising age at marriage, and the rapid pace of urbanization. Nevertheless Korea's age structure in 1970-with respect to such measures as median POPULATION GROWTH, PROGRAMS, AND POLICIES I I I Table 4.2. Age Composition of the Population, 1955-70 (percent) Age group 1955 1960 1966 1970 0-4 15.7 17.5 15.4 13.7 5-14 25.5 25.3 28.1 28.4 15-49 47.2 45.8 45.4 46.5 15-64 55.5 53.9 53.2 54.6 65 and over 3.3 3.3 3.3 3.3 Total (thousands of persons) 21,502a 24,989 29,160 31,435 Sources: Tai Hwan Kwon and others, The Population of Korea (Seoul: Seoul National University, Population and Development Studies Center, 1975), appendix I; EPB, 1970 Census Report. a. Includes foreigners. age (18) and the proportion of population under 15 (42.1 percent) and over 64 (3.3 percent)-was still young and typical of developing countries (table 4.3).1 Although the rate of population growth has slowed down and will Table 4.3. Selected Measures of Changes in Age Composition, 1955-70 (percent) Measure 1955 1960 1966 1970 Persons under 15 years 41.2 42.8 43.5 42.1 Persons 15-64 years 55.4 53.9 53.2 54.6 Persons over 64 years 3.3 3.3 3.3 3.3 Age dependency ratio' 80.4 85.5 87.9 83.3 Median age (years) 18.9 18.7 17.5 18.0 Sources: EPB, 1970 Census Report and earlier reports of the census. a. The ratio of children under 15 years and persons over 64 years to the population of working age (15-64 years). 1. Comparable figures for India were 18.8 years, 42 percent, and 3.4 percent respectively; those for Brazil 18.4 years, 42 percent, and 3.3 percent respectively. Comparable figures for Japan in 1965 were 29 years, 25.6 percent, and 6.3 percent; those for the United States in 1970, 28 years, 28.5 percent, and 9.9 percent. United Nations, Demographic Yearbook, 1973 (New York, 1974), table 6. 112 PART TWO: HUMAN RESOURCES continue to do so in the future, growth will take place for several decades because of the momentum built into the age structure. lhe fertility rate is expected to decline further, but the absolute number of births will increase because of the growing number of women in the childbearing ages. Their proportion rose from 46.1 percent in 1960 to 50 percent in 1975 and will continue to rise gradually to the end of the century, when it is expected to reach 53.1 percent. The rise in the proportion of women in the younger reproductive ages is particularly striking after 1970, and this will make it difficult to achieve a reduction in the number of births. Fertility and mortality Fertility rose sharply in the years after the Korean War. But unlike the fifteen-year baby boom in the United States after the Second World War, the upsurge in Korea lasted only a few years. The birth rate remained high until the early 1960s and then declined rapidly. Between 1960-65 and 1970-75 the crude birth rate-the annual number of live births per thousand population-fell from 42 to 29 (table 4.4). This decline has been analyzed to determine the relative importance of such contributory factors as age-sex structure, proportion of women married, and marital fertility for rural and urban areas. In all cases the decline in marital fertility-the average number of children born to a married Table 4.4. Vital Rates and Expectation of Life at Birth, Estimates for 1950-75 Expectation of Crude Crude life at birth birth rate death rate Period (per thousand) (per thousand) Male Female Total 1950-55 40 33 46.0 49.0 47.5 1955-60 45 16 51.1 54.2 52.6 1960-65 42 15 53.6 56.9 55.2 1965-70 32 13 56.0 59.4 57.7 1970-75 29 9 58.8 62.5 60.6 Note: Figures for 1975 are based on projections. Sources: Kwon, Population of Korea; United Nations Population Division, Selected World Demographic Indicators by Countries, 1950-2000 (New York, 1975), medium variant. POPULATION GROWTH, PROGRMS, AND POLICIES 113 Table 4.5. Age-specific Fertility Rates, 1950-70 (children per thousand women) Age group 1950-55 1955-60 1960-65 1965-70 15-19 45 38 20 12 20-24 289 308 255 180 25-29 287 335 351 309 30-34 233 270 274 223 35-39 168 194 189 134 40-44 83 96 92 59 45-49 15 18 17 10 Total fertility rate, (per woman) 5.60 6.29 5.99 4.64 Source: Kwon, Population of Korea. a. The total fertility rate is the sum of the age-specific fertility rates multiplied by the five-year interval. woman-was found to be the most significant. It accounted for 60 per- cent of the decline in the crude birth rate; changes in the age structure and the age at marriage accounted for 40 percent.2 The crude birth rate is a measure of the incidence of births over a short period. It thus yields few insights about completed fertility and different patterns of childbearing over the entire age span. Trends in age-specific fertility rates and the total fertility rate offer more informa- tion about changing fertility patterns over extended periods (table 4.5). The age-specific fertility rate for a given year is the annual number of births per thousand women in a five-year age group. The total fertility rate is the average number of children a woman would have if she sur- vived the reproductive years and bore children according to a given set of age-specific fertility rates. In Korea the total fertility rate has fallen from 6.3 in 1960-65 to about 3.6 in 1975. The reduced fertility rates for younger age groups are in part attributable to the postponement of marriage. In 1940 the average age at first marriage was 21.8 for males and 17.8 for females; by 1970 the comparable figures were 27.2 for males and 23.3 for females.3 As would be expected, the rise was greater for 2. Lee-Jay Cho, The Demographic Situation in the Republic of Korea, Papers of the East-West Population Institute, no. 29 (Honolulu, 1973), pp. 16-17. 3. Tai Hwan Kwon and others, The Population of Korea (Seoul: Seoul National University, Population and Development Studies Center, 1975), p. 46. I I4 PART TWO: HUMAN RESOURCES urban women than rural women. Given the already advanced ages at first marriage today, it is doubtful that further increases vill occur. In fact, some evidence suggests that the age at marriage may be dropping. Age-specific marital fertility-fertility rates computed for married fe- males only-has declined for all ages from the peak levels in the late 1950s and early 1960s (table 4.6). The decline for older age groups was substantially greater than that observed for younger age groups. This suggests that older married women, having reached their desired family size, are making use of contraception, abortion, and sterilization. Marital fertility of the younger age groups, although declining, has remained high. Korea's mortality transition is considered to have started in 1955, when mortality began a sharp decline because of the widespread use of antibiotics, the improvement in sanitation, and the extension of medical services. Between 1955 and 1970 the crude death rate-the annual number of deaths per thousand population-fell from an esti- mated 33 to about 9, and life expectancy at birth rose from 53 years to 61 years. The infant mortality rate-the annual number of deaths of children under one year per thousand live births-is estimated by the Ministry of Health and Social Services (MOHSA) to have been about 38 in 1975, but several surveys indicate that this estimate is low. Matemal mortality continues to be high, particularly in rural areas. Although it is presumed to be declining, the large number of un- attended deliveries makes it difficult to determine with accuracy the extent of maternal mortality. In addition to the desirability of reducing Table 4.6. Age-specific Marital Fertility Rates, 1950-70 (children per thousand married women) Age group 1950-55 1955-60 1960-65 1965-70 15-19 312 357 356 350 20-24 385 440 443 394 25-29 320 367 383 346 30-34 260 298 295 237 35-39 193 221 212 148 40-44 102 117 111 71 45-49 21 24 22 13 Marital fertility rate (per woman) 7.97 9.12 9.11 7.79 Source. Kwon, Populatin of Korea. POPULATION GROWTH, PROGRAMS, AND POLICIES I 1 5 infant mortality as a goal in itself, it can be argued that improvements in the survival rates of infants and young children would reduce fertility once parents realized they would need fewer births in order to assure the survival of their preferred number of children. Geographic distribution Since 1945 political and economic forces have in large part deter- mined the pattem of migration and the characteristics of migrants. Immediately after the Second World War a large number of Koreans returned from Japan and Manchuria. During and shortly after the Korean War the movement between North and South Korea was con- siderable. By the 1960s the motivation for improved economic oppor- tunity resulted in heavy rural-to-urban migration and accelerated the rate of urbanization. The proportion of population living in urban areas rose from 28 percent in 1960 to 48 percent in 1975.4 Migration to Seoul and Busan, the two special cities, was heavy and sustained, particularly during 1966-70. By 1975 they contained 27 percent of the national population. Of the provinces, Gyeonggi had the highest rate of growth in each intercensal period (table 4.7). As the province surrounding Seoul, it received the spillover of growth from Seoul's metropolitan area. During 1970-75 the growth rate for Seoul declined, but that for Gyeonggi markedly increased. In all provinces urban growth was faster than rural growth. Census data for the origin and destination of migrants reveal a substantial amount of migration to rural areas. This can be largely explained by the census definition, which classifies any area having fewer than 50,000 persons as rural. Thus many satellite towns of Seoul and the mining towns in Gangweon province are classified as rural. Population increases in urban areas have generally been a function of four factors: changes in administrative status (the number of cities rose from 27 in 1960 to 32 in 1966), the incorporation of areas peripheral to cities, net migration, and natural increase. Between 1960 and 1966 4. The Korean Census defines as urban a city (shi) having a population of 50,000 or more. Census data are given only for urban (shi) and rural (gun) areas. Towns (eup) with population ranging from 20,000 to 50,000 are classified as rural; these areas would be more accurately described as urban because of the life-style and economic activity of their inhabitants. This broad definition of rural residence often masks the true dimensions of the urban sector and complicates analysis of migration data. I I6 PART TWO: HUMAN RESOURCES Table 4.7. Geographic Distributian of Population, by Province, 1955-75 (percent) Percentage of national population Prorince 1955 1960 1966 1970 1975 Seoul' 7.3 9.8 13.9 17.6 19.8 Busan' 4.9 4.7 4.9 6.0 7.1 Gyeong gi 11.0 11.0 10.6 10.7 11.6 Gang weon 6.9 6.6 6.3 5.9 5.4 Chung bug 5.5 5.5 5.3 4.7 4.4 Chung nam 10.3 10.1 10.0 9.1 8.5 Jeon bug 9.9 9.6 8.6 7.7 7.1 Jeon nam 14.5 14.2 13.9 12.7 11.5 Gyeong bug 15.6 15.2 15.3 14.5 14.0 Gyeong nam 12.6 12.1 10.9 9.9 9.5 Jeju 1.2 1.1 1.2 1.2 1.2 Total 100.0 100.0 100.0 100.0 100.0 Total (thousands of persons) 21,502 24,989 29,160 31,435 34,681 Sources: EPB, Korea Statistical Yearbook, 1975; EPB, Preliminary Release of the Censuts, 1975. a. Seoul and Busan are special cities having provincial status. Busan became a special city in 1966; before that it was part of Gyeong nam province. The totals given for Busan in 1955 and 1960 were subtracted from totals for Gyeong nam province in those years. net migration and natural increase were about equally responsible for the increase in the urban population; between 1966 and 1970, how- ever, net migration became dominant, accounting for 73.5 percent of urban growth (table 4.8). A number of studies have found that improvement in economic status through job-seeking or job-acceptance is the dominant reason cited by migrants for moving. In one study 31 percent of migrants gave job- seeking as the reason for moving; 13.7 percent cited occupational reloca- tion or job acceptance.5 In a similar study census data for 1970 were used to support the hypothesis that migration reflects a decision of each individual to maximize the expected rate of return to investment in human capital.6 By using the county as the unit of observation, it was 5. Jong-Joo Yoon, "A Study of Fertility and Out-Migration in a Rural Area" (n.p., n.d.; processed). 6. Kong-Kyun Ro, "Macro Analysis of the Economic Determinants of Rural- Urban Migration in Korea" (n.p., n.d.; processed). POPULATION GROWTH, PROGRAMS, AND POLICIES II17 Table 4.8. Net Migration to Urban Areas, by Province, 1960-70 1960-66 1960-70 Net Increase Net Increase migra- trom migra- from tion net tion net to migra- to migra- urban Ratea tionb urban Rates tion' areas (per- (per- areas (per- (per- Province (persons) cent) cent) (persons) cent) cent) Seoul 784,203 20.7 65.7 1,400,564 25.4 80.9 Busan 42,268 3.0 19.3 321,624 17.1 71.4 Gyeong gi 72,369 9.9 43.3 119,213 13.1 66.5 Gang weon 22,550 6.8 33.0 22,690 5.9 45.8 Chung bug 295 0.1 1.1 12,557 5.4 45.7 Chung nam 22,492 5.8 30.8 73,157 14.9 69.2 Jeon bug 3,848 1.0 6.8 26,972 5.8 45.1 Jeon nam 66,974 9.0 41.2 78,387 8.9 58.4 Gyeong bug 115,307 10.3 46.5 181,113 13.0 66.3 Gyeong nam -5,722 -1.0 -8.6 67,224 10.0 58.6 Jeju 7,128 8.2 36.8 93,342 8.8 49.4 Total 1,131,713 11.6 49.2 2,312,813 17.9 73.5 Source: Kwon, Population of Korea. a. Calculated as net migration divided by the terminal year population for each period. The census survival method was used to estimate net migration. b. The proportion of the increase of urban population attributable to net mi- gration. found that the greater the income disparity between a county and an urban area, the greater was the likelihood of migration to an urban area. Although the regression analysis of the effect of income differences supported the hypothesis, the effect of investment in education was ambiguous, and the effect of investment in health gave negative co- efficients for five measures of investment in health. In an attempt to sort out the push-pull factors influencing migration, one investigator asked 404 heads of household, who had migrated to Seoul, which of nineteen factors were major influences in their moves.7 Economic factors figured predominantly in both the push and pull cate- 7. Man-Gap Lee, "The Facts Behind Seoul's Exploding Population," in Report of the International Conference on Urban Problems and Regional Development (Seoul: Yonsei University, 1970). 118 PART TWO: HUMAN RESOURCfES gories. Employment prospects were cited by 373 respondents as the primary attraction; educational opportunities were cited by 101 respon- dents. Cultural opportunities, the desire for urban life, and personal reasons lagged far behind. The push factors prompting migration in- cluded wartime dislocation, poverty, family bankruptcy, and the desire to escape from traditional life. Most other surveys examining the moti- vation for migration indicate that the pull factor predominates and the movement of people from rural agriculture to urban manufacturing accelerates when economic growth is rapid. While not disputing the primacy of economic motivations, many sociologists point out that a complex set of social aspirations often parallels the economic goals. Migration is seen as a means to escape from the traditional constraints of life in rural areas and to enter a social environment that is more mobile and open. Population Projections The projections of the Korean population presented here assume a decline in the average number of children per woman from 3.6 in 1975 to 2.3 in 2000, a gradual rise in life expectancy to 73.9 years for males and 77.5 for females, and a net emigration of 30,000 persons a year." The projections also assume that the thrust and intensity of population programs and policies will not substantially change. Although the rate of population growth will slow down over the period, the population will increase by more than 50 percent to 54 million in 2000 (table 4.9). The projections show that the proportion of the population aged between 15 and 64 years will increase, thus reducing the dependency burden (table 4.10). Refinements of these projections into rural and urban components indicate that urban areas will continue to grow throughout the period, but at a slower rate than in the past. The population of rural areas will decline throughout the period (table 4.11). The demographic situation promises to improve considerably by the 8. These projections were prepared by Lee-Jay Cho and used by government in the fourth plan. Cho made two series, one based on a very rapid decline in fertility, the other on a slower fall. The second series has been used in this report. Lee-Jay Cho, "Projections of the Population of the Republic of Korea, 1975-2000" (Honolulu: East-West Population Institute, 1976; processed). POPULATION GROWTH, PROGRAMS, AND POLICIfES II9 Table 4.9. Total Population, Projections for 1975-2000 Annual rate of increase for Population five-year period Year (thousands of persons) (percent) 1975 35,280 1.76 1980 38,531 1.89 1985 42,362 1.86 1990 46,484 1.64 1995 50,464 1.38 2000 54,077 Source: Lee-Jay Cho, "Projections of the Population of the Republic of Korea, 1975-2000" (Honolulu: East-West Population Institute, 1976; processed), assumption II. end of this century: in 1975 selected demographic indicators aligned Korea with other developing countries; by 2000 these indicators, based on projection results, will place Korea more closely in line with the developed countries (table 4.12). Government may nevertheless decide that the population growth associated with achieving a total fertility rate of 2.3 in 2000 still is too rapid and that attempts should be made Table 4.10. Selected Measures of Changes in Age Composition, Projections for 1975-2000 (percent) Measure 1975 1980 1985 1990 1995 2000 Persons under 15 years 38.3 34.3 32.2 30.9 30.1 28.1 Persons 15-64 years 58.2 61.8 63.6 64.3 64.4 65.5 Persons over 64 years 3.9 3.8 4.2 4.7 5.4 6.4 Age dependency ratio 72.4 61.7 57.2 55.5 55.1 52.5 Median age (years) 18.7 20.5 22.4 24.1 25.5 27.7 Source: Cho, "Projections of the Population of Korea." 120 PART TWO: HUMAN RESOURCES Table 4.11. Selected Measures of Changes in Age Composition, by Area, Projections for 1975-2000 Area/measure 1975 1980 1985 1990 1995 2000 Urban population (thousands of persons) 16,603 20,383 24,545 28,959 33,357 37,475 Percentage under 15 years 30.81 29.02 29.66 28.83 27.14 24.72 Percentage 15-64 years 66.49 68.30 67.50 67.80 69.03 70.39 Percentage over 64 years 2.70 2.68 2.85 3.37 3.83 4.89 Ratio of females aged 15-49 to total females (percent) 59.40 60.40 58.30 57.50 58.40 58.80 Rural population (thousands of persons) 18,678 18,148 17,847 17,524 17,107 16,602 Percentage under 15 years 45.48 39.33 31.62 26.71 27.77 29.84 Percentage 15-64 years 50.41 55.67 62.40 66.31 63,98 60.65 Percentage over 64 years 4.11 5.01 5.98 6.98 8.25' -9.51 Ratio of females aged 15-49 to total females (percent) 41.30 45.00 50.20 52.20 48.60 45.30 Note: The definitions of urban and rural in the census have been adhered to; all figures are mid-year estimates. Source: World Bank projections of urban and rural population. to reach replacement-level fertility in 2000.9 To accomplish this further reduction, government would have to adopt other population measures in addition to those for the family planning program-examples of such measures are described in the last section of this chapter. The National Family Planning Program After the Korean War popular interest in family planning was grow- ing, but the conservative Rhee government preserved the pronatal stance of colonial rule. Japanese laws against abortion and the distribution of contraceptives remained intact. Nevertheless the free distribution of 9. Replacement level fertility implies a net reproduction rate of one-that is, women on average would have one female child each and would only be replacing themselves. This net reproduction rate is associated with a total fertility rate of about 2.1 children. Even if Korea were to achieve a net reproduction rate of one in 2000, the population would continue to grow until about 2075, when the age and sex structures would become stable and the population stationary at about 78 million persons. POPULATION GROWTH, PROGRAMS, AND POLICIES IZI condoms to U.S. troops and the Korean army made contraceptives more available than ever before. During the period of rehabilitation a number of intemational agencies active in Korea pointed out the seriousness of the population problem and urged that contraceptive services be in- cluded in public health programs. In 1957 the women of the Methodist Church opened family planning clinics in urban areas. About the same time, members of the obstetrics and sociology departments at major universities began to promote the family planning concept, and many private doctors were providing contraceptive services to their patients. After the fall of the Rhee government in 1960, the political climate was more receptive to the idea of family planning. Newspapers, maga- zines, and radio stations offered information on contraception. The International Planned Parenthood Federation and the Pathfinder Fund supported family planning activities. And in 1961 the Korean govern- ment, correctly recognizing that a continuing rapid rate of population growth would impede economic development, first adopted an official population policy. To implement this policy, the national family plan- ning program was launched in 1962. Until recently, government's at- tempts to influence fertility were confined to this program. But with the realization that females born during the baby boom were entering the childbearing cohorts, it was recognized that other policies would be required in addition to conventional information and contraceptive services. In March 1976 government announced the establishment of a committee, headed by the deputy prime minister, to formulate population strategies and devise policies facilitating redistribution of the population. Operational aspects Four organizations have primary responsibility for implementing the national program: the Ministry of Health and Social Affairs (MoHSA), the Ministry of Home Affairs (MOHA), the Planned Parenthood Fed- eration of Korea (PPFK), and the Korean Institute of Family Planning (KIFP). The central government finances about 49 percent of the pro- gram, provincial governments about 17 percent, and foreign agencies about 34 percent. In 1974 the program accounted for roughly 11 percent of MOHSA'5 budget, which in tum constituted only 0.9 percent of the national budget. MOHSA is in charge of planning and coordination, but local authorities, which report to MOHA, run the health centers. Provin- cial governors appoint directors of each health center; county chiefs appoint fieldworkers. It has often been pointed out that MOHSA'5 lack of operating authority is a major weakness of family planning services and the health system generally. Reorganization of the health services 122 PART TWO: HUMAN RESOURCES Table 4.12. Demographic Indicators, Korea and Selected Countries, Projections for 2000 Total Crude Crude population Annual birth rate death rate (thousands growth rate Country of persons) (percent) (per thousand) Korea 54,077 1.30 18.0 4.9 Brazil 212,507 2.42 29.4 5.3 Chile 15,355 1.28 19.8 7.0 Ghana 21,164 2.85 39.8 11.4 India 1,059,429 1.77 26.3 8.7 Japan 132,929 0.51 14.5 9.4 Mexico 132,244 2.96 35.2 5.2 Sweden 9,930 0.44 14.3 12.0 United States 264,430 0.65 14.9 9.9 Sources: Cho, "Projections of the Population of Korea"; United Nations, Se- lected Demographic Indicators, medium variant. has been the subject of intensive discussion, and the family planning program may be affected by the outcome. The methods and scale of the national program have not changed significantly since 1968. At the end of 1975 the program employed 2,767 family planning workers. The ratio of workers to eligible couples was 1:1,200 in rural areas and 1:4,000 in urban areas.10 Fieldworkers recruit acceptors by making home visits and conducting group meetings. They distribute condoms and oral contraceptives and refer acceptors of intrauterine devices (IUDS) and sterilization to physicians authorized and trained by government. These physicians provide services at their offices and are reimbursed by government. The Maternal and Child Health Law of 1973 legalized abortion and authorized the insertion of IUDs by paramedicals. But by the end of 1975 government had trained only an estimated 175 nurses and midwives to insert rUDS, far fewer than the number required to enable this law to have a major impact. Recruitment is also carried out in connection with postpartum care and matemal and child health services, but only about 10 percent of de- liveries take place in hospital. Postpartum recruitment thus is not significant in Korea. 10. An eligible couple is defined as a married couple with the woman between the ages of 20 and 44 years. POPULATION GROWTH, PROGRAMS, AND POLICIES I23 Total fertility Life expectancy Age rate Male Female dependency Median (children ratio age per woman) (years) (percent) (years) Country 2.3 73.9 77.5 52.5 27.3 Korea 3.8 68.5 73.9 71.7 21.3 Brazil 2.4 67.4 73.5 53.4 27.3 Chile 5.3 54.2 57.8 72.6 17.3 Ghana 3.5 61.5 61.9 63.9 22.3 India 2.0 72.3 77.5 49.9 36.2 Japan 4.9 68.4 72.3 84.1 18.5 Mexico 2.1 72.3 78.1 54.3 36.1 Sweden 2.1 69.4 75.8 50.3 34.0 United States Fieldworkers devote about twenty days a month to home visiting. The rest of their time is allocated to a variety of tasks, such as partici- pating in special campaigns to prevent disease. On each initial visit they complete a household inventory form which assembles demographic, health, and economic information. They make approximately three follow-up visits in the ensuing six months. Because of the mobile char- acter of urban residents and the low ratio of urban fieldworkers to the population, home visiting has not been as satisfactory in urban areas as in rural areas. In view of the increasing urban concentration of the population, more intensive efforts wvill have to be undertaken to meet the demands of the urban market. Korea's fourth plan has emphasized the importance of such efforts, and a number of multipurpose clinics were recently established in Seoul. Fieldworkers must at least be nurses, a requirement which implies a three-year training course after completing middle school. When hired, they receive two weeks of training at ICIFP. The annual turnover is 25 to 30 percent. Part of this turnover is the result of marriage. But it also appears that the classification of the job as temporary, the absence of opportunities for career development, and modest pay provide little encouragement for workers to stay in their jobs or to perform them well. To keep their jobs and receive maximum bonuses, fieldworkers try to meet prescribed targets. This means that they frequently report 124 PART TWO: HUMAN RESOURCES the achievement of targets, regardless of the actual number of acceptors recruited. This distorts service statistics and greatly reduces the value of these statistics for purposes of management and evaluation. The emphasis on achieving targets also encourages fieldworkers to concen- trate more on bringing in new acceptors, whose ranks may be subject to rapid attrition, than on maintaining contact with past acceptors. The existing system of targets introduces a number of difficulties. Targets are arbitrarily set by MOHSA without regard to the preferences of a particular area. The amount of the incentive is pegged to the method adopted. Consequently fieldworkers promote the methods with the higher incentives rather than try to satisfy the wishes of the new ac- ceptor. Given the defects of this system, alternative procedures are being explored. One pilot project is experimenting with a system that assigns targets in points rather than by methods. Under this scheme, acceptors are more likely to receive the method of their choice, and continuation rates should improve. Such a system, although more flex- ible, may still not solve the problem of false reporting. Information, education, and communication The Planned Parenthood Federation of Korea (PPFK) is responsible for the information, education, and communication (IEC) functions of the national program. Its communication strategy has responded to changing conditions at various stages of program development. Initial efforts, aimed at spontaneous acceptors, were mainly informative. As this audience was absorbed into the program, a new communications ap- proach was developed to attract more reluctant groups. PPFK initiated long-range motivational campaigns to mitigate the influence of such tra- ditional attitudes as the preference for sons and large families. Its stop-at- two campaigns have emphasized the desirabilitv of the t,wo-child family, regardless of their sex, and the importance of raising them well. In view of increasing urbanization and the growing number of young people, PPFK began to direct specific campaigns, reinforced by the use of mass media, to urban slum groups and younger audiences. Although the media are not effective in actually recruiting acceptors, they perform the essential functions of keeping the issue before the public and culti- vating an environment conducive to the acceptance of family planning. Moreover media penetration in Korea is quite good. There are approxi- mately 5 million radios in the country and, because literacv is high, the press is an effective means for disseminating the family planning message. POPULATION GROWTH, PROGRAMS, AND POLICIES 125 In addition to using the mass media, PPFK exploits such interpersonal channels of communication as mother's clubs. The system of mother's clubs, established in 1968 to supplement the activities of government fieldworkers, has been particularly effective in rural areas. In 1975 a total of 705,000 women belonged to about 28,000 clubs. Mothers' clubs stress the importance of family planning as one aspect of responsible parent- hood and try to show women the impact their fertility behavior has on Korean society as a whole. In urban areas PPFK sponsors clinic-centered iEc activities and provides the necessary personnel and materials. It operates fourteen family planning clinics, staffed by medical personnel and fieldworkers, in areas which do not have ready access to government health centers. PPFK also uses other organizational networks, such as the Homeland Reserve Forces Project and the Enterprise Project, to spread the familv planning message. About 2.7 million men between the ages of 25 and 35 regularly attend reserve training sessions, and PPFK uses these sessions to motivate reservists to accept vasectomy. A similar program has been instituted for the regular army and civil ser- vice employees. Under the Enterprise Project, family planning informa- tion and services are provided to workers at their factories. In addition, the Korean Federation of Trade Unions sponsors family planning ser- vices at factory dispensaries. Because the target audience of all these motivational activities is predominantly male, the number of vasectomy acceptors should increase. Monitoring, research, and evaluation The Korean Institute of Family Planning (KIFP) was established in 1970 to undertake research, evaluation, and the training of program personnel. KIFP collects service statistics and reports to MOHSA on program achievement. Each month the 197 health centers submit infor- mation on personnel, registration, performance, and supplies of contra- ceptives to provincial headquarters for forwarding to KIFP. Health centers also assemble information from the seventy-five hospitals partici- pating in the national program. In order to evaluate the family planning activity in Seoul's slum areas, KFP compiles monthly reports for the ten urban family planning centers. These service statistics are used to eval- uate target attainment, the level of contraceptive use, and the compara- tive progress in various areas. KiFP also analyzes the coupons submitted by fieldworkers for each acceptor of an IUD or sterilization. A complete tabulation of coupons, which contain demographic information about the acceptor, has been made quarterly since 1974. Although progress 126 PART TWO: HUMAN RESOURCES has been steady, the research and evaluation could be considerably im- proved. The reliability of service statistics is questionable, and available data have been less than fully used. Although KIFP has primary responsibility for these functions, a num- ber of universities and PPFK also conduct research and evaluation. In 1976 PPFK launched a community-based distribution scheme in order to explore the extent to which Koreans are willing to practice family planning at their own expense. The project was felt to be in line with the increased prosperity in Korea and with the new social goals, pro- pounded by the New Community movement, which call for individual and collective efforts to improve the quality of life. The agenda for future research should include studies that explore reasons for non- acceptance, particularly among women who do not want additional children; community variables that may affect the level of acceptance; and ways to improve the performance of family planning workers. Program performance Korea's national family planning program has been acclaimed as one of the most successful of all national programs. The ambitious goal of reducing the annual growth rate from 3 percent in 1960 to 2 percent in 1970 was reached. During the early years of the program, the number of acceptors grew rapidly (table 4.13). But in view of the initial success, government became complacent and reduced the budget appropria- tions for the program. Acceptance rates for the IUD and sterilization markedly decreased between 1967 and 1968, and continuation rates were disappointing. Salaries of fieldworkers remained constant from 1966 until mid-1969, during which period their real incomes fell 22 percent. Also affecting the morale of family planning personnel was the proposal by MOESA in 1968 that workers be retrained to accommo- date the integration of family planning, maternal and child health, and tuberculosis services. Raising the qualifications for family planning workers forced many older and experienced workers to leave the pro- gram. They were replaced bv women who had more formal education, but who usually were younger, unmarried, and less adept at establishing rapport with their clients. Even with these difficulties, the program appeared to be doing well in the early 1970s. According to a 1973 survey, many major indexes revealed that the program had emerged from the doldrums of the late 1960s (table 4.14). The results also showed a steady rise in the propor- tion of women who had used contraception and in the number of POPULATION GROWTH, PROGRAMS, AND POLICIrS 1 27 Table 4.13. Achievements of National Family Planning Program, 1962-76 Method of contraception Intrauterine Condom Oral pill Year device Sterilization (monthly) (monthly) 1962 n.a. 3,413 59,000 n.a. 1963 1,494 19,866 129,804 n.a. 1964 106,397 26,256 156,301 n.a. 1965 225,951 12,855 191,706 n.a. 1966 391,687 19,942 168,868 n.a. 1967 323,452 19,677 152,724 n.a. 1968 263,132 15,988 135,184 26,264 1969 285,500 15,457 147,773 91,175 1970 295,100 17,321 162,986 170,512 1971 266,641 15,448 174,187 222,823 1972 300,309 19,698 155,605 213,948 1973 325,935 24,492 176,032 234,698 1974 349,510 35,627 173,918 242,075 1975 343,907 43,056 196,735 240,183 1976 297,872 44,881 158,070 203,402 n.a. Not available. Source: Data supplied by the Korean Institute of Family Planning (CIFP). women currently using contraception. Compared with an estimated 25 percent in 1971, 36 percent of married women were using contraceptives in 1973. The national average for ideal family size was 3.1 children. In Seoul it was 2.8; in other urban areas 2.9. The survey estimated the total fertility rate for 1973 to be 3.9, a considerable decline from about 4.5 during the last years of the 1960s. Notable differences in fertility values and family planning practices still remained, however, between urban and rural areas. The service statistics are adequate to show patterns in the acceptance and continuation of contraception. The IUD has been the principal method since the beginning of the program (table 4.15). Continuation rates after 12 and 24 months are much better for the IUD than for the pill. Thus, while the clinical effectiveness of the pill is superior to that of the IUD, the pill's low levels of use-effectiveness make the IUD the preferable method. The contraceptive mix being pursued under the fourth plan shifts the emphasis from methods with low use-effectiveness to those with high use-effectiveness. One sterilization has about three times the demographic impact as one IUD insertion. And it is generally 128 PART TWO: HUMAN RESOURCES Table 4.14. Selected Findings of the 1973 Family Planning and Fertility Survey Other Item Korea Seoul urban Rural Mean ideal number of cbildren, 3.1 2.8 2.9 3.4 Mean ideal number of sonsb 1.9 1.8 1.8 2.1 Desired family size (children) 3.7 3.0 3.4 4.2 Mean age at first marriage for females (years) 22.8 n.a. 23.3 22.2 Mean duration of marriage at first use of contraceptives (years) 9.2 7.3 8.1 10.7 Mean age at first use of contraceptives (years) 29.3 n.a. 28.6 30.0 Mean number of living children at first use 3.1 n.a. 2.7 3.5 Mean number of living sons at first use 1.7 n.a. 1.5 2.0 Proportion of contraceptive users at time of survey' (percent) 36.0 n.a. 39.0 34.0 Currently married women having used contraceptiond (percent) 55.0 58.0 55.0 54.0 Ratio of current users to ever users' (percent) 66.0 n.a. n.a. n.a. Current users using government program (percent) 60.0 n.a. 49.0 60.0 Total fertility rate' (children per woman) 3.9 n.a. 3.3 4.7 n.a. Not available. Source: KIFP, The 1973 National Family Planning and Fertility Survey (Seoul, 1974). a. The national mean declined by 0.6 children from 3.7 in 1971. b. The national mean declined by 0.3 sons from 2.2 in 1971. c. Past proportions of contraceptive users at the time of survey were 9 percent in 1964, 20 percent in 1967, and 25 percent in 1971. d. Past proportions of currently married women having used contraception were 12 percent in 1964, 28 percent in 1967, and 44 percent in 1971. e. Past ratios of current users to ever users were 75 percent in 1964, 72 percent in 1967, and 57 percent in 1971. f. The total fertility rate in 1971 was 4.8 children per woman. held that births prevented by the acceptance of IuDs and pills are in the ratio of about three to two. The plan envisions higher levels of sterilization and abortion, less use of IUDs, and a big decline in the use of oral contraceptives and condoms. The plan may be giving too much emphasis to abortion, which is most efficient as a backup for contracep- tive failure, not as a contraceptive method. POPULATION GROWTH, PROGRAMS, AND POLICIES 129 Table 4.15. Family Planning Acceptors, by Method, 1962-75 (persons) 1962-75 Method annual average 1975 Pill 141,000 97,559 Intrauterine device 350,000 280,733 Condom 217,000 n.a. Sterilizatione 31,000 57,588 Other methods' n.a. 3,877 Total 739,000 439,757 n.a. Not available. Source: Data supplied by KxFP. a. Pills became available in 1968. b. Includes tubectomy and vasectomy. c. Includes induced abortion. Projections of the number of women who will have to practice con- traception if replacement fertility is to be reached by 2000 provide some idea of the scope of the task lying ahead. By applying Bogues' fertility- contraception function to the national population projections, it is found that the proportion of married women practicing contraception must increase from 36.2 percent in 1975 to 51.1 percent in 2000 (table 4.16). The number of women practicing contraception must increase from 1.8 million in 1975 to 4.6 million in 2000. In fact the number would have to be substantially higher than 4.6 million because many contraceptors are not efficient and many subfecund women unnecessarily practice contraception. Determinants of Changing Fertility Between 1960 and 1970 the total fertility rate for urban areas fell from 5.4 children to 3.2; that for rural areas from 6.8 to 4.4.11 For a ten-year period these declines are substantial. The decline in urban fertility was to be expected, given the inroads of modernization and the easy access to contraceptive services. The decline in rural fertility, how- ever, was largely attributable to the efforts of the government family 11. Cho, Demographic Situation. 130 PART TWO: HUMAN RESOURCES Table 4.16. Projected Number of Contraceptors Necessary to Reach Replacement Fertility by 2000 (thousands) Item 1975 1980 1985 1990 1995 2000 Females aged 15-49 8,787 10,030 11,140 12,113 13,059 14,187 Married women 5,128 5,891 6,772 7,640 8,406 9,100 Married fecund women 3,861 4,439 5,149 5,825 6,350 6,659 Married fecund women who must be contracepting 1,857 2,530 3,221 3,871 4,370 4,648 Sources: Cho, "Projections of the Population of Korea"; World Bank projections. planning program to disseminate family planning information and pro- vide services in rural areas. The distinct contributions of the family planning program and general socioeconomic development to the reduction in fertility cannot be iso- lated. It can only be observed that the network of PPFK mother's clubs and the activities of govemment fieldworkers appear to have been effective in increasing public awareness of family planning services, particularly in rural areas. To this extent, the national program has facilitated the decline in fertility. The decline in the desired number of children, however, is likely to be more a reflection of fundamental socioeconomic changes than of any advertising campaigns mounted by the national program. One major variable accounting for urban-rural differences in fertility is the higher educational level of urban women. The Special Demo- graphic Survey of 1966 and the 1973 survey both shed light on the inverse relation between educational attainment and completed fer- tility.12 The more education a woman has, the later she is likely to marry, the more she is likely to practice contraception, and the lower her fertility norms are likely to be. The 1966 survey showed that married women with no education considered 4.5 children to be the ideal num- ber; women who completed high school 3.8; and women who with uni- versity education 3.1. Education influenced fertility norms more than 12. E. H. Choe and J. S. Park, Some Findings from the Special Demographic Survey, The Population and Development Studies Center Publication Series, no. 3 (Seoul, 1969); Korean Institute of Family Planning (Kipp), The 1973 National Family Planning and Fertility Survey (Seoul, 1974). POPULATION GROWTH, PROGRAMS, AND POLICIES 13I place of residence. When residence was controlled for, the inverse rela- tion between education and fertility still held. The 1973 survey indi- cated that the higher the wife's education, the fewer the children she expected. Those with no education expected a mean of 4.28 children; those with primary school education 3.71; those with middle school education 3.03; those with high school education 2.93; and those with college education 2.73. The inverse relation between fertility and educational attainment of women has been found in most countries. But it is not clear whether this relation holds because education raises the earning potential of women, and hence the opportunity cost of time devoted to bringing up children, or because education imparts different attitudes and values. Experience in Korea suggests that the opportunity-cost argument is more powerful in accounting for differences in the age of marriage by educa- tion than for differences in marital fertility by education. Despite mod- emization and the increases in educational attainment, it appears that the traditional cultural system, which decrees that women should leave the labor force when they marry, still prevails. The dominant pattern for most women employed in nontraditional activities-as secretaries, factory workers, and nurses-is to work until age 22 or 23 and then leave to get married. This may change in the future, particularly if the demand for labor increases. In addition, the cohort of women who postponed marriage to take jobs in the modern sector in the 1960s may break further with tradition and attempt to return to the labor market when their children are older. The most straightforward explanation of the rising age at marriage for women would seem to be that the growth of labor-intensive manu- facturing has offered them an alternative to marriage. But it would be a mistake to infer that employment and educational opportunities were the only reasons for the rising age at marriage. The age at marriage for men has also been rising. Military service and progressively longer periods spent in full-time education have delayed the entry of men into full-time employment. It may also be that men, marrying at an older age, seek older brides. A survey conducted in 1970 in connection with the nationwide value- of-children study provides some additional evidence on fertility norms. It attempted to discern the link between fertility behavior and value orientations. The hypothesis was that the value system of individuals ultimately causes them to favor, adopt, and continue the practice of family planning. This hypothesis was tested in the Korean setting by trying to identify the satisfactions and costs of having children and to 132 PARl TWO: HUMAN RESOURCES relate the value of children to other attitudinal and behavioral aspects of parents. The study revealed that the preference for sons is one of the strongest factors influencing fertility behavior in Korea. Girls were valued for such personal qualities as companionship and affection; boys for the continuity of the family name and economic contributions to the parents, particularly in old age. The expectation of financial support from sons was greater among rural respondents. Parental satisfaction and the cost of children were significant predictors of current use of contraceptives. To learn more about the determinants of fertility in Korea, further research could readily be conducted to examine the nature of urban and rural fertility differences and the extent to which these differences reflect different levels of educational attainment, different occupational structures, and different experiences of declining mortality. Research on fertility differences of migrants and nonmigrants would also be use- ful. Migrants usually exhibit lower fertility than nonmigrants at the place of origin, but higher fertility than the population at the place of destination. Thus, by migrating, they raise the fertility levels of both places. Because rural-to-urban migration has been extensive during the past fifteen years and is expected to continue, research examining migrant-nonmigrant differences might help to increase the understand- ing of fertility levels and the ways these are likely to evolve in the future. Policies Beyond Family Planning Although fertility and the declared ideal size of families have been falling, the continuing strong preference for sons suggests that it could be many years before the goal of a two-child family becomes the norm for most Korean parents. This likelihood has prompted government to adopt a strategy to influence the desired size of families. The measures associated with this strategy are termed "beyond family planning." They work in tandem with service-oriented programs and are intended to increase the demand for contraceptive services. Several measures that would promote a strong antinatal bias and make communities aware of the need to accept the small family norm have been advocated. Most of these measures are still in their formative stages; many are conceptual and lack supporting data. Consequently the applicability of any of these alternatives in the Korean context must be the subject of detailed dis- cussion and evaluation. POPULATION GROWTH, PROGRAMS, AND POLICIES 133 Population education Recent knowledge, attitude, and practice (KAP) surveys reveal that traditional ideas regarding the large family and the desirability of male children have been somewhat modified, but that they nevertheless con- tinue to influence fertility behavior. Because education was felt to be the most obvious means for recasting traditional values, population education was introduced in Korean schools in 1977. The Korean Edu- cational Development Institute (KEDI) developed a detailed curriculum to integrate population considerations with all aspects of study. Popula- tion is addressed from biological, demographic, environmental, and ethical points of view. The curriculum deals with the traditional factors influencing fertility ideals: Confucianism, family ethics, the preference for sons, and the patriarchal system. It also stresses the social equality of men and women, the emergence of a new role for women, and the relevance of family planning in this context. Community incentives Because decisions about family size normally are made privately by couples, government exhortations about modifying fertility norms in the interest of national welfare are for the most part incomprehensible to villagers. The community is better suited to act as an intermediary by translating national goals into terms that make sense at local level. In order to motivate individuals to modify their reproductive behavior, the benefits of lower fertility should be identified with benefits to individ- uals, families, and local communities. The idea of community incentives appears to be particularly suitable in Korea. The New Community movement provides different levels of public services to different grades of villages. This grading is based on judgments about a village's per- formance and potential. It might be advisable to have reduced fertility added to the criteria for judging villages and allocating resources. This would require improved registration of vital data, but this, too, could be included in the list of performance criteria. Other public policies Government could also manipulate income tax exemptions, access to public housing, and school fees to increase the cost of children to parents. But these measures would affect only a small proportion of the popula- tion. In contrast with such measures are those that reward the low fer- tility behavior of parents. Other countries have experimented with a 134 PART TWO: HUMAN RESOURCES variety of schemes: small cash payments to family planning acceptors; regular bonuses paid to married women for every year they go without bearing a child; fairly large payments made on evidence that completed family size has been kept small, either because one of a couple has been sterilized or the woman has reached the end of the childbearing ages.'3 Cash payments for sterilization are relatively simple to administer because problems of maintaining continuous enrollment and making repeated checks are eliminated. But immediate cash payments for steri- lization raise the objection that individuals in financial difficulty may be induced to make irreversible decisions which they later regret. Deferred incentive schemes, on the other hand, are more attractive. Payments are made only after society has reaped the benefits of lower fertility. Because such savings contribute to economic growth, the size of the deferred benefits can be much larger than immediate payments. But there are two major drawbacks. The remoteness in time of the benefits may make them unattractive; the schemes may run into substantial administrative difficul- ties. Implementing any deferred incentive scheme requires establishing an accurate record of fertility. This would be especially difficult in Korea, where registration of births is far from universal. 13. For an evaluation of this approach see Roberto Cuca and Catherine S. Pierce, Experiments in Family Planning: Lessons from the Developing World (Baltimore: Johns Hopkins University Press, 1977), pp. 61-63, 144, 210, 220. Chapter 5 Fredrick Golladay and Timothy King Social Development IN COMPARISON with many developing countries, the proportion of pub- lic expenditure committed to social development in Korea has been low. Most of this expenditure has been directed to education and the supply of piped water. The strategy was to focus the resources and efforts of government on activities that would accelerate economic growth and to leave many aspects of social development to the private sector. It appears to have worked. The rising incomes and higher absolute expen- diture for social development that accompanied rapid economic growth have led to improved health, high levels of educational attainment, and reasonable access to basic services. Even so, a number of problems persist. Health services are less acces- sible in rural areas than in urban areas, medical insurance barely exists, and the cost of hospitalization is beyond the reach of many Koreans. Mortality and morbidity could be further reduced by such low-cost interventions as prenatal care, immunization, basic preventive measures, sanitary water supplies, and sewage disposal. Primary education, al- though compulsory, is not yet free of all charges. Fees at middle and high schools constitute a heavy burden for poor families. In addition, housing conditions are far from satisfactory, working conditions in in- dustry frequently are poor, and social insurance to provide for old age or protect income in the event of death or disability has not been intro- duced. Consequently government is now exploring ways to promote social development. The investment program for social development is being substantially expanded under the fourth plan (table 5.1). Compared with the third plan, the increases in investment in education and training and in water supply and sewerage will be nearly threefold. The increase in investment in health will be more than fivefold. The figures presented in table 5.1 do not include the most dramatic increase in spending on human re- sources-a rise in recurrent public expenditure for education from 3.1 135 136 PART TWO: HUMAN RESOURCES Table 5.1. Investment Program for Social Development, by Service, 1972-81 1972-76 1977-81 Percentage Percentage Index Billions of total Billions of total (1972-76 Service of won investment of won investment = 100) Education and man- power development 254 2.2 749 3.9 295 Health, 38 0.3 211 1.1 555 Housing' 1,670 14.6 2,641 13.9 158 Water supply and sewerage 61 0.5 182 1.0 298 Total 2,023 17.7 3,783 19.9 187 Note: This table follows the classification of expenditure in the fourth plan and enables comparison of the third and fourth plan periods. Social security, which is classified among other services in the fourth plan, is not included in this table. All figures are in 1975 prices. Source: Economic Planning Board (EPB), The Fourth Five-Year Economic De- velopment Plan, 1977-81 (Seoul, 1976). a. Although health includes disease control and maternal and child health services, it apparently does not include family planning activities. b. Water supply and sewerage refer to programs of the Ministry of Construc- tion, which has the main responsibility for this sector. The Ministry of Health and Social Affairs (MOHSA) has responsibility for supplying piped water to vil- lages having fewer than 1,000 inhabitants and for some urban waste treatment. The investment program for health therefore includes W15 billion in 1972-76 and W62.9 billion in 1977-81 for piped water and waste treatment. percent of GNP and 12.1 percent of the government budget in 1976 to 5.1 percent of GNP and 16.7 percent of the budget in 1981. Health Conditions and Services Health conditions in Korea compare favorably with those in other developing countries. In East Asia, only the Republic of China, Hong Kong, and Singapore have markedly superior records, and this may be explained by their higher levels of income and unusually high concen- trations of health resources. The improvement of health conditions in Korea may in part be attributed to high levels of literacy, to seasonal cropping which interrupts the breeding and feeding of vectors and para- sites, and to good nutrition. These advantages are offset by problems of crowding and pollution, but only partially. In addition, supplies of SOC1AL DEVELOPMENT 137 health resources have grown spectacularly in recent years. The number of qualified physicians quadrupled between 1960 and 1975. The popula- tion served by sanitary water supplies quintupled over the same period. The proportion of deaths attributable to epidemics declined from an average of 2.7 percent in the early 1960s to less than 0.05 percent after 1971. Over the same period, the estimated crude death rate fell from about 15 per thousand population to about 9 per thousand. These statistics indicate that health conditions have significantly improved despite the modest scale of government initiatives. This apparent con- tradiction can be explained by two sets of facts. First, the private health care sector is large and growing rapidly. Second, nutrition, housing, and sanitation are improving because of the substantial growth in per capita incomes. This experience is consistent with the widespread belief that the environmental and nutritional improvements accompanying eco- nomic development significantly affect health conditions. The pattern of disease in Korea is highly predictable.' Gastroenteric, respiratory, and dermatologic disease, which reflect problems of sani- tation, crowding, water supply, and waste disposal, are the most com- mon (table 5.2). About two-thirds of incidents of disease have been diagnosed as respiratory or gastrointestinal infections; dermatologic complaints account for perhaps another 4 percent of cases. Chronic diseases-cancer, cardiovascular disorders, diabetes, and stroke-account for 5 to 7 percent of cases, but their chronicity makes their significance disproportionate to their incidence. Three-quarters of all incidents of disease apparently could be avoided, or would be trivial and self-limiting, if sanitation, nutrition, housing, and water supplies were further improved. Poor sanitation explains the high incidence of parasitic infections. Although it has been illegal since 1968 to use untreated nightsoil as fertilizer, a recent study of parasitic eggs in soil samples and vegetables revealed that the law is widely ignored.2 It is suspected that in poorer 1. Most health-related activities are performed by the private sector and by ministries other than the Ministry of Health and Social Affairs (monISA). Re- markably little is known about the operations and scale of activities of the private sector. The observations in this discussion are consequently based on fragmented and limited studies and provide only a rough sense of the social significance of nonfatal disease in Korea. 2. Moonshik Zong, "A Survey of Parasitic Eggs in Soils and Vegetables in Korean Rural Areas," in Chunseong Gun Community Health Program (Seoul: Seoul National University, School of Public Health, 1975), pp. 99-100. I38 PART TWO: HUMAN RESOURCES Table 5.2. Cause-specific Morbidity for Selected Subpopulations (cases per thousand population per month) Univer- Univer- Female sity Seoul sity Rural Cause workers families teachers students dwellers Infective and parasitic diseases 4.6 3.3 0.6 1.8 2.4 Neoplasms 0.4 1.1 0.9 n.a. 0.8 Endocrine, nutritional, and metabolic diseases 6.3 13.3 0.3 0.7 3.3 Diseases of blood and blood-forming organs n.a. 7.8 0.9 0.2 1.2 Mental disorders n.a. n.a. n.a. 0.0 1.4 Nervous disorders 5.4 28.9 1.2 2.8 9.4 Circulatory disorders 0.2 28.9 3.1 0.2 4.3 Respiratory disorders 22.3 86.7 32.0 9.4 61.9 Digestive disorders 12.4 86.7 11.5 6.2 25.7 Dermatologic disorders 3.0 11.2 0.9 3.6 4.9 Musculoskeletal disorders 5.6 12.2 3.1 0.4 6.3 Genitourinary disorders 0.7 5.6 1.9 1.4 2.4 Complications of childbirth n.a. n.a. 2.2 n.a. 3.9 Congenital anomalies n.a. n.a. n.a. n.a. 0.3 Diseases peculiar to newborn infants n.a. n.a. n.a. n.a. 0.3 Signs, symptoms, and ill-defined conditions n.a. 36.7 16.8 1.5 23.0 Injuries 2.6 2.2 1.2 n.a. 6.5 Total 63.5 325.6 81.3 28.3 158.0 n.a. Not available. Sources: Female workers from Kyung Kyun Kim, "A Survey on Sickness and Utilization of Medical Care of Working Girls in Pharmaceutical Companies," Korean Journal of Public Health, vol. 9, no. 1 (1972); university families from Ji Yong Yang, "A Study on the Health Care of Ewha Woman's University Staff, Faculty and Families," Korean Journal of Preventive Medicine, vol. 7, no. 1 (1974); Seoul teachers from Hee Yong Oh, Bang Bu Youn, and Ki Soon Kun, "A Study on Illness and Health Care Patterns of Seoul City Primary, Middle and High School Teachers," The New Medical Journal, vol. 18, no. 4 (1975); univer- sity students from Ewha Woman's University Health Center, "Student Health Status, 1972-74," University Health Annal, vol. 3 (1975); rural dwellers from Jong Huh and Ok Ryun Moon, "A Study of Demand and Need for Medical Care in Rural Areas in Korea" (n.p., 1975; processed). SOCIAL DEVELOPMENT 139 areas, where dependence on nightsoil as fertilizer is inescapable, the problem is more serious. The potability of water supplies is closely re- lated to the disposal of excreta. About 43 percent of the population was served by piped, centralized water supplies in 1973: in urban areas 71 percent received piped water; in rural areas only about 10 percent re- ceived this service. The rest obtained water from private wells, or in a few instances from untreated surface sources, which frequently are contaminated. The sanitation of water supplies in rural areas has never- theless improved considerably since the time of this study. Casual observation suggests that there is little gross malnutrition in Korea. But because of traditional dietary habits, certain nutrients are missing from the diets of Koreans, even among relatively affluent groups. As would be expected, the problem is more severe in rural areas and among the poor. Persons in the subsistence sector consume only small amounts of fruit and vegetables and almost no meat or fish, particularly during the winter months. Rice, augmented by modest quantities of fermented vegetables, dominates the typical diet; noodles in broth is a common midafternoon snack. According to estimates for 1975, annual consumption of foodgrains is about 228 kilograms a person, the equiv- alent of 27 ounces a day (table 5.3).3 The daily caloric intake is about 2,450 calories a person; the daily protein intake is about 70 grams a person. Both values appear high and probably reflect the underestima- tion of waste in marketing foodgrains. Although the Ministry of Health and Social Affairs (MoHSA) has technical responsibility for the health activities of the nation, the pro- grams that affect health are scattered among several other ministries. The operation of community-level health centers, the largest program- related component, is under the jurisdiction of the Ministry of Home Affairs (MOHA). Health centers have no official relationship, either administrative or medical, with the hospitals or clinics in their area. Hospital facilities are operated by a number of public agencies. Sub- specialty hospitals are attached to medical colleges and thus are under the jurisdiction of the Ministry of Education. Environmental health activities-such as water supply, sewerage, and pollution abatement- are under the authority of MOHA, which provides services directly to individuals. MOHSA thus has primary responsibility for health, but little 3. These estimates were prepared by a World Bank mission in 1973. Estimates of food consumption reported in chapter 7 differ slighdy from those presented here, but nevertheless support the general observations made about nutrition levels. 140 PART TWO: HUMAN RESOURCES Table 5.3. Average Korean Diet, by Selected Nutrient, 1975 Consumptionz Daily intake Grams Kilograms Percentage of Nutrient a year of total Calories protein Rice 126.0 29.8 1,243 23 Wheat 40.9 9.7 408 12 Barley 25.3 6.0 242 6 Other grains 23.3 5.5 230 5 Total foodgrains 215.5 51.0 2,123 46 Fruit 15.2 3.6 24 1 Vegetables 105.1 24.9 40 3 Potatoes 27.6 8.9 86 2 Total fruit and vegetables 147.9 37.4 150 6 Pulses 1.0 0.2 9 1 Beef 1.9 0.4 9 1 Pork 4.7 1.1 48 2 Chicken 1.9 0.4 6 1 Fish 24.9 5.9 68 11 Total meat and fish 33.4 7.8 131 15 Milk 11.3 2.7 19 1 Eggs 3.5 0.8 16 1 Total 412.6 99.9 2,448 70 Note: All figures are estimates. Source: World Bank, "Agricultural Sector Study, Republic of Korea," report no. 213a-KO (a restricted-circulation document) (Washington, D.C., 1973; pro- cessed), p. 34. capacity to affect health, either through its own activities or through the other agencies that perform health-related functions. Until a co- ordinating mechanism is established with the authority and stature to influence significantly the programs of all these agencies, the efforts to deliver effective health services will continue to be diffuse. Korea has a smaller ratio of health providers to population than other Asian countries at comparable levels of economic development, except for pharmacists for whom the ratio is more than double the average. Pharmacists constitute the largest category of health professional in SOCIAL DEiVELOPMENT 141 Korea; they are a major source of primary diagnostic and therapeutic services. In 1973 there were about 18,000 licensed pharmacists, about 10,000 licensed physicians, more than 20,000 licensed nurses, about 2,500 dentists, and about 2,700 herbal doctors.4 Medical manpower is concentrated in Seoul and Busan. The two special cities have roughly double the national average of health providers per unit of population, except for midwives, who are more common in rural areas. The private sector dominates the health care system in Korea: about 93 percent of licensed physicians are in private practice; about 97 per- cent of the 11,000 establishments providing care operate privately. Services are provided largely in privately owned clinics and hospitals, and health care is financed almost totally by individual households on an episodic basis. It has been estimated from household expenditure surveys that Korean households spent about 2.5 percent of disposable income on personal health care in 1973. Only a miniscule proportion was covered by any form of life insurance: the seven health insurance schemes operating in 1974 provided benefits only to about 38,000 sub- scribers. Private health expenditure in 1974 was about W168.1 billion, a considerable rise from W4.8 billion in 1960.1 With GNP expected to increase by 78 percent in real terms between 1974 and 1981, annual health expenditure is likely to more than double and exceed W300 billion, or approximately W9,000 a person, by 1981. Oriental conceptions of disease and economic and spatial barriers in- fluence Koreans in their selection of health care resources. Diseases causing obvious changes in normal physical states-such as elevated temperature, diarrhea, or convulsions-are generally regarded to be within the domain of oriental medicine. Accidents, injuries, sores, and other disorders having specific and visible signs are regarded to be within the domain of western medicine. It nevertheless appears that the domains of oriental and westem medicine are not clearly demar- cated. As a result, valuable treatment time is often lost because of the improper election by patients of the mode of care. This pattern reduces the efficacy of both medical traditions. A major reason for heavy re- liance on traditional cures and drug therapy is economic. The cost of modern medical care is high in relation to family income, and the pressure is strong to rely on self-care and use whatever drugs are avail- 4. MOnSA, Yearbook of Public Health and Social Statistics, 1974 (Seoul, 1974) Pp. 180-88. 5. Bank of Korea (BoK), Economic Statistics Yearbook, 1975 (Seoul 1975). 142 PART TWO: HUMAN RESOURCES able. The poor accessibility to modern health facilities, particularly in rural areas, boosts costs even higher because of the expense of transport and the earnings sacrificed while seeking care. Major Issues of Health Policy The foregoing description of health conditions and services in Korea produces a fairly consistent set of implications. Improvements in en- vironmental sanitation, nutrition, immunization, and basic health care would eliminate much of the pathology that now prevails. Reorganiza- tion and redistribution of health care resources would broaden access and improve treatment, even without committing additional resources to the health sector. Furthermore the continuing rapid economic devel- opment and parallel rises in incomes will heighten the intolerance to disease that can be treated or avoided. As a result, the ratio of health expenditure to GNP will continue to increase. The policies put into effect in anticipation of the growth in private demand for health services will influence the productivity of future private and public outlays. Clearly, increasing the total supply of health care resources is desirable and economically justifiable. But because the problem of inappropriate and delayed selection of services already is serious, attention should also be given to health education, patient edu- cation, and the identification of persons needing professional care. Al- though such efforts are unlikely to produce profound short-term effects, they will nevertheless become important determinants of the efficiency with which health services are used. Korea has already initiated programs and activities which are certain to have a major impact on future health conditions. The national pro- gram to improve water supplies probably is the most important health activity under way. In the fourth plan period government intends to spend W180 billion to develop supplies of piped water; this figure compares with W61 billion spent in the third plan period. The program should increase the population served to 60 percent by 1981. In addi- tion, the New Community movement is installing or upgrading village water supplies and waste disposal systems at a cost of W47 billion. By 1981 all urban areas are to be fully served by sanitary sewerage systems; in 1975 only about one-third of the urban population was served. Also during the fourth plan period, programs of health manpower education will significantly increase the stock of qualified providers of SOCIAL DiEVELOPMENT 143 care (table 5.4). The supply of physicians is to increase by 57 percent, that of nurses by 177 percent, and that of dentists by 66 percent. The supply of pharmacists is to grow by only 26 percent. The implications of these increases for the per capita availability of health manpower are considerable. The amount of physician's time available for every Korean will increase by about 40 percent during the plan period; the amount of nurse's time will increase by nearly 150 percent. Future economic and demographic trends will also tend to mitigate health problems. Higher per capita incomes, which are projected to rise by 40 percent during 1977-81, may be expected to lead to improved housing and nutrition. These improvements, in turn, should reduce the threats of respiratory and airborne diseases and of nutritional stress, but the magnitude of these changes is impossible to predict. General improvements in health conditions will be partially offset by the effects of urbanization and industrialization on the quality of air and water. Pollution is already recognized to be serious in Seoul and Busan. It will become more serious unless strong measures are introduced, particularly to control motor vehicle emissions and the use of coal briquettes for heating and cooking. The rapid development of the economy, in addition to continuing urbanization and broader education, will contribute to the formation of more sophisticated preferences for health care. This growth in de- mand will strain the capacity of the health care system and may intensify the already serious problems associated with having services concentrated in wealthier urban areas. The expected growth in expenditure and Table 5.4. Supply of Health Manpower, Projections for 1974-81 Year Physicians Nurses Dentists Pharmacists 1974 9,964 10,298 1,798 14,024 1975 10,682 12,562 1,912 14,535 1976 11,492 15,284 2,033 15,040 1977 12,183 18,062 2,161 15,538 1978 13,072 20,776 2,375 16,272 1979 13,947 23,428 2,584 16,751 1980 14,810 26,019 2,788 17,224 1981 15,659 28,550 2,988 17,692 Source: Adapted from Kong-Kyun Ro, "Analyses of Health Resources in Korea," prepared under contract for the U.S. Agency for International Development (Seoul, 1975; processed). 144 PART TWO: HUMAN RESOURCES services in urban areas will reinforce differences in access to care and produce mounting political pressure to improve health care for the poor. Nevertheless manpower training programs will expand the total supply of manpower, weighted by salaries by profession, by only about 45 percent during a five-year period in which the expenditure on health is expected to double. Health insurance Government has begun implementation of a program of compulsory health insurance for firms with more than 500 employees and for teachers and government workers. These programs will eventually cover about one-fifth of the population. The aim is to improve access to health care for those who cannot now afford it. Because manpower resources will be expanding less rapidly than private demand during the fourth plan period, the effect of such a scheme might be to increase nominal expenditure by nearly the full amount of the program. The danger is that an insurance scheme would simply redistribute health resources, not expand their supply, and cause substantial price inflation. Any in- surance program to reduce economic barriers to health care must there- fore be accompanied by a scheme to increase the supply of services and to coordinate supply and demand. The mechanisms now being considered by government to finance health care must, as a matter of political ideology, preserve the private delivery of services. The introduction of a national scheme of health insurance thus raises a large number of complex issues. * Determining the intent of the scheme: Is insurance to redistribute real income, or merely to pool risks within an income class? Is the insurance mechanism to influence the allocation and geographic distribution of real resources? * Devising mechanisms to provide economic discipline: How is pa- tient utilization to be monitored? What constraints are to be im- posed on the use of costly technology? How are the quality of services and the legitimacy of claims to be audited? What incentives are to be introduced to assure an adequate supply of inputs? * Avoiding problems of excess demand: How will expansion of de- mand and supply be synchronized? Who will be included first and last under the evolving insurance scheme? How will introduction of insurance be related to the problem of geographic inequities in the supply of care? SOCIAL DEVELOPMENT 145 Developing a system for administration: Will insurance be pro- vided by private or public institutions? Will premiums be based upon claims experiences for subgroups, or will there be deliberate cross-subsidization? Will providers of care be compensated on a fee-for-service, episodic, or year-of-care basis? Or will they be em- ployed by health care institutions which are themselves financed by membership fees? The answers to these questions will determine the financial barriers to health care and influence the quality, price, supply, organization, and geographic distribution of care. ACCESS TO CARE. The major political and economic issue in health policy is access. Economic and geographic barriers exclude many Koreans from the health care system. The cost of care under the private system is comparable to that in economically advanced countries. Many families must now borrow money at high interest rates in order to obtain ad- mission to hospitals and clinics for care. In addition, transport to health facilities is inconvenient and expensive relative to rural incomes. Prob- lems of motivating providers to live in rural areas further impede access to health care. The choice of location of practice by physicians is not well understood, but several factors are known to militate against their choosing a rural location-isolation from colleagues, heavy case loads and responsibility, poor schools for their children, and smaller financial rewards. A health insurance scheme could be used to equalize net com- pensation of health workers. For example, higher payments might be made to persons working in isolated areas. But such a scheme would remedy only the financial disadvantages of rural living, not the others. It also seems that merely expanding the supply of traditional providers offers no solution. QUALITY OF SERvICEs. The quality of care already is a problem and is certain to become worse as access to care increases. The underlying problem, true for any country, is that only a minority of patients is sufficiently informed to choose providers and levels of services rationally. This ignorance places patients at the mercy of providers. Just as signifi- cantly, it affects the behavior associated with seeking care. Particularly in Korea, where scientific notions of western medicine are only partially accepted, the problem of underreliance on preventive medicine and early treatment persists. The economic efficiency of health care is largely determined by patterns of use. Furthermore the cost and quality of care are inter- 146 PART TWO: HUMAN RESOURCES related. Providers of care are likely to define quality as it relates to the care of individual patients; they lavish resources on the patient until the marginal benefit vanishes. The larger perspective of society would argue for allocating resources in a way that equates marginal costs and benefits. Unless reimbursements made through an insurance mechanism are care- fully designed, providers of care will be able to pursue benefits without regard for costs. They would have no reason to be concerned about the patient's capacity to pay. Several mechanisms can be used to ensure the quality of health care. Although none of them will fully resolve the problem, each may be expected to contribute to improving the situation. The insurance scheme can be structured to reimburse a smaller fraction of charges for care for persons who neglect preventive measures. More specifically, patients might be required to make a copavment if they become victims of a disease for which they should have sought immunization or if they have not submitted to recommended periodic examinations. The Koje Health Plan introduced this device with some success; elsewhere, prepaid health schemes have required periodic mass-screening for asymptomatic dis- ease. Another economic incentive to cost-effective use is copayment, which requires that a percentage of all charges be borne by the patient. But the effectiveness of this device depends on the medical knowledge possessed by the patient population. Some control over the cost and quality of services rendered by pro- viders may be obtained by limiting and monitoring access to inputs essential to particular aspects of care. Under a system of health insur- ance and private delivery of care, supplies of real inputs should be planned. By limiting the availability of selected inputs-such as hospital beds, technicians, supplies, and drugs, which facilitate particular types of care-providers may be forced to make allocative decisions. Certifi- cates of need and public convenience might be required before new health care facilities are built or existing facilities are substantially modified. Another way to upgrade quality is to monitor the qualifica- tions and capabilities of providers and institutions, defining these by law in order to minimize incompetence. Korea's system of registration and licensure now is permissive. Limited licensure, under which the range of functions a provider can perform is clearly defined, might be considered. Periodic reexamination of skills might be introduced to protect patients and to motivate providers to continue their education. These initiatives could be enforced by introducing legal sanctions, re- quiring clear advertising to prospective patients of the limits of a prac- SOCIAL DEVElLOPMENT I47 tice, and compensating providers only for services they are certified to perform. Control over the quality of care could also be obtained through direct review of services. Korean patterns of use of a particular facility could be compared with established norms in order to identify such gross problems as excessive surgery or revisits for a given class of diseases. This review could be easily appended to the billing and reimbursement system of a national insurance scheme, and Korean norms could be established from experience. In addition, samples of patient records could be periodically examined, either by colleagues or by government- appointed auditors, to determine whether the procedures of care are plausible. Such over-the-shoulder supervision has been attempted else- where, but it has proven difficult to impose enough surveillance to com- pensate for the loss of the individual provider's sense of professional responsibility. Consequently the mechanisms to assure quality should not be presented as management schemes. CONTAINING COSTS. Total expenditure for health services would almost certainly grow spectacularly with the introduction of national health insurance. Some costs could be contained by phasing the implementa- tion to control excess demand. But even with phased expansion, the problem is likely to persist. Imposing an official schedule of charges to control prices may seem to be a solution, but because providers could characterize services rendered however they wish, the likelihood is that services would be itemized in excessive detail to raise the actual price of care. Two attractive alternatives to reimbursing for specific services would be to compensate providers for episodes of illness or person-years of comprehensive care. These arrangements are relatively unambiguous and are not amenable to excessive itemization. In addition, by paying for output-oriented items, not for inputs, these forms of payment shift the locus of technical decisionmaking to the professional provider. Rather than emphasize the sale of services to unqualified consumers, these units of reimbursement would place the responsibility for select- ing effective modes of care on the providers. Health manpower The second major issue of health policy confronting Korea relates to health manpower. In order to provide a sense of the magnitude of the need for medical manpower, requirements have been calculated on the 148 PART TWO: HUMAN RESOURCES basis of rules of thumb thought to be appropriate to developing coun- tries with fairly high birth rates and young populations. In studies of personal health care in Latin America, Africa, and South Asia, a con- sensus has emerged that comprehensive care for individuals in a rela- tively young population requires about 4.5 to 5 encounters with the formal health care system annually. This parameter takes into account the program of maternal and child health, the prevention or cure of infectious and parasitic disease, and the treatment of accidents. Male adults are expected to use the system less than average; women of child- bearing age more than average. Of all encounters, approximately 70 percent are for the diagnosis and treatment of common conditions. In the majoritv of these cases the con- dition is self-limiting, not responsive to any of the tools of modern medicine, or simply requires monitoring to avoid unnecessary complica- tions. Upper respiratory infections, normal pregnancies, and macro- parasitic diseases dominate this group. An additional 20 percent repre- sent significant problems which can be effectively dealt with by a qualified medical practitioner possessing general qualifications. This class of problems includes injuries, complicated deliveries, and severe parasitic infections. The remaining 10 percent of encounters demand sophisticated practitioners and institutionalized care for successful treatment. By using these crude parameters, the adequacv of health manpower resources in Korea may be analyzed at national and provincial levels. These values suggest that the number of encounters between individ- uals and the health care system in 1973 should have been about 160 million (table 5.5). Of these, 112 million could have been resolved without the aid of a professional; about 32 million would have required professional care; and about 16 million would have required specialty care. A physician, if assisted by two persons responsible for the business operations of the practice and for the care of instruments, can comfort- ably deal with 140 patients in a five-day week. With considerable effort and a full six days, a physician can treat 200 patients in a one-week period. If it is assumed that Korean physicians see 120 patients a week for 48 weeks a year, it can be concluded that the stock of physicians in 1973 could have provided approximately 54 million encounters. These speculative calculations reveal that the Korean system of health care could provide a fairly adequate level of professional care to the nation- if it were highly organized. It thus appears that the problem of access to care may not be as serious as is sometimes described. But even without national health insurance, SOCIAL DEVELOPMENT 149 Table 5.5. Professional Capacity and Needs, by Province, Estimates for 1973 (thousands of patient encounters) Professional Level of care needed" cap2acity Profes- _minus sional Pri- Second- Ter- professional Province capacity' mary ary tic.ry needs' Seoul 24,906 19,139 5,486 2,734 16,685 Busan 5,817 6,347 1,813 906 3,097 Gyeong gi 3,594 10,709 3,059 1,529 -995 Gang weon 1,635 5,870 1,677 838 -880 Chung bug 731 4,748 1,356 678 -1,303 Chung nam 2,217 9,135 2,610 1,305 -1,697 Jeon bug 1,825 7,676 2,198 1,099 -1,472 Jeon nam 4,429 12,835 3,667 1,833 -1,071 Gyeong bug 6,278 14,767 4,219 2,109 -50 Gyeong nam 2,039 9,936 2,838 1,419 -2,219 Jeju 316 1,200 342 171 -197 Total 53,792 102,386 29,253 14,626 9,913 Note: Figures may not reconcile because of rounding. Source: World Bank estimates. a. The estimates of professional capacity assume that each doctor has 120 pa- tient encounters a week and works 48 weeks a year. b. The estimates of care needed assume that each Korean has about five en- counters a year with the health care sector. Of these encounters, 70 percent are pri- mary encounters for the diagnosis and treatment of common conditions not requir- ing a highly trained health professional; 20 percent are secondary encounters for conditions requiring a medical doctor or other highly qualified professional; the remaining 10 percent are tertiary encounters requiring specialists and institutional care. c The surplus or deficit of professional capacity in relation to the total of sec- ondary and tertiary encounters requiring a doctor or other highly qualified pro- fessional. supply is not likely to meet demand over the fourth plan period. With any scheme of public finance or prepayment, excess demand would be substantial. About one-quarter of Korean physicians are now practicing abroad, but this proportion is likely to decline. The growing surplus of physicians in Europe and the United States is likely to produce agitation for protection against competition from foreign medical gradu- ates. This pressure should reduce the outflow from Korea in the future. The weak attachment of nurses to the labor force presents a more serious challenge. It is estimated that nurses with full baccalaureate qualifications work fewer than two years after completing their training. 1 50 PART TWO: HUMAN RESOURCES Those possessing only nursing certificates practice for three or four years. This pattern is attributable to strong social pressures on women to with- draw from the labor force when they marry. Although this custom appears to be disappearing, it still is a major determinant of the supply of nurses. In addition, there now is agitation in professional nursing circles to discourage certificate-level training for nurses and to shift to four-year collegiate programs instead. If this occurs, the effect will be to reduce further the average period of service. The brief duration of service of nursing personnel has already produced a situation in which the cost of training nurses per year of service is substantially greater than that for qualified physicians. Moreover the brief period of service severely limits the professionalization of nurses through experience. There is little opportunitv for physicians to develop working relation- ships with nurses or to establish confidence in the professional judg- ments of nurses. As a result, the contributions of nurses to medical care in Korea, in contrast with other countries, are modest. This modest contribution is reflected by the exceptionally low ratio of nurses to physicians in practice and the limited role of nurses in patient care. Korea should thus give serious consideration to increasing the effi- ciency with which it uses health manpower, particularly in primary care. Efforts to increase the duration of service of qualified nurses could re- duce training costs relative to the years of service, increase the effective supply of nurses, and encourage a team approach to care. The develop- ment of a cadre of male health workers-corresponding to the assistant medical officer in Africa or the physician's associate in North America- could reduce costs, improve the use of physicians, and promote a rational team approach to care. It is likely that such health workers would be more willing to live in rural areas, particularly if trained near their future place of employment. Health authorities should also re- assess the roles of the community health workers now responsible for tuberculosis control, maternal and child health, and family planning, particularly with a view to improving care of the poor. Above all, gov- ernment should restructure the administration of the entire health sec- tor. MoHsA's lack of authority and its inability to provide leadership and coordinate health activities has perhaps been the greatest impedi- ment to the effective delivery of health services in Korea. The supply of medical manpower, which would nearly be adequate if it were highly organized, clearly is inadequate under the present system of organiza- tion. It would become even more inadequate if a national scheme of health insurance were introduced. SOCIAL DEVELOPMENT I 51 Education and Skill Formation After 1950 the growth of enrollment at all levels of education was dramatic (table 5.6). Primary schools expanded rapidly, and secondary and tertiary systems of education were built up. The six years of primary education were made compulsory in 1949, and essentially full enroll- Table 5.6. Educational Enrollment, by Level, 1950-75 Level/measure 1950 1960 1965 1970 1975 Primary Total enrollment (thousands) 2,669 3,621 4,941 5,749 5,599 Rate of increase (percent) - 3.1 6.4 3.1 -0.5 Enrollment ratio' (percent) 83.0 96.0 100.0 104.0 n.a. Proportion of females in total (percent) n.a. 45.0 46.0 48.0 48.0 Secondary Total enrollment (thousands) 436 875 1,201 1,935 3,176 Enrollment ratio' (percent) 16.0 29.0 34.0 41.0 n.a. Proportion of females in total (percent) 19.0 26.0 35.0 38.0 41.0 General secondary Total enrollment (thousands) 381 749 1,005 1,634 2,675 Rate of increase (percent) - 7.0 6.1 10.2 10.4 Vocational secondary Total enrollment (thousands) 55 126 196 301 501 Rate of increase (percent) - 8.6 9.2 9.0 10.7 Tertiary Total enrollment (thousands) 36 101 142 201 297 Rate of increase (percent) - 10.9 7.0 7.2 8.7 Proportion of females in total (percent) 11.0 17.0 25.0 24.0 27.0 - Not applicable. n.a. Not available. Sources: Ministry of Education, Statistical Year Book of Education, various years; UNESCO, Statistical Yearbook, various years. a. Enrollment ratios for 1950-70 are those calculated by UNEsco. See table 5.7 for age-specific enrollment ratios for 1975. 152 PART TWO: HUMAN REJSOURCES ment at primary level was achieved by the early 1960s. The level of human resource development at that time already was well above the level typical for countries at Korea's income level.6 Since then the main growth point has been secondary education. In the 1960s the growth of three-year middle schools was particularly fast. They now provide academic education to a large proportion of the 12-15 age group and have reached all but the most remote rural areas. In the late 1960s and early 1970s vocational school enrollments grew much faster than those of general high schools. This trend reversed in 1973. The enrollment in general high schools increased by 57.7 percent between 1973 and 1975, while that in vocational high schools grew only by 8.8 percent. Although the children born during the postwar baby boom caused much of this expansion, the growth in enrollment was much faster than the growth in cohort size. Today the differences in the size of cohorts between the ages of 6 and 18 are small (table 5.7). These small differences mean that government can now direct new educational investment to enlarging the proportion of the school-age population enrolled in school and to raising the expenditure on each pupil. Enroll- ments will begin to rise again only when the children of parents born in the baby boom enter school in the late 1980s. Literacy of those over 12, estimated to have been 22 percent in 1945, rose to about 90 percent in 1975. A major adult education campaign, begun in the late 1940s and continued after the Korean War, generated much of this improvement. Nevertheless literacy still varies by sex and place of residence. According to the 1970 census, only 2 percent of urban males 15 and over were illiterate, compared with 26.6 percent of rural females 15 and over. Census figures also show that the proportion of the economically active population without schooling dropped from 32 per- cent of males and 55 percent of females in 1960 to 16 percent and 31 percent in 1970. The proportion of males with middle school education rose over the same period from 30 percent to 49 percent; that of females, from 11 percent to 26 percent. Steady improvements in the educational attainment of the labor force have affected all occupational groups (table 5.8). 6. Harbison and Myers regressed a composite index of human resource develop- ment on GNP per capita. Korea exhibited the largest absolute deviation from the regression line, indicating its atypically high education levels for a country at its income level. Frederick Harbison and Charles A. Myers, Education, Manpower and Economic Growth: Strategies of Human Resource Development (New York: McGraw-Hill, 1964), pp. 42-43. SOCIAL DEVELOPMENT 1 53 Table 5.7. Enrollment Ratios, by Age and Sex, 1975 Males Females Enrollment Enrollment Population' ratiab Population' ratiob Age (thousands) (percent) (thousands) (percent) 6 469 87.0 440 85.6 7 456 97.6 430 97.0 8 453 100.8 422 102.0 9 456 100.3 424 100.6 10 453 103.6 420 102.8 11 469 102.2 432 103.5 12 488 69.7 451 64.3 13 483 83.7 449 68.0 14 490 82.5 456 66.9 15 497 66.9 462 52.3 16 474 54.1 440 40.2 17 446 46.5 416 33.3 18 420 32.7 394 18.3 19 400 17.9 380 8.8 20 381 13.7 361 6.6 21 337 11.4 316 5.2 22 310 8.1 290 3.2 23 288 5.9 279 1.3 24 251 3.9 255 0.5 Note: Enrollment ratios were not computed for those aged 25 and over. About 28,000 males and 2,150 females aged 25 and over were enrolled in a program of education. Sources: EPB, "Korean Population Projection, 1975-85" (Seoul, n.d.); Ministry of Education, Statistical Year Book of Education, 1975. a. Mid-year estimates based on preliminary census data. No adjust- ment was made for underenumeration, which may explain the apparent discrepancies between the population estimates and enrollment ratios for those of primary-school age. b. Computed for students enrolled on April 1. Girls on average still receive less education than boys, but the gap is narrowing. The larger number of boys in primary school was for years a reflection of the sex structure of the population, not of social factors. But the discrepancy remains for those over 12. Approximately two-thirds of girls now receive a middle-school education, compared with four- fifths of boys, and their advancement rate to higher secondary school is slightly lower. The importance of educating more girls goes beyond upgrading the quality of the labor force. Empirical evidence suggests that it also is likely to be a major determinant of fertility ideals and the propensity to adopt family planning and health practices. 154 PART TWO: HUMAN RESOURCES Table 5.8. Educational Attainment of the Employed Population, by Occupation, 1960-70 (percent) Professional and technical Administrative ClericaZ Sales Year/level workers workers workers workers 1960 No education 3.2 14.8 1.3 33.5 Primary education 9.3 36.7 20.8 44.0 Secondary education 53.9 37.0 58.0 19.9 Higher education 33.5 11.4 19.8 2.4 Unknown 0.2 0.4 0.2 0.4 1970 No education 1.3 1.3 0.4 13.3 Primary education 5.5 14.1 9.3 39.6 Secondary education 37.7 44.9 58.8 40.0 Higher education 55.6 39.9 37.6 7.3 Unknown 0.1 Zero or negligible. Sources: EPB, 1960 Population and Housing Census Report (Seoul, n.d.); EPB, 1970 Population and Housing Census Report (Seoul, 1973). Formal education Enrollment rates in Korea for secondary school and university now are very high. The proportion of 17-year-olds in school, although behind that in Japan, exceeds that in the United Kingdom and is close to that in several other European countries. Enrollment rates at university level are lower than those in most developed countries, but they still are high by developing-country standards. The growth of enrollment in general high schools is perhaps the most striking, but junior technical (vocational) colleges and graduate schools are also growing very fast. The grovth of enrollment in other secondary institutions and colleges and universities continues to be steady (table 5.9). The table also illus- trates the diversity of the educational system. Nevertheless several types of school are of minor importance, and some are on the verge of disappearing. One reason Korea has managed to afford high levels of enrollment relative to per capita income is that it has kept educational costs per student low. Dropout and repetition rates in Korea, compared with SOCIAL DEVELOPMENT 1 55 Agricultural Production Service wtorkers and workers workers fishermen All orkers Year/Ilvel 1960 26.2 22.9 55.5 45.5 No education 54.0 47.3 37.0 39.4 Primary education 18.5 25.7 6.7 13.4 Secondary education 1.1 3.9 0.5 2.4 Higher education 0.4 0.4 0.5 0.4 Unknown 1970 9.0 10.0 38.9 23.8 No education 48.1 50.7 49.1 43.7 Primary education 40.0 35.6 11.5 26.5 Secondary education 3.0 3.9 0.6 6.2 Higher education 0.1 0.1 0.1 0.1 Unknown many developing countries, are low. More than 90 percent of entrants to primary, middle, and secondary schools graduate without repetition. Class sizes are very large by international standards. Despite the low costs per student, education is so extensive that the proportion of GNP devoted to education is higher in Korea than in many developing coun- tries. Although not all sources agree on aggregate educational expendi- ture, particularly with respect to private expenditure, it has been esti- mated that between 5 and 6 percent of GNP is devoted to education. This figure excludes out-of-school expenditure. Public expenditure on educa- tion has recently been only about 3 percent of GNP and, as a proportion of total costs, is highest at primary and tertiary levels. Household expenditure on fees and out-of-school expenses appears in recent years to have increased relative to total household expenditure. In cities with more than 50,000 inhabitants, educational expenditure rose from 4.8 percent of total expenditure in 1966 to 8.1 percent in 1973. For farm households the rise was from 5.8 percent to 7.1 percent. Educa- tional expenditure in Seoul reached 9.2 percent of household expendi- 156 PART TWO: HUMAN RESOURCES Table 5.9. Enrollment in Public and Private Institutions, 1965-75 Annual 1965 1975 growth rate Public Private Public Private Public Private Type of institution (thousands) (thousands) (percent) Primary school 4,917 25 5,529 70 1.2 10.9 Middle school 418 333 1,204 823 11.1 9.5 General high school 105 149 257 391 9.1 10.4 Vocational high school 37 26 226 249 19.9 25.4 Junior technical vocational school 4 3 19 40 16.1 27.8 College, university, andothertertiary' 35 99 70 169 7.1 5.5 Other 12 84 8 150 -4.3 5.9 Totalb 5,528 720 7,312 1,892 2.8 10.1 Sources: Ministry of Education, Annual Survey of Education, 1965 (Seoul, 1965); idem, Statistical Year Book of Education, 1975. a. Includes junior colleges, junior teachers' colleges, four-year colleges and uni- versities, graduate schools, miscellaneous undergradute institutions, and nurse training courses. b. Excludes preprimary institutions. ture in 1973. Only primary education is tuition-free.7 At other levels, tuition and other fees now have been equalized at private and public schools. There is some local variation, and a small proportion of students are exempted in full or part. But for most students the fees are substantial: in 1975 annual fees reached almost W60,000 ($124) at middle schools and W83,700 ($171) at high schools.8 This evident wiillingness to pay high fees reflects the traditional respect for education in Korean culture and the appreciation that private rates of return from education have until recently been high. One in- vestigator calculated cross-sectional social rates of return for different levels of schooling using 1971 data for educational costs and earnings 7. A parental contribution to a fund operated by a local parent-teacher associa- tion is required for expanding facilities and school needs even at this level; the fourth plan proposes to eliminate this. 8. These are legal maxima; legal minima were W37,000 and W61,000 re- spectively. SOCIAL DEVELOPMENT 157 differences.9 The social rate of return averaged 8.2 percent for middle school, 14.6 percent for high school, and 9.3 percent for college; the rate of return was not calculated for primary education. Because some costs were publicly borne, private returns would have been higher. Although current earnings differences may be a misleading guide to future differ- ences, they are the only indication parents have of the economic value of sending their children to higher levels of education. Thus it is not surprising to find that high school enrollments have been rapidly ex- panding in recent years and that competition to obtain coveted places at higher levels of education has been considerable. The reliance on fees for much educational finance has encouraged private institutions to respond to the rising demand for education. Pri- vate schools are of negligible significance at primary level, but become increasingly important at each higher level of education. Private voca- tional training institutions have outstripped even the extremely rapid growth of their public counterparts. The characteristics of public and private schools appear to be much the same. Rapid growth and low costs perhaps suggest that the quality of educa- tion has been sacrificed. Classes are large, and learning depends heavily on rote memorization, at least at the lower levels. Obviously such a system cannot provide individual attention or opportunities for personal creativity that many people regard as inseparable from education of high quality. In relation to less exalted objectives, however, the system per- forms efficiently. Teachers appear to manage the large classes without excessive problems, disciplinary or otherwise. It can be argued, more- over, that the authority, discipline, and routine fostered by the educa- tional system are valuable traits in an industrial setting. Excessive competition among students has been another threat to the quality of the educational process, especially where advancement from one educational level to the next has depended on admissions examina- tions. Before 1968 entrance to middle schools was based on examination results. A small number of middle schools appeared to offer much greater promise of eventual admission to the best high schools, which in tum offered better prospects of acceptance by the most prestigious university departments. In consequence, impending examinations distorted the educational objectives of the last two years of elementary school. To 9. Chang Young Jeong, "Rates of Return on Investment in Education: The Case of Korea," Korea Development Institute Working Paper, no. 7408 (Seoul, 1974). 1 58 PART TWO: HUMAN RESOURCES remove these distortions and counter elitism, government abolished the examination for entrance to middle schools in Seoul in 1968 and that to all middle schools by 1970. Children in each school district are now assigned by lottery to an individual school, whether public or private. In addition, the fee differences between public and private schools were eliminated. Skill formation Korea has been remarkably successful in developing a stock of skills to meet the needs of a development process whose course was far from predictable when rapid industrialization started. A diverse array of training mechanisms-private and public, formal and nonformal, insti- tutional and on-the-job-were developed to meet growing needs for skills, particularly in manufacturing. Although there are a few trade schools at middle level, vocational training becomes important only at senior secondary level. Vocational high schools, which include agricultural, commercial, and technical high schools, account for more than 40 per- cent of enrollment in secondary schools.10 These schools grew at the exceptionally fast average rate of 21.1 percent a year during the late 1960s, and the growth rates were higher for private schools than public' schools. Not all students in vocational high schools receive a primarily vocational course. Some follow a curriculum identical with that in academic high schools. Vocational training at senior secondary level also includes several trade schools run by the Ministry of Education and a variety of non- formal training institutions, including the vocational training institutes (vrIs) administered by the Office of Labor Affairs. In 1974 approxi- mately 46,000 trainees were trained at 166 centers, nearly half of which were in plants. Since January 1975 firms with more than 200 employees have been legally obligated to give training to 15 percent of the total number of employees. The Office of Labor Affairs and the Office of Rural Development also provide some nonformal agricultural education. Vocational training continues at junior technical colleges, which usually are classified as tertiary institutions but overlap with vocational high schools to a considerable degree. Some have lower levels which admit middle school graduates after they pass an entrance examination. Junior technical colleges continue to be one of the fastest growing components of the education system. Early in the 1970s a study suggested that only 20 percent of vocational 10. Ministry of Education, Statistical Year Book of Education, various issues. SOCIAL DEVELOPMENT 159 school graduates received jobs relevant to their training, that 30 percent received less relevant jobs, that 10 percent were unemployed at the time of the study, and that 40 percent went on to higher education or mili- tary service." These findings generated concern about unemployment of the graduates of the vocational schools and criticism of the lack of coordination between the Ministry of Education, which regulates intake quotas for these schools, and the business and industrial community. It is not clear to what extent the much slower growth of vocational schools in recent years is the result of changes in policy or of market forces that suggest to students and their parents that academic high schools might be a better bet. The Ministry of Education establishes admission quotas to determine the size of tertiary institutions and individual departments. This prac- tice makes it possible to encourage the growth of some disciplines and discourage the growth of others. Engineering students accounted for a major share of expanded enrollment, although to a much greater extent during 1965-70 than recently (table 5.10). Education and engineering students represented 64 percent of the total increase during 1965-70 and 43 percent during 1970-75. The recent spurt in enrollment in fine arts and agricultural sciences and the absence of expansion in the natural sciences are rather surprising. But the extent to which the apparent technological bias in the development of tertiarv education should be credited to market forces, to decisions of educational administrators, or to student preferences has not been explored. Some of the most recent growth of college enrollments mav reflect the very rapid increase in cohort size after the Korean War. Even though the growth rate of the university-age cohort will be falling in the late 1970s and early 1980s, it will be offset by the imminent and planned growth in the number of graduates from academic high schools. The traditional objective of these schools has been to prepare students for higher education. Because high school graduates already appear to have difficulty finding jobs without first obtaining industrial or other train- ing, the proportion vho normally seek college or university entrance is high. Consequently the pressure for the further expansion of higher education is likely to increase. The proportion of disappointed applicants for places in tertiary institutions rose from 62 percent in 1970 to 74 percent in 1975.12 More than one-third of each year's applicants re- apply in subsequent years. 11. World Bank, Korea: Review of Education and Training (Washington, D.C., n.d.), p. 30. 12. Statistical Year Book of Education, various issues. 16o PART TWO: HUMAN RESOURCES Table 5.10. Enrollment in Tertiary Institutions, by Field of Study, 1965-75 Per- Annual cent- Enrollment growth rate age 1965 1970 1975 1965-70 1970-75 of total Field of study (students) (percent) 1975 Humanities 21,674 19,686 23,529 -1.9 3.6 7.9 Education 12,339 27,932 45,212 17.8 10.1 15.2 Fine arts 7,567 7,746 12,351 0.5 9.8 4.2 Law 6,789 6,006 6,218 -2.4 0.7 2.1 Sodal sciences 27,135 36,821 37,247 6.3 0.2 12.5 Natural sciences 12,495 17,090 17,022 6.5 -0.1 5.7 Engineering 26,929 49,784 73,743 13.1 8.2 24.8 Medical sciences 12,256 19,332 29,248 9.5 8.6 9.8 Agricultural sciences 14,452 17,039 27,436 3.3 10.0 9.2 General - - 25,213 - - 8.5 Total 141,636 201,436 297,219 7.3 8.1 100.0 - Not applicable. Note: Institutions of tertiary education include junior technical (vocational) colleges. Source: Ministry of Education, Statistical Year Book of Education, various years. Because four-year colleges and universities have been growing at the expense of junior colleges and junior teachers' colleges, some part of the growth in numbers enrolled reflects the replacement of two-year courses by four-year courses. This replacement is likely to be particularly important in the remarkable expansion in the number of education students and is permitting a significant upgrading in the educational attainment of teachers. There already have been reports of a surplus of primary school teachers. Because the size of the primary school cohort has leveled off, this surplus is likely to increase. Major Issues of Educational Policy Korea now is at an interesting stage of educational development. A diverse and extensive network of institutions and mechanisms has been established. Enrollments at all ages have grown much more rapidly than population. This growth, in turn, has raised the education and skills of the population to much higher average levels than in the vast majority SOCIAL DEVELOPMENT I6I of developing countries. Despite all that has been achieved, however, the situation appears far from satisfactory to many Koreans. There are complaints about the high burden of fees, the large size of classes, particularly at primary level in Seoul and Busan, and the poor quality of facilities. The demand for places, particularly in university, exceeds the supply. Finally, doubts about the ability of the education and train- ing system to meet future manpower needs have been widely expressed. These things combined make it difficult to select priorities. The proposals for education in the fourth plan are ambitious. Re- current expenditure of the Ministry of Education is to rise from 3.1 percent of GNP in 1976 to 5.1 percent in 1981. The proposed salary increase for teachers is the largest element in this rise. Salaries for teachers in government primary, middle, and high schools account for about two-thirds of the projected recurrent budget for education. Teachers' salaries thus are of considerable importance to the entire government budget. The proposed increase appears generous. Teachers now are not among the government employees whose salaries are sig- nificantly below those for comparable employment in the private sector. Teachers are not in short supply, except for vocational schools, and there appears to be an oversupply of primary school teachers. The proposed wage increase could jeopardize the viability of private schools. Some administrators of private schools are already complaining about the inadequacy of revenue to pay existing salaries. Thus the private investment expected in secondary schools may not occur. Gov- ernment may be forced to raise tuition fees substantially, despite its desire to reduce the burden of private expenditure, or to subsidize private high schools. Pressure to allow private schools to reinstate examinations and charge higher fees already exists and will grow. To accede to this pressure would be a step back from the remarkably egalitarian system that now exists. The matter is important enough to justify further analysis. Investment outlays in the fourth plan, including the training expen- diture of the Office of Labor Affairs, are to be nearly triple those in the third plan and almost double their share of total outlays (table 5.11). The main policy issues raised by these outlays fall into three categories: the size and financing of the formal educational system; the adequacy of the skill formation; and the quality of education. Size and financing of formal education To what extent should enrollment at various levels of education be expanded still further? What proportions of educational costs should be I62 PART TWO: HUMAN RESOURCES Table 5.11. Investment Program for Education and Manpower Development, 1977-81 Percentage of private Proposed outlay investment Type of institution or training (billions of won) in total Primary school 139.2 n.a. Middle school 114.2 14.4 General high school 58.5 62.1 Vocational high school 185.0 15.8 Junior technical vocational college 102.8 20.9 Junior teachers' college 2.4 n.a. College and university 74.9 n.a. Other, including educational broadcasting 24.1 n.a. Vocational training by Office of Labor Affairs 41.6 n.a. Employment security and industrial safety' 6.0 n.a. Total 748.7 13.8 n.a. Not available. Note: All figures are in 1975 prices. Source: aPB, Fourth Plan. a. Includes labor exchanges and other informational services. borne publicly and privately? These questions are related. The higher the school fees, the lower will be the demand for education. But to rely on fees to ration scarce places works against egalitarian objectives and, in certain circumstances, might lead to underinvestment in education, though this does not appear to be a danger in Korea. MIDDLE SCHOOLS. Government has a long-standing commitment to move toward compulsory, and basically free, middle school education. Considerations of equity support this commitment: equality of oppor- tunity and reduced earnings differences are justifiable objectives. Con- siderations of economic efficiency probably support this commitment as well, but they are less easy to establish. True, the near universality of primary education achieved by the early 1960s provided an abundant pool of easily trainable, low-cost workers for labor-intensive industries. These workers made it possible for exports of manufactured goods to be- come the engine of economic development. More recently, with the rate of advancement from primary to middle school approaching 80 percent, employers have been able to find workers with even better basic educa- tion. It is difficult to establish, however, the extent to which high levels SOCLAL DEVELOPMENT }63 of educational attainment were essential for Korea's rapid economic progress. In the fourth plan period government is attempting to move closer to the objective of free and compulsory nine-year education, not to reach it. Fees of all sorts are to be eliminated at primary level. For middle school, the target is 90 percent enrollment in 1981, up from 76 percent officially estimated in 1975. All of the increase is to be in the public sector. By aiming only for 90 percent enrollment, government avoids the need to make major changes in the system of financing. Compulsory education would constitutionally have to be free. Even the 90 percent target is ambitious by the standards of developing countries. SENIOR SECONDARY SCHOOLS. Despite the increased enrollments in middle school, the advancement rate from middle to high school is expected to rise from 75 percent in 1975 to nearly 80 percent in 1981. Whereas 32.3 percent of those aged 18 were graduates of high school in 1975, this proportion will increase by about 10 percent a year and reach roughly 50 percent in 1981. This prospect gives cause for some concern. Social retums to high school education were relatively large in 1970, but this may no longer be the case. Although there is little statistical evidence, the employment prospects for those who stop their education after general high school are widely thought to be limited. The considerable pressure parents exert on children to enter high school may not reflect a realistic appraisal of future labor market prospects relative to alternative returns from the resources invested. This pressure may instead reflect an appraisal of past labor markets and the desire to provide children with a relatively favorable standing in the labor market. The result in Korea, as elsewhere, is likely to be that the formal qualifications expected of applicants for all jobs will steadily increase and that the opportunity cost of remaining in school will fall correspond- ingly. Students reaching their target level of attainment will find their ex- pectations disappointed; increasing proportions will try to enter the next highest level. This phenomenon explains the growing number of un- successful college applicants who stay out of the regular labor market and reapply for admission the following year-a pattern that has elicited considerable official concern. The proposed addition of vocational classes, primarily in office skills, to the high school curriculum may help some- what, but it is unlikely to eliminate growing pressure for college en- trance. To the extent that the growth of senior high schools is privately financed, the reason for concern is diminished. Market forces may even correct this growth naturally. But because substantial public expenditure 164 PART TWO: HUMAN RESOURCES is involved, particularly for vocational high schools, the cost-effectiveness of this expenditure is compromised. One way to increase its cost- effectiveness would be to expand and diversify the vocational training system run by the Office of Labor Affairs. COLLEGES AND UNIVERSITIES. It is planned that the growth of tertiary education will continue to be fast and that the most rapidly growing institutions will continue to be junior technical colleges. Private institu- tions, with 68 percent of enrollment, now dominate this sector. But because of heavy public expenditure during the fourth plan period, their share will drop to 58.7 percent in 1981. The intake of two-year institu- tions was less than half that of four-year colleges and universities in 1976 and will nearly double during the plan period. Enrollment in four-year institutions, in contrast, will grow by onlv 2.5 percent a year, primarily at local colleges. The economic case for continued fast growth at this level of education is weak. Korea already has high levels of en- rollment, and the returns to college and universitv education are rela- tively low. The advanced technicians who will emerge with a back- ground of completed general high school and junior vocational college could no doubt be produced more cost-effectively. But given the pressure from the largely self-financed orowth of general high schools, the expan- sion of two-year institutions may be the best wav to handle the mounting pressure for advanced training. Adequacy of skill formation Although many Korean planners are worried that critical bottlenecks in skills may arise, the fourth plan optimistically forecasts potential surpluses at most training levels. It must nevertheless be recognized that in no countrv, advanced or developing, have methods of manpower fore- casting been developed that can be regarded as particularlv reliable or as a satisfactory guide to educational planning.'3 Too manv things can change: the rate and pattern of economic growth may turn out to be different from those expected; new technologies may be invented; trained persons may be able to apply their skills in different occupa- tions. For Korea, which is undergoing rapid structural changes, the task is even more difficult. Fortunately the possibilities for skill substitution 13. Bashir Ahamad and Mark Blaug (eds.), The Practice of Manpower Fore- casting (New York: Elsevier Scientific Publishing Company, 1973). SOCIAL DEVELOPMENT I65 that make prediction difEcult also make errors in prediction less serious. This is particularly true for a large economy that has been well served in the past by an established and diverse educational structure and a variety of educational institutions and training mechanisms. Korea's systems of training have proved responsive to changing cir- cumstances. Consequently there is good reason to believe they will in the future be equally responsive to the inevitable shifts in the pattern of demand for skills. The proportion of total employment in manufac- turing constituted by the current stock of trained manpower-craftsmen, technicians, engineers, and scientists-already is high by international standards. By 1981 Korea will be even more advanced. Another measure of the adequacy of technical manpower is the elasticity of the propor- tion of technical and professional workers in the labor force relative to changes in productivity. For the manufacturing sectors of advanced countries as a whole in 1960, it has been estimated that this elasticity was 0.58 for engineers and 0.51 for technicians. The projected elasticities for Korea in 1975-81 are 0.65 and 0.56 respectively.'4 Although there certainly is no need for alarm, and needs probably will be more exaggerated than neglected, training will have to be exten- sive during the fourth plan period. Much of this training, but by no means all, will take place on the job. One difficult question is the extent to which the remainder of the training required is best served by techni- cal high schools or by VTIs and other public training centers offering short courses. Both sorts of training institution clearly have a role. The short courses offered by the vocational training system of the Office of Labor Affairs may be more flexible in their output and more able to provide retraining in response to structural changes in industry. On the other hand, the theoretical knowledge imparted to high school graduates may make them better equipped not only to move upward to an ever- increasing level of skills, but to respond to changing needs. 14. Comparing the fourth plan's figures for 1975 and 1981 with Zymelman's figures for a number of developed countries in the early 1960s, Korea comes out well ahead for craftsmen and technicians and lags behind only the most advanced countries for engineers. The elasticity was derived by estimating bt, where log xij ai + b,, log(Q,/L). The proportion of persons with occupation i in industry j is denoted by xij. Total output in industry j is denoted by Q,; employment by L>. Zymelman calculated blj for a large number of industries and for the manu- facturing sector as a whole. Manuel Zymelman, "Productivity, Skills and Educa- tion in Manufacturing Industry," in Planning for Advanced Skills and Technol- ogies, Industrial Planning and Programming Series, no. 3 (New York: United Nations, 1969). I66 PART TWO: HUMAN RESOURCES Technical high schools are important today, but the recurrent and capital costs of other public training centers are much lower. The courses in technical high schools are much longer, and much of the students' time is spent on regular academic subjects. The recurrent cost per student-hour also appears to be substantially higher in technical high schools. As a method of training this could only be justified if the out- put of technical high schools were clearly superior to that of VTIS. This does not appear to be the case. Given the greater cost-effectiveness of vTIs, the already high levels of general education, and the continuing expansion of junior technical colleges, it would seem that giving greater emphasis to vTIs than to technical high schools is the right strategy for the present. A survey by the Korea Educational Development Institute (KEDI) found that a substantial majority of employers preferred to hire middle school graduates having one year of experience or vocational training for lower skill jobs and, to a lesser extent, for higher skill jobs.'5 Employers felt that the advantages of greater practical knowledge and immediate productivity outweighed the greater theoretical knowledge and the po- tential to acquire higher skills that characterized graduates of technical high school. The survey also found that many employers believed that graduates of academic high schools could, with a year's training in indus- try, perform as well as those of technical high schools. Those with such training earned more than the graduates of vocational high schools and still more than vTI trainees. But without this training, the graduates of academic high schools performed worse than those in the other two groups. This pattern suggests that an expanded and diversified voca- tional training system could play an important role by providing basic industrial training to graduates of academic high schools. The survey also pointed to a number of deficiencies in the present system of training. Graduates of vTIs and technical high schools com- plained that the equipment they used was different from that which they were trained to operate. The overwhelming majority of graduates of technical high schools felt they had not received adequate practical knowledge and experience. This finding suggests that technical high schools should reduce general training and increase job-related training. One problem for technical high schools and VTIs alike has been a short- 15. Seventy employers and 600 workers were interviewed. Private communica- tion to the authors from Manuel Zymelman, a World Bank staff member assisting REDI with the survey. SOCIAL DEVELOPMENT I67 age of good teachers with appropriate industrial experience. Reforms are needed to increase the industrial experience of teachers, perhaps at the expense of their academic training, and to extend to experienced craftsmen the opportunity to be trained as teachers. Quality of education At primary level the thrust of programs under the fourth plan will be to reduce the size of classes and improve schools. Because research sug- gests that reducing the size of classes is not a cost-effective way to improve the educational achievements of children, these measures may be questionable. But even classes of sixty students, the norm for most primary schools, are large by international standards, and overcrowding apparently is extreme in metropolitan schools. Criticism of middle and high school programs has focused on the inadequacy of instructional materials, the gearing of the pace of instruc- tion to the average student, and the failure to exploit modern educa- tional technology. KEDI is administering a pilot project to examine the ways these deficiencies can best be remedied. The experience gained will shape new efforts to improve the facilities for science and vocational training, to subsidize experimentation and the development of teaching materials, and to develop educational broadcasting-efforts which will be extended to junior vocational colleges as well. At the end of the fourth plan period Korea will be well placed to meet the educational needs of the future. Throughout the 1980s the educational system will benefit from the decline in fertility. The school- age population will continue to shrink until the early 1990s, when chil- dren of those born in the postwar baby boom will pass through primary and middle school. The 1980s will therefore be a decade to concentrate on improving educational quality. In some instances this will simply mean additional laboratories and equipment. But more radical changes are possible. If it is felt that Korea should reduce class sizes to those common elsewhere, or that team-teaching of even larger classes is more promising, this will be the period in which to attempt it. Naturally re- search should be conducted to determine the appropriate direction for the education and training system to follow. Whichever direction is followed, further expansion of high schools will still be necessary if the twin targets of 90 percent enrollment in middle school and 80 percent advancement from middle school are to be met, especially if dropout rates remain low. The implication of meeting these targets and having 70 to 75 percent of each cohort complete twelve years of education is i68 PART TWO: HUMAN RESOURCES that Korea would have one of the most extensive systems of formal general education anywhere. Social Development Priorities It is instructive to review briefly some of the issues associated with the fourth plan's proposed expenditure for social development in the light of four social goals-redistributing income and wealth (equity), raising future incomes (efficiency), reducing fertility, and improving the location of economic activity. Equity For many proposed schemes, at least some of the justification must be sought in their impact on the distribution of income. Moreover deci- sions about the magnitude and direction of income redistribution have to be based on political criteria, not on objective economic analysis. To illustrate, the objective of the proposed elimination of certain educa- tional fees and the increased subsidies to private middle schools is solely redistributive. Decisions about educational finance thus require a politi- cal judgment about the extent to which government should ensure the equality of opportunity for all children or insist that the social costs of children are fully borne by their parents. No attempt is made here to judge the political criteria affecting the selection of priorities, but it can be asked whether the proposed schemes are likely to achieve their equity objectives and whether they achieve these objectives at minimum social cost. Equity is closely bound to the pricing and financing of services. All developed countries, whatever their political ideology, have a variety of subsidized or free social services, arrangements for social insurance, and complex labor legislation. Some of these countries lean more toward the concept of a welfare state, in which access to certain social services is regarded as a basic right of citizenship. Such a view tends to reject any pricing policy that might lead either to restricted access for certain groups or to discrimination on the basis of income. It is argued that the social stigma attached to proving eligibility for the lower priced services discourages those in need from using them. In others of these countries, social services are seen to provide a minimum standard, below which citizens need not fall. But such services are to be provided only to those in need, not to evervbody. The current attitude of the Korean govern- ment is closer to the second view than to the first. Government is SOCIAL DEVELOPMENT I69 anxious to alleviate the problems of acute poverty, not to provide general social welfare schemes. Even if Korean goals in the long term become similar to those of a welfare state, it is clear that the measures associated with such goals are too expensive at this stage of Korea's development and inconsistent with the country's ambitious targets for economic growth. It always is politically easier to give subsidies than to take them away and to remove taxes than to levy them. Thus, whatever the field, the initial steps should be cautious and should, to the extent possible, focus on the poorest groups. Two major redistributive measures in the plan are the proposed abolition of charges for primary school and increased subsidies for middle school. Without implying that these measures are misguided or that they are not administratively efficient, it may be more equitable to give scholarships to students from poorer homes than to reduce tuition fees. Also working in favor of redistribution will be the general increase in middle school enrollment in areas other than Seoul. This increase should improve the opportunities for underprivileged groups. But raising teachers' salaries in real terms will increase the costs of private high schools and result in growing pressure to revert to en- trance examinations and different fees. Considerations of equitv suggest that these pressures should be resisted. In other sectors of social development, redistribution is likely to prove more difficult. Like most countries, Korea is finding it tempting to finance its schemes by levying payroll charges on large firms. Of all possible methods of financing such schemes, this probably is the least desirable at this stage of Korea's development. First, employers would pass most or all of the cost of such schemes to their consumers. This would raise domestic prices, which runs counter to government's objec- tive of price stability, and reduce the price competitiveness of ex- ports. Second, during a period when rapid labor absorption by the manufacturing sector must remain a major policy objective, raising labor costs will encourage employers to use inappropriate technology. Third, although it may appear that the amounts involved are not large and that the effect on prices and technological choices would not be signifi- cant, neither is certain. In some Latin American countries the social insurance charges have become a significant part of the costs of employ- ing labor. In Chile and Uruguay these charges average almost 50 per- cent of the wage bill.16 Fourth, the administrative complications of 16. Organization of American States, Guidelines for Achieving Maximum Employment and Growth in Latin America (Washington, D.C., 1973), table 51, p. 180. 170 PART TWO: HUMAN RESOURCES introducing such schemes make it necessary to start with larger firms. For example, the medical insurance scheme is starting with firms having more than 500 employees and gradually extending to smaller firms. Large firms in Korea already pay higher wages to workers of comparable skills and education than do smaller firms. Thus the benefits of such schemes, organized for workers in larger enterprises, do not go to workers most in need. Services do have to be financed. For services that require some sub- sidization, either out of general revenue or by taxing richer consumers of the same service (cross-subsidization), the case normally is strong for charging some fees for the service in order to prevent or reduce waste and abuse. Sometimes the appropriate policy may be to charge minimum fees roughly equal to short-term marginal costs and to leave fixed costs to be paid by richer consumers and the state. It should be noted that cross-subsidization has no particular economic merit. For richer consumers it involves a tax on the same service, which may or may not be thought desirable. Given Korea's relatively equal distribu- tion of income, the richer consumers may not themselves be very rich. There would thus appear to be little justification for raising high school fees to pay for scholarships or raising public hospital fees to pay for free services to the relatively poor. In some cases, however, cross- subsidization may be justified on grounds of administrative simplicity. Care must also be taken to ensure that schemes do not develop an effective demand for services at a much faster rate than they can be supplied. For example, if a medical insurance scheme is put in operation before an effective system of providing health services is developed, the scheme would raise the high price of medical care even higher. Far more urgent in the health field is the need to provide improved services for preventive health, family planning, and primary care. These services should be available in all areas at prices that all Koreans can afford. Housing is another sector in which it is proving difficult to reach the poorest members of the community. In contrast with other indicators of social development, the adequacy of the Korean stock of housing does not appear to have been improving. Numerical increases may neverthe- less mask qualitative improvements. The number of households in- creascd from 4.2 million in 1960 to 6.5 million in 1975; the number of housino units increased over the same period from 3.5 million to 4.8 million.17 Thus, although the number of housing units substantially 17. Raymond J. Struyk and Kyung Nak Song, "Korean Housing: Economic Appraisal and Policy Alternatives," Korea Development Institute Working Paper, no. 7603 (Seoul, 1976). SOCIAL DEVELOPMENT I71 increased, the ratio of housing units to households declined from 82 per- cent to 75 percent. With the proposed construction of 1.3 million housing units under the fourth plan, this ratio is expected to rise to 80 percent in 1981. The average cost of one unit of urban public housing is projected at W1.5 million ($3,100) compared with W2.5 million ($5,200) for private construction. Even at this cost, houses will be out of the reach of families in the lower deciles of income distribution. Because the beneficiaries of the public housing program will be the middle and higher income groups, any subsidization should be avoided. The same applies to various schemes being considered to improve the mortgage market and encourage private home construction and owner- ship. Efficiency, fertility, and migration In addition to attempting to determine the impact alternative pro- grams of social development will have on the distribution of income and wealth, Korean planners should evaluate the effects of such programs on efficiency, fertility, and rural-to-urban migration. It is difficult to quantify such effects, but this does not mean they should be ignored. With respect to efficiencv, the major set of expenditures designed to raise productivity, growth, and incomes is for education. An attempt is being made to raise the quality of education and to expand enrollment ratios at secondary and tertiary levels. Many parts of the educational program, especially the attempts to improve technical education and to increase middle school enrollment, may well have a reasonably high economic payoff. Nevertheless overinvestment is also a danger, partic- ularly for general high schools. Health programs may also have a high economic payoff, but this is easier to suggest than prove. Morbidity in Korea continues to be high. Its reduction would have an impact on efficiency and must be a major health objective. There is, however, no wav at present to compare the economic return from expenditure on health with that from expenditure on education. With respect to population objectives, the determinants of fertility in Korea were discussed in chapter 4, along with specific proposals de- signed to influence fertility. Programs proposed for social development also have an impact on fertility, though knowledge of the determinants of fertility is insufficient to assess the extent of this impact. Broadening middle school education may be of value in reducing fertility, especially when augmented by the program of population education now being worked out. Dropping primary school charges might have the opposite effect. The schemes themselves do not offer much opportunity to apply 172 PART TWO: HUMAN RESOURCES effective disincentives to fertility. Equity considerations argue against different school fees by birth order. Housing schemes, which might dis- criminate against large families, are still rather limited in coverage. If, however, a social insurance scheme, such as the National Welfare Pension scheme, is brought into operation, consideration should be given in its design to experiments that explore whether such a scheme can be used to reduce fertility. Government is also concerned about the impact of proposed schemes on the rates and patterns of migration and on the location of economic activity. It has repeatedly stressed the goals of limiting the growth of Seoul and decentralizing industry. There is little doubt, however, that the relatively high levels of educational facilities and social services are among the attractions of Seoul and Busan. Primary school classes may be most crowded in these cities, but advancement rates to higher levels of education are better than elsewhere and their universities are the most prestigious. Once migrants do arrive, it is difficult to resist supply- ing them with services. As a result, the two largest cities retain their attraction. To counter this trend, government did much during the third plan period to improve the amenities in rural areas, and the actions taken may have been a factor in the abnormal slowing of migration to metropolitan cities. The programs associated with this policy are to be continued during the fourth plan period. Chapter 6 Ricardo Moran Labor Availability and Employment THE POOL OF EMPLOYABLE MANPOWER in Korea, as in many less devel- oped countries, grew at historically fast and increasingly rapid rates during the 1960s and early 1970s (table 6.1). What distinguishes Korea from most of these countries is that it appears to have successfully in- corporated a growing proportion of employable persons into production at rising average real wages. This absorption entailed a shift in employ- ment toward manufacturing industries and skillintensive technologies. The progress made in education, skill formation, and health also con- tributed. Yet, despite these achievements, much room remains for im- provement. The proportion of women in remunerated activities still is low by international comparison; the productivity and conditions of emplovment of the lowest paid workers can be upgraded. The characteristics of the labor supply behavior of various groups among the working-age population are diverse. Females are particularly sensitive to labor market conditions in their decisions to seek employ- ment-their labor force participation rates (LFPRS) are positively and strongly correlated with an index of employment opportunities. In addi- tion, the combined effects of age, sex, and residence are extremely powerful in accounting for variations in labor force participation be- havior among members of the working-age population. Demographic trends indicate that the rate of growth of the working- age population recently peaked and will gradually slow down during the next two decades. In order to sustain secular improvement in aggre- gate indicators of manpower use, such as the proportion of the working- age population employed, it will be necessary for employment to continue to grow at rates of at least 3.2 percent a year until the mid- 1980s. Then this minimum critical rate of employment growth will gradually ease toward the level of 2 percent a year for the rest of the century. Fears of general or persistent bottlenecks in the labor supply in the coming years seem exaggerated, unless government introduces drastically different labor and educational policies that might weaken I73 174 PART TWO: HUMAN RESOURCES Table 6.1. Labor Force and Employment, 1963-75 Popula- Official tion Iabor force unem- 14 years participation rate Total Em- ploy- and over Aver- Fe- labor ployed ment (thousands age Male male force persons rate of (thou- (thou- (per- Year Persons) (percent) sands) sands) cent) 1963 15,085 55.3 76.4 36.3 8,343 7,662 8.1 1964 15,502 54.5 76.4 35.4 8,449 7,799 7.2 1965 15,937 55.6 76.6 36.5 8,859 8,206 7.4 1966 16,367 55.4 76.5 36.2 9,071 8,423 7.1 1967 16,764 55.4 76.0 36.8 9,295 8,717 6.2 1968 17,166 56.2 76.1 38.2 9,647 9,155 5.1 1969 17,639 56.0 76.6 37.5 9,888 9,414 4.8 1970 18,253 55.9 75.1 38.5 10,199 9,745 4.5 1971 18,984 55.5 74.2 38.5 10,542 10,066 4.5 1972 19,724 56.0 74.7 38.9 11,058 10,559 4.5 1973 20,438 56.8 73.9 40.8 11,600 11,139 4.0 1974 21,148 57.1 74.8 40.6 12,080 11,586 4.1 1975 21,833 56.5 74.5 36.9 12,340 11,830 4.1 Sources: Economic Planning Board (EPB), Monthly Statistics of Korea, May 1976 and earlier issues; Office of Labor Affairs, Yearbook of Labor Statistics, 1975, pp. 4-9. the mechanisms for skill generation and allocation that have worked so well for so long. Most of the propositions in this account of the historical and prospec- tive contours of manpower use in Korea are widely accepted. But other propositions-particularly those concerning the prospects of labor short- ages during the period of Korea's fourth plan-differ from views held by many Korean policymakers and by researchers in Korea and abroad. There are three principal reasons for some disagreement over labor market issues. First, although a great deal of data has been collected, no set of consistent data is sufficiently complete and reliable to provide an agreed description of historical trends. Instead, competing time series have been independently adjusted or derived. Operational definitions of the same terms sometimes differ from one source to the next and, in the demographic censuses, change from one census to the next. These diverse definitions make intertemporal comparisons within categories difficult. Second, views of labor economists, manpower specialists, and statisticians diverge over the way many variables related to the labor market are to be conceptualized and measured. Third, alternative assumptions conceming future labor market behavior of employers and LABOR AVAILABILITY AND EMPLOYMENT 175 potential workers have been made by the various analysts. This results in correspondingly different projections of such variables as labor force, employment, and unemployment. Because of these considerations, an attempt is made in this chapter to be particularly explicit about opera- tional definitions, underlying assumptions, data adjustments, deriva- tions, and the like. Fortunately not all the relevant variables and in- dicators are subject to the same degree of definitional ambiguity and abuse. In trying to assess the historical and prospective labor market trends in Korea, it thus seems sensible to focus on variables that appear conceptually and operationally most robust. Historical Trends and Patterns A substantial body of evidence supports the widely held and intui- tively compelling view that the Korean economy has been using avail- able manpower more fully and productively since the early 1960s, when annual rates of economic growth began to take off. The extent and rates of labor absorption nevertheless vary according to operational yard- sticks and data sources. It is practical to begin by examining the evolu- tion of a fairly straightforward, if crude, index of manpower use: the proportion of the employment-eligible population employed, or the employment ratio for short. Korean data sources distinguish the working-age population, which in Korea comprises all persons 14 and over, from the employment- eligible population, which is the working-age population minus institu- tionalized persons. The census presents the working-age population; the labor force surveys of the Economic Planning Board (EPB) present the employment-eligible population.1 In 1970 the employment-eligible population was 96.4 percent of the working-age population. If this proportion is assumed to remain constant, the growth rates of the two populations are by definition identical. In the interest of precision, how- 1. According to the operational definitions used by EPB, the employment- eligible population, misleadingly labeled as the population 14 years old and over, is the set of all individuals residing in Korea at the time of enumeration, excluding foreigners, soldiers, prisoners, and presumably such other institutionalized persons as inpatients. Employed persons constitute the subset of the eligible population reported to be working for wage or profit, plus unpaid family workers and those who had a job or business but were temporarily away from work. The period of reference was the week preceding the interview. EPB, Annual Report of the Economically Active Population, 1974 (Seoul, 1975), pp. 18, 20, 25, 39, 45. 176 PART TWO: HUMAN RESOURCES ever, the distinction between the two populations is maintained through- out this chapter. Also to be distinguished is the labor force, which comprises persons employed and persons actively seeking employment. It differs from the employment-eligible population in that it excludes eligible persons who are neither employed nor actively seeking em- ployment. The data derived from labor force surveys conducted by EPB appear to be the most reliable and internally consistent. According to this source, the employment ratio grew at a slow, steady pace from 51 percent in 1963 to 55 percent in 1974 (figure 6.1). The evolution of this index nevertheless masks certain dimensions of the progress achieved in man- power use. The eligible population grew at a rapid and accelerating pace during the entire period: at about 2.7 percent a year from 1963 to 1970 and 3.7 percent a year from 1970 to 1974 (figure 6.2). But the rapid growth rate of students among the eligible population diminished the performance of the employment ratio. Thus the student-adjusted Figure 6.1. Indexes of Manpower Use, 1963-74 Percent 70 Student-adjusted employment ratio 66 62 Employment ratio 58 54 50 10 Unemployment rate 6 2 1963 1966 1968 1970 1972 1974 Source: Table D.36 in appendix D. LABOR AVAILABILITY AND EMPLOYMENT 177 Figure 6.2. Grouwth of Selected Categories of the Working-age Population, 1963-74 Average annual rate of change (percent) 10 _ ______________ 6 4 2 _ 1963 1966 1970 1974 |i Attending school | | Employed population | Employment-eligible population Source: Table D.36 in appendix D. employment ratio gives a more accurate indication of the productive use of Korea's manpower throughout the period. It seems reasonable, after all, to consider study as an economically productive activity leading to human capital formation. The student-adjusted employment ratio rose from 58 percent in 1963 to 67 percent in 1974. In Korea, as in other developing countries for which the necessary data is available, empirical research has shown a strong statistical rela- tion of labor market conditions to the survey responses of certain classes of individuals, not employed at the time, as to whether they are "seeking work" or "inactive."2 This relation, often called the discouraged-worker 2. Or whatever other labels and operational definitions are applied to distinguish the unemployed from those not in the labor force. 178 PART TWO: HUMAN RESOURCES or additional-worker effect, makes the measured rate of unemployment imperfect as an index of aggregate manpower use, and it is not con- sidered as such an index in the following discussion.3 The demographic census is the other major primary source of nation- wide data on manpower variables.4 The census has been taken five times since 1955. Operational definitions of relevant variables have changed between censuses, as have the months in which they were conducted. There have, in addition, been some large differences between the values of certain variables in the census and those estimated by ErPB from its labor force surveys taken at approximately the same time. These in- consistencies have led to several efforts to attempt to reconcile the differences between the two sources.5 All data in the quarterly time series of the EPB surveys were recently revised, and some of the more glaring discrepancies have been reduced. Even so, many discrepancies with earlier censuses still remain. Despite the remaining discrepancies and underlying methodological problems, the estimates generated from the census essentially corroborate the view that manpower use in Korea, as measured by the employment ratio, impToved between 1960 and 1970. Three alternative series of employment ratios, derived or adjusted by three separate investigators, show that the employment ratio increased appreciably.6 3. Lee documented a strong, net discouraged-worker effect: the proportion of jobless individuals declaring themselves inactive increases as labor demand slackens and decreases as labor demand picks up. Lee also discussed the implications of these findings for the interpretation of the rate of unemployment in Korea. One such implication is that the reported rate of unemployment understates the true rate, which would incorporate discouraged workers with the unemployed. The data are not in a form that permits quantifying the underestimation or computing an adjusted unemployment rate. Kyu Sik Lee, "Labor Force Behavior in Korea" (paper read at ILCOsR Conference, February 20-24, 1974, Busan, Korea; processed). 4. Strictly, this data source is not the census, but a sample of households covered by the census. 5. The analysis and methodology developed by Lee is particularly useful. Kyu Sik Lee, "Ratio Estimation of the Labor Force: A Preliminary Study for Illustra- tion," (Seoul: Korea Development Institute, 1973; processed). 6. Kyu Sik Lee, "How Much Labor Absorption in Korea," Memorandum 7202 of the Korea Development Institute (Seoul, April 1972), table 13; Chuk Kyo Kim, "Exports and Productivity Trends in Korean Manufacturing Industries," Journal of East and West Studies, vol. 3, no. 2 (1974), table C2; Wontack Hong, "Exports and Employment in Korea," Discussion Paper Series, no. 75-04 (Quezon City, Philippines: University of Philippines, Council for Asian Manpower Studies, July 1975; processed), app. table B4. LABOR AVAILABILITY AND EMPLOYMENT 179 In addition to employment ratios, the agricultural real wage can be an important indicator of manpower use. This premise stems from the theoretical framework for the development of labor surplus economies spawned by Lewis and elaborated most prominently by Fei and Ranis.7 Fei and Ranis dealt with the relation of the agricultural real wage to manpower use in the empirical context of Korea and Taiwan for the 1955-70 period. They argue that the agricultural real wage in Korea during the mid-1950s exceeded the value of the marginal product of agricultural labor and that its level was determined by institutional factors.8 As long as the value of the marginal product of labor remained below the level of the agricultural real wage-which in their model is the essential characteristic of what they term the labor surplus condition or phase-it would have been unlikely for the agricultural real wage to rise at a pace as rapid as 5 or 6 percent a year for a period of three or more consecutive years. This would hold even if the value of the maxginal product of agricultural labor rose at such rates for such an extended time period. Thus an essentially stagnant agricultural real wage over time would indicate that the labor surplus phase has not yet been overcome. Conversely a takeoff of the agricultural real wage into a sustained growth path roughly approximating the rates of growth achieved, say, in overall output per worker would signify an end to the labor surplus. Although the authors date this takeoff point in Korea toward the end of the 1960s on the basis of their interpretation of the temporal profile of an agricultural real-wage series running from 1955 to 1969, they do so with some reservations. They question whether the rise in agricultural real wages may have been the result of a short-term deterioration of the terms of trade of the industrial sector, not the result of the end of labor surpluses. With additional data current through 1975 it was possible to examine a series of index numbers, corresponding to an extended real-wage series, in conjunction with a corresponding series of average real manufactur- ing earnings (figure 6.3). The evidence from these extended time series 7. W. Arthur Lewis, "Development with Unlimited Supplies of Labour," Manchester School of Economics and Social Studies, vol. 22, no. 2 (May 1954); John C. H. Fei and Gustav Ranis, Development of the Labor Surplus Economy: Theory and Policy (Homewood, Illinois: Richard D. Irwin, 1964). 8. The marginal product of agricultural labor is the amount by which total agricultural output would fall if one less worker were to be employed in this sector and all other things were to remain equal. In practice, this is not a direcdy observable magnitude. 180 PART TWO: HUMAN RESOURCES Figure 6.3. Annual Change in the Real Agricultural Wage and the Real Earnings in Manufacturing, 1962-75 Percent 20- 15 1- 10 5 0 11963 11965 1967 11969 1971 1 973 175 1962 1964 1966 1968 1970 1972 1974 -5 -10 - Real agricultural wage - - - Real earnings in manufacturing Source: Table D.39 in appendix D. confirms that the agricultural real wage, after an abrupt but short-lived spurt in 1963, appears to have taken off in about 1968 and sustained rapid growth rates until 1973.9 It subsequently declined by 9.5 percent 9. The derivation of these figures from the original sources was, incidentally, quite independent of the derivations presented by Fei and Ranis. John C. H. Fei and Gustav Ranis, "A Model of Growth and Employment in the Open Dualistic Economy: The Case of Korea and Taiwan," Journal of Development Studies, vol. 11, no. 32 (January 1975). LABOR AVAILABILITY AND EMPLOYMENT I8I between 1973 and 1974 and then rebounded modestly in 1975. Much of this decline is attributable to the slowdown of overall economic growth and consequently of urban demand for labor. As a result, the increased residual supply of rural labor probably depressed agricultural wages. With the recovery of the economy after 1975, it is likely that agricultural real wages also recovered and that their growth rate once again became positive. Real manufacturing wages, on the other hand, continued to increase year after year. The manufacturing wage index outperformed the agricultural real wage series in most years after 1965 and in the entire 1965-75 period. Wages in manufacturing rose almost twice as fast as those in agriculture, by 8.3 percent a year compared with 4.7 percent.10 It seems that Fei and Ranis were well advised to hedge their conclusion that the takeoff in the agricultural real wage might have been the result of other factors and a false signal of an end to the labor surplus in Korea. Thus, despite Korea's considerable progress in labor absorption since 1965, it is not fully established that the labor surplus has been entirely overcome, even when applying the rather restrictive connotation of this term adopted by Lewis and by Fei and Ranis. Projections of Labor Supply What is the range of reasonable magnitudes by which the labor supply is likely to grow during the rest of this century? What are the dimensions associated with having an increasing proportion of the working-age population in production, where increases in productivity of the lowest paid workers will be commensurate with overall output per worker? The concept of aggregate labor supply has traditionally been operationalized in empirical macroeconomic analyses and projections as the so-called labor force, which consists of those employed plus those unemployed at a particular point in time or over a short period. Because of the conceptual and methodological problems associated with the notion of unemployment and such derivative concepts as labor force 10. In contrast with this pattern, real earnings of farm workers in the United States grew at an average annual rate of 4.5 percent between 1950 and 1975, while those of manufacturing workers grew at 3.8 percent. Office of the President of the United States, Employment and Training Report of the President (Wash- ington, D.C.: U.S. Government Printing Office, 1976), pp. 161, 357-58. i 82 PART TWO: HUMAN RPESOURCES and labor force participation rates (LFPRS), the discussion in this section breaks partially with tradition by using an alternative approach to make projections of aggregate labor supply. Corresponding estimates based on conventional methodology are nevertheless given as well. Several labor force projections for Korea extend to 1981, and one extends to 2000. Six of these have been examined." With the exception of the projection by the Ministry of Science and Technology (MST), the methodologies, assumptions, and primary data sources are stated fairly explicitly. Significant differences regarding one or more of those under- lying factors exist across projections, but all share the traditional feature of having been derived without explicit regard to future labor demand or wage conditions. Despite these differences in base numbers and methodologies, the corresponding annualized rates of growth in labor force for the 1976--81 period cluster around 3.1 percent. The range is between 2.7 and 3.2 percent, and three coincide at 3.1 percent. To the extent that labor force projections derived independent of labor demand are valid, these projections can be constructively interpreted for pur- poses of policy and planning. Even though applying a broad range of reasonable estimates of initial conditions and alternative assumptions. about the future evolution of age- and sex-specific LFPRS, they imply that the Korean economy will have to generate jobs at the rate of ap- proximately 3.2 percent a year between 1977 and 1981 if the aggregate open unemployment rate is not to rise above its officially estimated 1976 level of 4.1 percent.12 The unemployment rate would therefore increase to the extent that annual rates of employment growth were lower than 3.2 percent and decrease to the extent these rates were higher. After 1981, according to the MST projection, the rate of annual growth of employment necessary to maintain a constant unemployment rate would decrease to about 2 percent by the end of the century. It 11. World Bank, "Current Economic Position and Prospects for the Republic of Korea," Report no. 332-KO, 2 vols. (Washington, D.C., February 1974; processed), vol. 2, table 1.5; Ministry of Science and Technology (MST), The Third Five-Year Manpower Development Plan, 1972-76 (Seoul, 1971), p. 54; Korea Development Institute, Korea's Economy: Past and Present (Seoul, 1975), pp. 173, 175, and 176 for methods I, j, and K respectively; EPB, The Fourth Five-Year Economic Development Plan, 1977-81 (Seoul, 1976), annex table 4, pp. 104-05. 12. The labor force participation rate (LFPR) is the proportion of employment- eligible individuals in a particular category either working or seeking work. The official estimate of unemployment was derived from figures supplied by the Korean planning team that visited Washington, D.C., in May 1976. LABOR AVAILABILITY AND fMPLOYMENT I83 Table 6.2. Manpower Availability and Deployment, Projections for 1976-81 Annual rate of change 1976- 1978- 1976- Item 1976 1978 1981 78 81 81 Employment category (thousands of persons) (percent) Eligible population' 22,751 24,175 26,123 3.0 2.6 2.8 In school 3,048b 3,585 4,478 8.0 8.0 8.0 Labor force 12,650 13,510 14,870 3.3 3.2 3.2 Employed 12,188 13,012 14,341 3.3 3.2 3.3 Unemployed 462 498 529 3.8 2.0 2.7 Ratios (percent) Labor force to eligible population 55.6 55.9 56.9 - - - Unemployed to labor force 3.7 3.7 3.6 - - - Employed to eligible population 53.6 53.8 54.9 - - - Employed to student- adjusted eligible population 61.9 63.2 66.3 - - - - Not applicable. Note: Compare with table D.36 in appendix D, in which official definitions of employment categories are presented and comparable figures are given for 1963-74. Sources: EPB, The Fourth Five-Year Economic Development Plan, 1977-81 (Seoul, 1976); World Bank projections. a. Excluding institutionalized persons. b. This figure assumes an 8 percent growth rate from the 1974 base. should be noted that historical rates of employment growth have gen- erally been 3.2 percent or higher since 1963.13 The projections of em- ployment growth generated by the macroeconomic model described in appendix A are 3.3 percent a year for the 1977-81 period (table 6.2). Because this rate of growth is approximately equivalent to the projected labor force growth rate of 3.2 percent a year, the conclusion to be drawn from these two sets of projections is that the aggregate unemployment 13. In only three of eleven year-to-year observations for 1963-74 did this rate fall below 3.1 percent. For the entire period employment grew at an average annual rate of 3.8 percent a year. See table 6.1. 184 PARTTWO: HUMAN RESOURCES rate will not change significantly during the fourth plan period if the assumptions underlying the projections materialize. The simplest way to circumvent objections to the use of labor force projections that ignore demand is to replace them with projections of the working-age population, which is insensitive to labor demand con- ditions. The corresponding policy criterion would then be that the growth rate of employment would, as a minimum, have to match the growth rate of the working-age population.14 To the extent this rate of growth falls short of the growth rate of the working-age population, the employment ratio would deteriorate over time. Conversely, to the extent this first rate exceeds the second, the employment ratio would improve. Following this method, and on the basis of the most recent population projections, it turns out that the minimum target for employ- ment growth is again on the order of 3 percent a year between 1976 and 1981. Corresponding minimum targets in the longer term would be about 2.5 percent a year between 1980 and 1985, and would then slow down to the 2 percent range for the rest of the century. It should be stressed that meeting such minimum targets implies no improvement whatever in the employment ratio, which by international standards and Korean expectations cannot be considered satisfactory, particularly for females.15 Given Korea's recent performance, however, the tasks of fulfilling or exceeding such targets do not appear awesome. Demographic characteristics and labor supply Because labor supply behavior is strongly influenced by the demo- graphic characteristics of the working-age population, some of these characteristics-particularly age and sex-are often taken into account in labor force projections. Previous work has shown that rural or urban residence, in addition to age and sex, is an important correlate of labor 14. If attending school is deemed for certain purposes to be more like working than not working, then employment would be augmented by those in school and of working age. 15. In Japan 64 percent of the population 15 years and older was employed in 1970; for females this proportion was 50 percent, compared with 40 percent in Korea in 1974. International Labour Organisation (ILO), Yearbook of Labour Statistics, 1974 (Geneva, 1974), table 1. An MST publication states: "By the year 2000 there will emerge in Korea a society in which almost all the people, except students, old people, and patients, will participate in economic activities." Envisaged is an aggregate LFPR of 65 percent in 2000, compared with 56 percent in 1976. MsT, Third Manpower Development Plan, p. 54. LABOR AVAILABILITY AND EMPLOYMENT I85 Table 6.3. Average Labor Force Participation Rates, by Age, Sex, and Place of Residence, 1963-71 (percent) Male Female Age group Rural Urban Rural Urban 14-19 50.9 33.9 37.2 33.9 20-24 85.4 70.6 48.0 44.9 25-34 92.0 95.8 51.2 24.3 35-44 92.8 97.3 57.6 33.3 45-54 89.2 92.8 50.9 31.9 55-64 73.4 65.0 35.6 18.7 65 and over 30.1 19.8 10.1 4.8 Source: Kyu Sik Lee, "Labor Force Behavior in Korea" (paper read at ILCORK Conference, February 20-24, 1974, Busan, Korea; processed), table 3. force participation in Korea.", To explore these relations, the mean LFPRS of twenty-eight population groups of working-age for the 1963-71 period were obtained by crossing seven age categories by sex and resi- dence (table 6.3). The comparisons show on average that only one-third of urban males under 20 years of age, but more than one-half of their rural counterparts were in the labor force.17 Similarly the mean LFPR for women in the 35-44 age group in urban areas is less than three- fifths of the corresponding rate for their rural counterparts. Generally, with the exception of males in primary working-age groups 25-34 for which the urban LFPR is somewhat higher, rural LFPRS are appreciably higher.'8 It has also been found that the propensity to move in and out 16. Lee, "Labor Force Behavior." 17. These differences arise in large part from the higher school enrollment ratios of working-age urban males; in 1970 these ratios were almost double those of their rural counterparts. EPB, Economically Active Population, pp. 40, 45. 18. Given the low proportion of landless agricultural workers in Korea and the correspondingly high proportion of rural family units constituting agricultural production units, it is not surprising that LFPRS for most of the fourteen age-sex groups should be higher for rural residents. Not obvious, however, is why the opposite should be true for males in the primary working ages from 25 to 44. An odd hint to this apparent paradox is that the proportion of persons whose main activity was dassified as housekeeping, which accounts for the bulk of those out of the labor force, was 15 percent for males in farm households and 2 percent for males in nonfarm household in 1970, a typical year in this period. What seems to happen in periods of slack agricultural activity is that men are working in the house, not in the field, and this gets them classified outside the labor force. 186 PART TWO: HUMAN RESOURCES of the labor force differs between certain subpopulation groups. Rural residence, for example, is associated with a relatively high propensity to shift in and out of the labor force in apparent response to labor market conditions. Without exception, the variation in LFPRS for rural age-sex groups is much higher than those for corresponding urban groups; in sev- eral cases, the coefficients are more than four times higher. As would be expected, LFPRS for males tend to be more stable than those for females. The most salient implication of these findings is that place of residence, in addition to age and sex, is an important correlate of labor supply behavior among the Korean working-age population. It should thus be taken into account in any attempt to interpret past trends or project future evolution of labor supply. Although a breakdown of LFPRs by age, sex, and place of residence is not yet available for years after 1971, yearly data from 1963 to 1974 have been compiled by sex and place of residence. These series indicate that the substantial differences between mean LFPRs for rural and urban males across age categories are not preserved when all age groups are lumped together. This suggests significant differences in the rural and urban age structures of the male working-age population. The average LFPRS for rural females have been sharply increasing-from about 41 percent in the mid-1960s to about 54 percent in the mid-1970s. This could be interpreted as a sign of impending labor scarcity in rural areas. The LFPRS for urban females have fluctuated around an average of about 30 percent. Information on time trends between 1963 and 1971 for age-, sex-, and residence-specific LFPRS indicates significant reductions for the 14-19 age group in all categories.19 These reductions are expected be- cause of rapidly rising educational enrollment rates throughout the period. Another remarkable feature is that, with the notable exception of rural females, increasing participation is reported for only three other groups: urban males 20-24 and 55-64 and urban females 55-64. The rank correlation of LFPRS between 1963 and 1971 is strong.20 These findings corroborate the widely accepted view that women account for most of the potential labor supply in Korea. These findings also highlight the less well-known feature that LFPRS are much lower for urban women than rural women. What is more important, rural 19. Lee, "Labor Force Behavior," table 4, p. 11. 20. Spearman's rank correlation coefficient is equal to 0.85 with a confidence level above 99 percent. LABOR AVAILABILITY AND EMPLOYMENT 187 women have been making steady and substantial progress in increasing their labor force participation during the past fifteen years or so, but urban women made no appreciable gains during the same period.21 Because labor force participation among urban women was found to be strongly and inversely correlated with their fertility, improving their job opportunities should be a top priority in labor market policy, if for no other reason than to facilitate attainment of government's objec- tives to reduce fertility. This seems to be the most relevant impli- cation of this empirical analysis of demographic characteristics and labor supply in Korea. Another important implication of the foregoing dis- cussion, again to the extent that labor force projections are deemed useful, is that some effort should be made to incorporate distinctions of rural and urban residence into the models generating these projections. The salient features and results of one such exercise are summarized below. Rural-urban labor force simulations As argued earlier, labor force forecasts are unreliable because of the sensitivity of age-, sex-, and residence-specific LFPRS to changes in labor demand. At the same time, projections of the employment-eligible population are insensitive to all factors influencing labor supply be- havior. Korea's future labor supply may nevertheless be further explored in simulations yielding plausible boundaries for the growth rate and size of the labor force during the 1975-2000 period. Such simulations have distinct advantages over the other two methods. By using the labor force, not the employment-eligible population, as the index of labor supply, it is possible to capture the effects of changes in the age, sex, and resi- dential composition of the working-age population. By establishing boundaries the reliability of results is less vulnerable to unpredictable changes in labor demand. A set of nationwide population projections produced by the East- 21. Although intertemporal patterns of LFPRS for females in Korea are not unique, they certainly seem atypical. Taking a sample of fifty-four countries, of which approximately one-half are LDCs, Durand found that in only one-third of them did the female share in agricultural employment increase appreciably over time. Their share remained virtually unchanged in fourteen countries and declined in twenty-three. In nonagricultural employment the female share increased appreciably in twenty-six countries; declines occurred in only fourteen. John D. Durand, The Labor Force in Economic Development (Princeton, N.J.: Princeton University Press, 1975), table 6.3, p. 130. i88 PART TWO: HUMAN RESOURCES West Center was taken as the starting point.22 These projections run from 1975 to 2000 and are regarded as the best projections now available for Korea. They disaggregate the population by age and sex, but not by rural and urban residence. To make this added distinction, the 1975 base population was divided into rural and urban categories according to the preliminary retums for the 1975 census.23 Each group was dis- aggregated by age and sex on the basis of corresponding distributions in the 1970 census. Fertility and mortality schedules for rural and urban groups were based on recent patterns; specified future population shares in each location were derived from rural-urban projections made by the United Nations. The resulting rural-urban proportions for given future years were applied to the national population projections produced by the East-West Center. Two sets of LFPRS for specific age, sex, and residence groups were applied to these projections. The first set corresponds to the mean values of LFPRS for 1963-71 and yields labor force simulation A (table 6.4). The second set is identical to the first with regard to rural males and females and urban males, but it assumes a linear increase in the LFPR of urban females from 31 percent in 1974 to 55 percent in 2000.24 This set yields labor force simulation B (table 6.5). Simulation A appears extreme because it assumes that the LFPRS of urban females will not increase from current or previous levels. As noted, these levels are quite low in comparison with rural women and with urban women in several other countries (table 6.6).25 Simulation 22. These projections are the second series prepared by Lee-Jay Cho and described in chapter 4. Lee-Jay Cho, "Projections of the Population of the Republic of Korea, 1975-2000" (Honolulu: East-West Population Institute, 1976; processed). 23. The characterization of rural and urban residence used here follows the official definition which presumably is also used in the EPB surveys and cor- responds to their terminology of farm and nonfarm households. According to this definition, urban areas are those circumscribed by the perimeter of cities having more than 50,000 inhabitants. Because the most recent demographic census is the sampling frame used in the EPB surveys, there probably is a bias toward under- enumeration of nonfarm households in view of the rapid rates of net rural-to- urban migration throughout the 1963-74 period. 24. This 1974 rate corresponds to the average of their age-specific LFPRS given in the first set, weighted by the number of women in each age category in 1974. 25. Although intemational comparisons of LFPRS are fraught with pitfalls, the differences between the orders of magnitude for Korea and selected countries reported in table 6.6 probably are not spurious. As this table also shows, LFPRS for Korean urban males, when compared with rates in other countries, are rather low as well, except for those aged between 30 and 44. LABOR AVAILABILITY AND EMPLOYMENT I89 Table 6.4. Growth of Labor Force, Simulation A, Projections for 1975-2000 (percent) Annual rate of change Category 1975-80 1980-85 1985-90 1990-95 1995-2000 Total labor force 3.3 3.1 2.5 1.8 1.4 Rural males and females 1.0 1.9 1.1 -0.6 -1.3 Urban males and females 5.6 4.0 3.5 3.3 2.8 Rural males 1.0 2.0 1.3 -0.4 -1.3 Urban males 6.1 4.7 3.6 3.2 2.8 Rural females 1.0 1.6 0.7 -0.9 -1.3 Urban females 4.4 2.3 3.2 3.6 2.9 Source: World Bank projections. A also assumes an appreciable decline in the LFPRS of rural females from recent levels of about 54 percent to their mean 1963-71 levels of about 44 percent. This projection may thus be regarded as the minimum reasonable growth of the labor force. It could conceivably be approached if the demand for labor is weak and the aggregate employment ratio falls from its current level of about 54 percent to its 1963 value of 51 percent. Simulation B may be regarded as being comparably extreme in the opposite direction. It assumes that urban females will account for a Table 6.5. Growth of Labor Force, Simulation B, Projections for 1975-2000 (percent) Annual rate of change Category 1975-80 1980-85 1985-90 1990-95 1995-2000 Total labor force 3.8 3.7 3.0 2.4 2.1 Urban males and females 6.4 5.1 4.3 4.0 3.6 Urban females 7.1 6.0 5.7 5.3 4.8 Note: The projections for rural males and females and for urban males are iden- tical with those in table 6.4 and are not presented here. Source: World Bank projections. 190 PART TWO: HUMAN RESOURCES Table 6.6. Urban Labor Force Participation Rates, by Age and Sex, Korea and Selected Countries (percent) Age group Sex/country 15-19 20-29 30-44 45-64 65 and over Males Korea 34 82 97 80 20 Japan 41 92 98 93 49 Panama 39 91 96 86 36 Sweden 48 85 96 92 20 United States 43 90 96 89 29 Females Korea 34 36 40 24 5 Japan 40 55 49 46 14 Panama 32 48 44 30 9 Sweden 42 52 47 43 5 United States 30 43 43 44 11 Note: The figures for Panama and the United States are for 1960; those for Sweden and Japan are for 1965. The figures for Korea are adjusted from 1963-71 mean values in table 6.3 to fit the age groups presented here. Source: John D. Durand, The Labor Force in Economic Development (Princeton, N.J.: Princeton University Press, 1975), pp. 194-207. growing proportion of the working-age population throughout the pro- jection period, because of the sustained net rural-to-urban migration, and that they will attain by 2000 an average LFPR thus far observed in only a handful of predominantly urban countries.26 This projection may be regarded as the maximum reasonable growth of the labor force. It is likely to be approached only if the labor demand for urban women is tight and sustained and if sociocultural pressure inhibiting married urban women from working for wages is relaxed. Unless drastic unfore- seen circumstances intervene, the outcome should fall between these two extremes. 26. Poland, a somewhat extreme example, reported an LFPR of 62 percent for women 15 years and older in 1970. ILo, Yearbook of Labour Statistics, 1974, table 1. LABOR AVAILABILITY AND EMPLOYMENT I 9I Under simulation A the aggregate labor force will grow at an average annual rate of 3.3 percent during 1975-80. Its growth will slow down to 3.1 percent a year during 1980-85 and then decline to 1.4 percent a year for the last five years of this century (table 6.4). The labor force is expected to grow at a much faster pace in urban areas than rural areas. By the early 1990s the rural labor force will begin to shrink, and during 1995-2000 it will decline at the rate of 1.3 percent a year. One apparent quirk in these projections is the acceleration in the annual rate of growth of the rural labor force from 1 percent a year during 1975-80 to 1.9 percent a year during 1980-85. This can be attributed to the aging of the rural baby-boom cohorts. In the 1975-80 period they are in the 14-19 age group for which LFPRS are low-51 percent for males and 37 percent for females-relative to the 20-24 age group they will move into in 1980-85-85 percent for males and 40 percent for females. Under simulation B the aggregate labor force will grow at an average annual rate of 3.8 percent during 1975-80. This rate remains virtually unchanged until the mid-1980s and then gradually slows down to 2 percent a year by century's end (table 6.5). The differences between these rates and those projected under simulation A are attributable to the alternative LFPR schedules assumed for urban females. Although the urban female labor force will grow at 4.4 percent a year during 1975-80 under simulation A, its corresponding rate under simulation B is 7.1 percent. It then gradually slows down to 4.8 percent a year in 2000. The results of simulation B provide a reasonable upper bound for rates of labor force growth during the next twenty-five years. If aggre- gate labor demand were geared to elicit employment growth rates higher than these at matching future time periods, the probable result would be a correspondingly strong wage-push on the economy. The effect of such a wage-push might nevertheless be mitigated by the government's ex- traordinary capacity to implement tough economic policies, such as limiting wage increases despite excess labor demand. One policy implication emanating from these simulation exercises is this: in formulating macroeconomic policies, rates of growth of aggre- gate employment should be generated within the bounds of simulations A and B for matching future periods. To the extent aggregate labor demand is associated with employment growth rates below the boundary traced by simulation A, it can be expected that the urban unemployment rate will tend to rise asymptotically toward, say, 20 percent and that real wages will stagnate or even deteriorate, particularly among lower paid workers. Conversely, to the extent aggregate labor demand is geared to employment growth rates above the boundary traced by simulation 192 PART TWO: HUMAN RESOURCES Table 6.7. Labor Force, by Sex and Place of Residence, Simulations A and B, Projections for 1975-2000 (millions of persons) Category/simulation 1975 1980 1985 1990 1995 2000 Total labor force Sirmnlation A 12.5 14.7 17.1 19.4 21.2 22.7 Simulation B 12.6 15.2 18.2 21.1 23.7 26.4 Rural males Simulations A and B 4.1 4.3 4.7 5.0 4.9 4.6 Urban males Simulations A and B 4.2 5.6 7.1 8.4 9.9 11.4 Rural females Simulations A and B 2.4 2.5 2.7 2.8 2.7 2.6 Urban females Simulation A 1.9 2.3 2.6 3.0 3.6 4.2 Simulation B 1.9 2.7 3.6 4.8 6.2 7.9 Note: These figures correspond to the labor force projections presented in tables 6.4 and 6.5 Source: World Bank projections. B, it can be expected that the urban unemployment rate will fall asymptotically toward 2 or 3 percent and that wage-push pressures will increase, particularly among higher paid workers. In addition, women might have fewer babies. To facilitate comparison of the two simula- tions, the changing size of the labor force by sex and residence is shown in table 6.7. Changing Composition of Employment Because of technological change and shifting patterns of output mix, the composition of employment in Korea has markedly changed (table 6.8). In 1963 about 63 percent of workers were attached to the agricul- tural sector; about 28 percent to the services sector; only 9 percent to the manufacturing sector. By 1974 the share of agriculture in employ- ment fell below 50 percent, while that of services grew to 33 percent LABOR AVAILABILITY AND EMPLOYMENT 193 and that of manufacturing doubled to 18 percent. Two features of this change in the sectoral composition of employment between 1963 and 1974 are noteworthy. The change was accompanied by substantial, though widely divergent, rates of productivity growth in all three major sectors, ranging from 2.9 percent a year in agriculture to an impressive 6.7 percent a year in manufacturing. The rates of employment growth were strongly correlated with the initial relative productivity of each sector and, what is more important and striking, with the rates of productivity growth among sectors. Korea's economic progress between 1963 and 1974 can be viewed as the result of the growth of employment and productivity in each major sector. In the manufacturing sector, employment grew at 10.3 percent a year and productivity at 6.7 percent a year. Corresponding annual increases in services were 5,5 percent and 4.5 percent; those in agricul- ture were 1.3 percent and 2.9 percent. The 9.8 percent annual rate of GDP growth was thus happily shared by employment growth of 3.8 percent a year and productivity increases of 6 percent a year. Because of the small proportion of manufacturing employment in the total for the base period, the employment growth directly attributable to man- ufacturing was not the highest of the three sectors. Manufacturing nevertheless played a crucial and dynamic role in stimulating the growth of productive employment in the services sector and, perhaps, in in- creasing the productivity of the agricultural work force. This last point is persuasively argued by Fei and Ranis who point out that the manufacturing sector generated the foreign exchange required to import a broad array of agricultural inputs not domestically produced during the early part of the period and later produced some of the inputs, such as chemical fertilizer and farm machinery.27 The view that the manufacturing sector contributes to employment growth in services is supported by two considerations: a priori expectations of a rich net- work of linkages between manufacturing activities and services; em- pirical studies that show a significant relation of growth in manufactur- ing output to growth in services employment.28 Before examining the 27. Fei and Ranis, "Model of Growth and Employment." 28. The theoretical and casually empirical literature on these relations goes back at least to Clark's work in 1957; the statistical analysis of cross-country com- parisons to Galenson's in 1963. Colin Clark, The Conditions of Economic Progress, 3d ed. rev. (London: Macmillan; New York: St. Martin's Press, 1957); Walter Galenson, "Economic Development and the Sectoral Expansion of Em- ployment," International Labour Review, vol. 87, no. 6 (1963). 194 PART TWO: HUMAN RESOURCES Table 6.8. Value Added, Employment, and Labor Productivity, by Sector, 1964-74 Item/sector 1963 1966 1970 1974 Value added (billions of won)' All sectors' 1,320 1,704 2,577 3,858 Agriculture' 532 668 725 848 Manufacturingd 187 275 591 1,212 Services' 601 761 1,262 1,799 Employment (thousands of persons)' All sectors 7,662 8,423 9,745 11,586 Agriculture 4,837 4,876 4,916 5,584 Manufacturing 667 913 1,395 2,062 Services 2,158 2,634 3,434 3,940 Value added per worker (thousands of won) All sectors 172 202 264 333 Agriculture 110 137 147 152 Manufacturing 280 301 423 588 Services 279 289 368 457 Note: Index numbers for value added:per worker are the product of value added per worker in each sector and that sector's share in total employment divided by value added per worker for all sectors. Consequently the sectoral index numbers do not add up to 100. Sources: EPB, Monthly Statistics of Korea, January 1976, p. 137; EPB, Economi- cally Active Population, p. 32. empirical aspects of this relation in the Korean context, some prospective inferences can be drawn from past and present patterns of the broad sectoral distribution of employment and output. Growth of agricultural employment The secular growth of the agricultural work force after 1963 calls for some explanation. This growth, however slow and irregular it may have been, is remarkable for several reasons. The rates of growth in output, employment, productivity, and real wages registered in the non- LABOR AVAILABILITY AND EMPLOYMENT 195 Annual rate Index number Ahsolute of change 1963 1974 change 1963-74 (all sectors 1963-74 (percent) = 100) Item/sector Value added (billions of won)' 2,538 9.8 100 100 All sectorsb 316 4.2 40 22 Agriculture' 1,025 17.0 14 31 Manufacturing' 1,198 10.0 46 47 Services' Employment (thousands of persons)' 3,924 3.8 100 100 All sectors 747 1.3 63 48 Agriculture 1,395 10.3 9 18 Manufacturing 1,782 5.5 28 34 Services Value added per worker (thousands of won) 161 6.0 100 100 All sectors 42 2.9 64 46 Agriculture 308 6.7 163 176 Manufacturing 178 4.5 162 137 Services a. In 1970 constant market prices. b. Equal to GNP minus the rest-of-the-world sector. c. Includes forestry and fishing. d. Includes mining. e. Includes social overhead capital and other services. f. Yearly average of quarterly observations. agricultural sectors during this period were high and sustained. Man- to-land ratios in agriculture, compared with other countries, were ex- tremely high at the beginning of the period and now stand among the highest in the world. The spread of mechanization in agriculture, particularly during the later part of the period, was substantial. In addition, much of this growth of agricultural employment occurred during 1970-75, when the growth of urban employment, particularly in services, was slowing down. This pattern suggests that agriculture is a residual employer, absorbing labor when urban demand is slack, but releasing labor when urban demand is brisk. Rates of open un- I96 PART TWO: HUMAN RESOURCES employment in urban areas fell from 16 percent in 1963 to 6.8 percent in 1974. With the upsurge in urban labor demand after 1975, it can be expected that the growth of agricultural employment subsequently slowed down and possibly turned negative. In addition to the possibility that the growth of agricultural employ- ment reported during this period was a statistical aberration, there are three possible, and complementary, explanations of this apparent para- dox. The first is that an array of policies was developed by govern- ment to discourage heavy migration out of rural areas. These were implemented through a combination of economic incentives and dis- incentives. For example, one important economic component of these policies was government's successful program after the late 1960s to tilt the rural-urban terms of trade in favor of agriculture by subsidizing the prices of major agricultural inputs and crops. A second possible explanation is that technological change in Korean agriculture during this period was not labor-saving. Some of the inputs, clearly recogniz- able as labor-using, have been fertilizer, higher-yielding seed varieties, and additional multiple-cropping made possible by growing seedlings in hothouses. Even the most ubiquitous form of mechanization in Korean agriculture-the power tiller-tums out to be fairly labor intensive. Great physical effort is required to operate these tillers, and the speed at which a field can be power-tilled is not significantly greater than that achieved with animal traction. A third explanation is that the proportion of males in the agricultural work force declined from 62 percent in 1963 to 57 percent in 1974. It therefore seems probable that the participation of females increased to meet seasonal peaks in the demand for agricultural labor.29 Of these three possible explanations for the growth of agricultural employment, only the second-the labor intensiveness of technological change-would tend to offset the effects on labor productivity of in- creasingly severe diminishing marginal returns resulting from the rising man-to-land ratio. It appears that Korean agriculture is at or near the forefront in applying this type of technology and that its dissemination is fairly well advanced. There is little room for further improvement and, barring unforeseen technological breakthroughs, additional progress 29. Another intuitively plausible explanation might be an increase in part-time farming or a decrease in the average hours worked per week in agriculture. But this hypothesis is contradicted by data that show an increase in the average work week in this sector from 42.9 hours in 1963 to 45.9 in 1973. EPB, Economically Active Population, pp. 50, 59. LABOR AVAILABILITY AND EMPLOYMENT I97 is likely to be much slower in the future.30 Indeed, the average rate of increase in output per worker slowed in constant market prices from 4.2 percent a year for 1963-68 to 2 percent a year for 1968-74.31 Strong economic forces consequently are likely to impede increasingly any further simultaneous growth of employment and productivity in Korean agriculture. If this sector must continue to absorb labor in the future, agricultural productivity will soon stagnate. But if market forces are allowed to influence future intersectoral deployment of labor, the evi- dence strongly suggests that the most critical factor governing this deployment during the next five to ten years will be the rate of growth of manufacturing output. In the longer run the changes in composi- tion of output and technology in the manufacturing sector are likely to play an increasing role. Effects of manufacturing output growth Using an input-output framework to capture indirect employment effects, two investigators estimated that manufacturing output ac- counted for 1,730,000 jobs in 1966 and 2,151,000 in 1975 .32 Of this net increase of 421,000 jobs, 122,000 were in manufacturing and the remaining 299,000 in all other sectors. Related estimates reported in that study suggest that it is unlikely that more than 15 percent of these additional nonmanufacturing jobs could have been generated in agri- culture. It therefore seems reasonable to assume that at least 254,000 of those jobs were in services. This figure is equal to 32 percent of em- ployment growth in services between 1966 and 1970 (table 6.8). This calculation underestimates the impact of manufacturing activity on employment growth in the services sector in two ways: the additional employment generated by manufacturing activity induces migration 30. To the extent that current crop composition can be shifted toward crops with higher value but similar requirements for labor intensity, the marginal revenue product of agricultural labor will tend to be increased. 31. Value-added figures are from EPB, Monthly Statistics of Korea, Januarv 1976, p. 136; employment figures are from aPB, Economnically Active Population, p. 32. 32. In relation to final demand 20 percent of these jobs were attributable to exports in 1966 and the remainder to domestic sales; the proportion attributable to exports rose to 24 percent in 1970. David C. Cole and Larry Westphal, "The Contribution of Exports to Employment in Korea," in Trade and Development in Korea, eds. Wontak Hong and Anne Krueger (Seoul: Korea Development Insti- tute, 1975), pp. 89-102. 198 PART TWO: HUMAN RESOURCES from rural areas to urban centers, where the greater demand for services induces a second-round expansion in services employment; the greater output in manufacturing causes real growth in domestic income and greater demand for services which typically are income-elastic. Because both processes are strongly operative in Korea, the underestimation of jobs directly and indirectly attributable to manufacturing output ex- pansion and those induced by second-round effects could be substantial. Projections of sectoral employment According to the macroeconomic model, sectoral employment during the 1976-90 period is projected to evolve in accord with the figures pre- sented in table 6.9. What are the implications of these simulations for possible labor shortages? Even under the high growth path, more than 30 percent of total employment would still be in agriculture in 1990. This proportion is commensurate with the corresponding figure for Japan in 1960. It thus appears unlikely that Korea will be confronted with a binding, overall labor shortage before the 1990s-that is, a shortage in the sense of there being competition for lower skilled labor within the manufacturing sector or between the manufacturing and modern services sectors-or with a condition that would sharply drive up the structure of real wages in those sectors. Labor productivity and wage differences As Korea shifts its dominant pattem of industrial growth from light manufacturing to heavy industries such as chemicals, machine tools, Table 6.9. Sectoral Employment, Projections for 1976-90 (millions of persons) High growth path Low growth path Sector 1976 1981 1990 1976 1981 1990 All sectors 12.2 14.3 18.9 12.2 14.2 18.2 Agriculture 5.5 5.6 5.8 5.5 5.6 5.8 Manufacturing 2.5 3.3 5.2 2.5 3.3 4.6 Services 4.2 5.4 7.9 4.2 5.3 7.8 Note: The high growth path assumes economic growth rates of 9.1 percent a year for 1976-81 and 9.5 percent a year for 1981-90; the low growth path 8.4 percent a year for 1976-81 and 7.4 percent a vear for 1981-90. Source: Projections of the macroeconomic model presented in table 3.4 and described in appendix A. LABOR AVAILABILITY AND EMPLOYMENT 199 and shipbuilding, a corresponding shift will occur in the composition of labor demand. Because the requirements for highly skilled labor will grow relative to the requirements for semiskilled workers who now con- stitute the bulk of manufacturing employment, there is some concern that the result of this process will be bottlenecks in the supply of certain types of skilled labor. These bottlenecks, it is feared, would hamper the growth of the emerging industries and generate wage-push inflation and gaping differences between the earnings of workers in short supply and the rest. In attempting to assess the likelihood of such shortages, it is useful to examine the past performance of Korea's labor markets in adapting to changing patterns of labor demand. According to official data, wage earnings accounted for 37 percent of national income in 1962. In 1974 their proportion of national income was again 37 percent (see table D.40 in appendix D). In the inter- vening years wage earnings fluctuated from a low of 28 percent of national income in 1964 to a high of 39 percent in 1970 and 1971. Although the real wage per worker grew by the same proportion as value added per worker during the entire 1962-74 period, there was much leapfrogging between these two variables. For example, between 1966 and 1970 the real average wage grew more than 75 percent faster than output per worker; between 1970 and 1974 productivity grew almost 50 percent faster than the real average wage (table 6.10). Aver- age real wages have apparently been growing in a typical lagged-response pattern to spurts in productivity through the period (figure 6.4). The patterns suggest that factor shares in output in Korea, as in several other countries, tend to be stable over the medium and long run despite the lack of any convincing reasons for this to be so and Table 6.10. Growth of Productivity and the Average Wage, 1963-74 (percent) Annual rate of change Item 1963-66 1966-70 1970-74 1963-74 Value added per worker' 5.4 6.7 5.8 6.0 Average wageb 4.3 11.8 3.9 6.9 Sources: Value added and compensation of employees from EPB, Monthly Sta- tistics of Korea, January 1976, p. 139; total employment from EPB, Economically Active Population, p. 139. a. Equal to GDP in constant 1970 prices divided by total employment. b. Equal to total compensation of employees in constant 1970 prices divided by total employment. 200 PART TWO: HUMAN RESOURCES Figure 6.4. Real Value Added per Worker and the Real Average Wage, 1963-74 Index (1963 = 100) 200~~~~~~~~~~~~~~~.- 150 100 1963 1965 1967 1969 1971 1973 1964 1966 1968 1970 1972 1974 Real value added per worker - ~ - Real average wage Source: Table D.38 in appendix D. despite the dramatic structural changes in the Korean economy after 1960.33 Because there is no compelling argument for Korea to depart from its historical pattem of secularly stable factor shares, it is plausible that this pattern will be preserved in the future. Consequently the best working assumption appears to be that average real wages will continue to rise at approximately the same secular rate as output per worker. Empirical evidence indicates that wage differences generally have not widened on a trend basis and that ever-widening wage differences 33. There are fairly compelling arguments, such as those of Kuznets, for labor's share to have shrunk during this period, but it did not. Simon Kuznets, "Economic Growth and Income Inequality," American Economic Review, vol. 45, no. 1 (March 1955). LABOR AVAILABILITY AND EMPLOYMENT 201 Table 6.11. Earnings Differences, by Occupational Category, 1971-74 1971 1974 1971-74 Ratio (1) (2) (2)/(1) Chemical engineers to: Machine tool setters 3.43 3.89 1.14 Phone and telegraph installers 1.79 3.05 1.70 Plumbers and pipefitters 2.22 2.03 0.91 Transport conductors 6.77 5.83 0.86 General farmworkers 9.35 2.90 0.31 Machine tool setters to: Transport conductors 1.98 1.50 0.76 General farmworkers 2.74 0.75 0.27 Note: The basis for comparison is total monthly wages per hour; wages include annual bonuses. Source: Office of Labor Affairs, Yearbook of Labor Statistics, 1972, pp. 101-13, and 1975, pp. 136-55. are unlikely in the future. This is true despite the immense differences in productivity and technological modernity within and across indus- tries in the early 1960s, the extremely rapid and diverse patterns of modernization since that time, and the possible widening that may have resulted from the upsurge in economic growth in 1976-77. The only apparent exception to this generalization is that the wage difference between males and females seems to have widened substantially in recent years.34 Periods of brisk growth in labor demand, when the composition of the skills being sought by employers is rapidly changing, provide an ideal time frame in which to assess the allocational efficiency of labor markets and the response elasticity of skill formation. A simple, but powerful, indicator for such an assessment is the evolution of wage differences between occupations characterized by skills subject to widely diverging labor market conditions. Extant data allows this comparison only be- tween 1971 and 1974 (table 6.11). Consequently the inferences 34. For the empirical evidence see Bertrand Renaud, "Economic Growth and Income Inequality in Korea," World Bank Staff Working Paper, no. 240 (Wash- ington, D.C.: World Bank, February 1976; processed), p. 160. 202 PART TWO: tHUMAN RESOURCES drawn from data for such a short period cannot be considered as well- established trends. This comparison of earnings differences by occupational category in- dicates two interesting results. First, relative to the wage rates for chem- ical engineers, who are taken to represent university-trained professionals in technical fields, the hourly earnings of highly skilled technicians, such as machine tool setters and telephone and telegraph installers, have declined. This finding does not support the view that labor shortages were developing among the second occupational group relative to the first. Second, wage differences between the highest and lowest paid occupations presented in these comparisons (chemical engineers and lower skilled occupations) narrowed appreciably between 1971 and 1974. The larger the difference in 1971, the more the narrowing by 1974. This finding suggests that labor markets and skill-formation mechanisms performed efficiently during a period characterized by rapid labor demand and structural changes in the skills required. Combined, these findings indicate that labor markets in Korea have been powerful forces toward a more egalitarian distribution of income. This evidence justifies considerable optimism about the ability of labor market mechanisms to continue to allocate and generate skills in response to the demands of a growing and changing economy. PART THREE Major Sectors Chapter 7 Randolph L. P. Harris Agriculture THE LAND REFORM ACTS of 1947 and 1948 improved the distribution of land and contributed to the growth of farm income. The reforms virtuallv eliminated tenancy, put a ceiling of three hectares on paddy holdings, and established a structure of small, owner-operated farms (table 7.1). Tenancy arrangements have since returned on a small scale, but the broad characteristics of land distribution remain un- changed. Despite the migration of agricultural workers to urban areas, labor productivity in agriculture is still low and is rising more slowly in relation to other sectors. Government has used several policy instru- ments to offset this difference, including increased investment in agri- culture, price support for grains, subsidies for agricultural inputs, and the New Community movement which improves rural living standards through a program of self-help. The distribution of income in Korea today is considered to be among the best in the developing world, and there is little doubt that rural living standards have significantly improved. Even so, average rural incomes are considerably below average urban incomes. The challenge for the future will be to prevent widening disparities in the face of rapid productivity growth in manufacturing. Several conditions militate against this. The extent to which government can afford to shift the terms of trade in favor of agriculture is limited. Rural income from nonagricultural sources has not grown as expected, remaining instead at about 20 percent of farm household income during the first half of the 1970s. Programs to locate industrial estates outside existing industrial areas, along with the factory program of the New Community move- ment, have met only with limited success. It is clear that fresh measures will have to augment conventional programs if rural incomes are to keep pace with those in urban areas. 205 zo6 PART THREE: MAJOR SECTORS Table 7.1. Distribution of Farms, by Size, 1974 Percentage of Percentage of farm cultivated Farm size households land area Landless 4.5 Less than 0.5 hectare 29.4 10.6 0.5 to I hectare 34.9 29.7 I to 3 hectares 29.8 52.9 More than 3 hectares 1.4 6.8 ... Zero. Source: Ministry of Agriculture and Fisheries (MAF), Year- book of Agriculture and Forestry Statistics, 1977, pp. 26-27. Historical Perspective Only 23 percent of Korea's 99,000 square kilometers of relatively infertile and ruggedly mountainous land is cultivable; 67 percent is forest; 10 percent satisfies requirements for urban areas, industrial es- tates, and roads. Paddy fields, which produce rice in the summer and barley where climatic conditions permit in the winter, cover 58 percent of cultivable area. Unirrigated upland fields, which produce barley, wheat, rye, millet, peppers, tobacco, and numerous other summer and winter cash crops, cover 42 percent. Korea's population density in 1976 was one of the world's highest: 363 persons per square kilometer and 16 persons per hectare of farm- land. Population pressure has led to intensive land development, in- cluding irrigation, drainage, and consolidation.' The cropping intensity of cultivable land is almost 140 percent; 84 percent of paddy land is irrigated. Population pressure has also led to the annual loss of about l1,000 hectares (0.5 percent) of cultivable land to urban growth, fac- tories, and roads. But the conversion of upland to paddy land and the reclamation of agricultural land from forests and tidal flats have more than offset these losses. The total area cultivated, albeit of declining quality, increased from 2 million hectares in 1960 to 2.24 million hec- tares in 1975. Taking double-cropping into account, the total area under crops in 1976 was 3.1 million hectares. The relative importance of agriculture, which includes forestry and 1. Land consolidation consists of replacing the uneven pattern of small, irregular plots with larger, level plots in a rectangular grid of ditches, farm roads, and drains. It facilitates mechanization, improves water control and drainage, and enables easier access and better management. AGRICULTURfE 207 fishing, has been declining steadily. Its contribution to GNP fell from 44 percent in 1961 to 20 percent in 1976. Its contribution to export eamings also declined-from 30 percent in 1964 to about 9 percent in 1976. Nevertheless the growth of value added in the entire sector aver- aged a respectable 4 percent a year for the 1966-76 period. The expan- sion of deep-sea fishing accounted for a good part of this growth (table 7.2). Excluding fishing, value added in agriculture grew at about 2.5 percent a year over the same period. Much of this growth was the result of two changes: a shift in sectoral composition from barley, wheat, and potatoes to higher valued rice; and the increased share of high-value, labor-intensive fruits and vegetables. Livestock production accounts for about 6 percent of value added in the agricultural sector. It declined substantially in 1975 because of the rapid escalation in the cost of imported feedstuffs, but recovered somewhat in 1976. Although the pork industry depends on imported feed, export trade in pork has been de- veloping with Japan. Despite agriculture's declining significance in the economy, it con- tinues to play a crucial role in providing the main source of income for the rural population and in providing food for the growing and increas- ingly affluent population. The proportion of Koreans living in villages fell steadily from 55 percent in 1965 to 36 percent in 1976; the rural population peaked at 16.7 million in 1964 and declined to 12.8 million Table 7.2. Value Added in Agriculture, Forestry, and Fishing, 1966-76 (millions of won in constant 1970 prices) Sector 1966 1969 1972 1974 1976 Agriculture 600,589 646,381 644,887 673,042 779,037 Agricultural crops 562,729 588,852 571,753 599,234 707,362 Livestock 24,972 42,542 51,326 52,845 49,322 Side business 232 350 1,375 964 1,074 Attached services 12,656 14,637 20,433 19,999 21,279 Forestry 36,449 43,379 43,129 45,346 44,381 Fishing 30,867 41,717 72,899 115,521 149,626 Inshore 25,556 29,932 42,918 59,292 56,604 Aquiculture 2,687 3,694 8,075 19,324 18,002 Deep-sea 2,624 8,091 21,906 36,905 75,030 Total 667,905 731,477 760,925 833,909 973,054 Source: MAF, Yearbook of Statistics, 1977, pp. 398-99. 208 PART THREE: MAJOR SECTORS in 1976. Forty-three percent of the country's labor force of 13.1 million was employed in agriculture in 1976, a considerable decline from 58 percent in 1963. The principal grain in the Korean diet is rice, which accounted for 43 percent of total food consumption by weight in 1975. Barley accounted for 21 percent, and wheat 19 percent. Foodgrains thus continue to constitute more than 80 percent of the Korean diet. According to the official definition, foodgrains comprise the staple crops of rice, barley, wheat, and potatoes and the secondary crops of rye, millet, sorghum, corn, and pulses. Using this definition the average foodgrain production for 1965-66 was about 6.8 million tons; that for 1975-76 about 7.9 million tons (table 7.3). The loss of output resulting from the reduction of cropped area was more than offset by higher yields, the effects of research and extension efforts, and the increased area of irrigated and consolidated land. The area sown to winter grains declined after rice prices began to increase in 1968, but recovered slightly during 1975 and 1976 because of government policies to support grain prices and increase barley production. Barley yields and production have risen; wheat production has fallen despite higher yields. Available wheat varieties are not well suited to crop rotation because they take too long to mature. If wheat is allowed to come to full maturity, the yield of the succeeding rice crop is reduced because of late sowing. As with wheat and barley, considerations of timing are important deter- minants of the decline in potato production. Table 7.3. Average Area, Yield, and Production of Foodgrains, 1965-76 Area Yield Production 1965-66 1975-76 1965-66 1975-76 (thousands of (kilograms per 1965-66 1975-76 Foodgrain hectares) hectare) (thousands of tons) Rice 1,230 1,216 3,016 4,064 3,710 4,942 Barley' 802 711 1,928 2,433 1,546 1,730 Wheat 94 40 2,000 2,250 188 90 Potatoesb 210 142 4,717 4,968 990 706 Other 557 407 - - 348 471 Total 2,893 2,516 - - 6,782 7,939 - Not applicable. Source: MAF, Yearbook of Statistics, 1977, pp. 74-85. a. Includes naked barley. b. White and sweet potatoes. AGRICULTURE 209 Rice, the dominant crop, accounted for 64 percent of foodgrain pro- duction in 1976. Its output, aided to some extent by favorable weather conditions, rose from 4.4 million tons in 1974 to 5.2 million tons in 1976. The output of barley rose from 1.4 million tons in 1974 to 1.8 million tons in 1976. Harvests of rice and barley were sufficient to meet domestic requirements for these grains in 1975 and 1976. This self- sufficiency was attained with the help of regulations requiring res- taurants not to serve rice two days a week; requiring institutions to use a 70:30 rice-barley mix; requiring that rice be coarse-milled to 7 degrees, not fine-milled to the preferred 9 degrees; controlling the relative prices of rice and barley. Government is now relaxing its restrictions on rice consumption so that rice can supplant imported wheat. The steady upward trend in rice production is a function of three basic changes: the rapid acceptance by the literate rural population of new technology, such as fertilizer, weedicide, and pesticide; the extension of irrigation and drainage; the introduction of locally adapted, higher yielding varie- ties which by 1976 were sowvn on 43 percent of the sown area. These factors combined to raise average rice yields despite the loss of prime riceland to industry and its partial replacement by less fertile converted forest and upland. Agriculture under the Fourth Plan Two major goals of government policy expressed in the fourth plan are to increase economic self-sufficiency and to improve the welfare of all sections of the community, especially rural dwellers. The first implies additional investment in agricultural production. The second implies improvements in education, rural electrification, roads, telecommunica- tions, housing, water supply, public health, and social security. Specific goals of the plan for agriculture are: v Sustained growth of agriculture and fishing to ensure a growing domestic supply of food. These objectives are to be attained by in- creased irrigation, drainage, and land consolidation and reclama- tion; the use of high-yielding varieties such as Tongil, Yushin, and Milyeng; and an increase in livestock production. Self-sufficiency in rice and barley is to be maintained. * Increased farm and fishing household income. This is to be at- tained as a result of agriculture's sustained growth, the develop- ment of rural industry, and improvement of the agricultural mar- keting system. 210 PART THREE: MAJOR SECTORS * Enhancement of the rural living environment. This is to be attained through rural infrastructure activities ranging from small-scale im- provements to village drains to fairly sophisticated large-scale utility or irrigation projects. Since 1971 Korea has undertaken a massive program of rural infrastructure under the New Community movement. Although the movement is nationwide, it concentrates its activities on the less-developed villages which tend to be in the more remote parts of each region where poor transport and lack of services hamper development. The movement has two essential features: decentralized decisionmaking and self-help, either through volunteer labor or financial contributions. The movement places heavy emphasis on the development of self-reliance. In the fourth plan the proposed investment in agriculture is W1,979 billion, or 10.4 percent of total investment (table 7.4). This investment Table 7.4. Investment Program for Agriculture, 1972-81 Third plan 1972-76 Fourth plan 1977-81 Billions Percentage Billions Percentage Sector/itenm of won of total of won of total All agricultural sectors' 1,341 100.0 1,979 100.0 Agriculture 1,027 76.6 1,500 75.8 Increased grain production 52 3.9 80 4.0 Cash crops 75 5.6 36 1.8 Livestock and sericulture 119 8.9 133 6.7 Land development 359 26.8 600 30.3 Farm mechanization 85 6.3 279 14.1 Marketing and storage 42 3.1 166 8.4 Research and extension 38 2.8 63 3.2 Environmental inmprovemene 190 14.2 104 5.3 Other 67 5.0 38 1.9 Forestry 91 6.8 194 9.8 Fishing 223 16.6 285 14.4 Note: All figures are in constant 1975 prices. Source: Economic Planning Board (EPB), The Fourth Five-Year Economic De- velopment Plan, 1977-81 (Seoul, 1976). a. The proportion of this investment in total investment was 11.8 percent in 1972-76; it is projected to be 10.4 percent in 1977-81. b. By the New Community movement. AGRICULTURE 211 will maintain the momentum generated in the third plan period, when investment in agriculture was W1,341 billion, or 11.8 percent of total investment and three times the expenditure under the second plan. The basic change between the third and fourth plans is the substantial in- crease in mechanization. The expenditure associated with the New Community movement is for community centers, common workshops, warehouses, village roads, and bridges. The heavy government expendi- ture on such projects in recent years largely satisfied the most pressing needs, and future activities will increasingly rely on self-help. In addi- tion to these activities, ministries other than the Ministry of Agriculture and Fisheries (MAF) have substantially increased their planned expendi- ture in rural areas. The fourth plan's targets for a number of agricultural indicators are shown in table 7.5. Table 7.5. Targets of the Fourth Plan for Agriculture, 1975-1 Item 1975 1981 Cultivated land area (thousands of hectares) 2,240 2,335 Paddy fields 1,277 1,333 Irrigated paddy fields 1,072 1,248 Consolidated paddy fields 251 419 Upland 963 1,002 Production (thousands of tons) Rice 4,669 5,472 Barley 1,700 1,875 Wheat 97 283 Meat 235 368 Milk 160 499 Fruit 644 1,150 Cocoons 36 49 Eggs (millions) 2,896 4,377 Fertilizer consumption (thousands of nutrient tons) 886 1,083 Pesticide consumption (thousands of tons in element) 21 30 Average farm household income (thousands of won in constant 1975 prices) 873 1,400 Rural electrification (thousands of households) 1,643 2,532- Reforested land (thousands of hectares) 4,434 5,530 Production of marine products (thousands of tons) 2,135 3,562 Export of marine products (millions of dollars) 429 824 Source: Data supplied by EPB. a. To be completed by 1978. 2I2 PART THREE: MAJOR SECTORS Table 7.6. Food Consumption, Korea and Japan, Selected Years (kilograms per year per capita) Japan Korea Item 1954-56 1972 1975 2000' Rice 104 92 113 102 Barley 17 1 53 34 Wheat 25 31 48 53 Potatoes and starch 48 25 49 26 Fruit 16 44 14 38 Vegetables 75 118 76 108 Beef, pork, and chicken 3 15 9 18 Milk 12 52 4 30 Note: All figures have been rounded to the nearest kilogram. Sources: Thodev, "Food and Nutrition," table 5.3; World Bank cal- culations. a. Targets. Major Issues of Agricultural Policy The major agricultural issues confronting the Korean government are maintainingy the balance between food demand and production, sus- taining rural employment while reducing seasonal peaks in the demand for farm labor, and preventing the gap between rural and urban incomes and living standards from widening. Directly related to these basic issues are side issues of farm mechanization, prices, and subsidies. Matching food production with demand The estimates of food demand in Korea presented here are based on a detailed study of 1965-74 farm and nonfarm food consumption pat- terns, nutritional requirements, projected changes in population size and structure, and a comparison with Japanese diets (table 7.6).2 The target diet for Korea in the vear 2000 assumes a rise in total food consump- tion for farm and nonfarm consumers; an increase in fruit, vegetable, animal, and marine products as a proportion of all foods; and a decline in 2. Alan R. Thodey, "Food and Nutrition in Korea, 1965-74," Korean Agri- cultural Sector Study Special Report 11 (Seoul: National Agricultural Economics Research Institute; East Lansing: Michigan State University, n.d.). AGRICULTURE 213 the relative weight of grain. Korea's per capita rice consumption is pro- jected to rise until about 1978-80 and then decline. Food consumption patterns for Korea in 1990 were calculated by interpolating the figures for 1975 and 2000 and making adjustments for changes in the popula- tion (table 7.7). One significant change is the projected rise of the working-age population from 50 percent to 60 percent of the total popu- lation. Changes in population and consumption will affect each com- modity differently. Rice demand by 1990 is projected to be 36 percent greater than in 1975; barley demand 10 percent greater; wheat demand 48 percent greater; milk demand 725 percent greater. Korea's foodgrain output can be increased by two basic means: in- Table 7.7. Food Consumption, Projections for 1975-90 (thousands of tons) Total demand Item 1975 1981 1990 Grains 10,797 13,214 14,289 Rice 4,304 5,416 5,857 Barley 2,117 2,555 2,324 Wheat 1,693 2,001 2,510 Other cereals 106 106 102 Potatoes and starch 2,222 2,904 2,938 Pulses 355 432 558 Fruit 600 785 1,395 Vegetables 3,140 3,885 5,252 Livestock products 1,693 2,534 4,086 Beef 71 98 163 Milk 141 510 1,023 Pork 106 137 214 Chicken 71 78 116 Eggs 176 259 432 Fish 1,129 1,452 2,138 Note: Gross consumption, which includes provisions for seed requirements and storage losses, is the basis of total demand. Demand is based on a 1975 population of 35.3 million and population growth as described in chapter 4. Total population is projected to be 39 million in 1981 and 46.5 million in 1990. Source: World Bank calculations. 214 PART THREE: MAJOR SECTORS creasing the area of cropped land by converting forest to upland or paddy, reclaiming tidal flats, and intensifying cropping on land already cultivated; increasing the output from existing cropland by improving land with irrigation and consolidation and by improving yields with better varieties, greater inputs, sound practices, and selective mechani- zation. The purpose of the discussion that follows is to determine the appropriate level of land and water development, which government directly controls.3 This development is subject, of course, to variations in cropping intensity and yields, which government can influence only indirectly through various policies for research, extension, technology, and incentives, as well as through land and water development. Al- though investment, cropping intensity, and yields are highly interrelated, an attempt was made to isolate the effects of each variable on production in order to come up with a recommended level of investment. The conclusions should be viewed as preliminary, subject to confirmation as more experience is gained and analytical tools are sharpened. They nevertheless indicate the level of investment that appears justified on the basis of existing knowledge and information. ALTERNATIVE INVESTMENT IN LAND AND WATER RESOURCE DEVELOP- MENT. Estimates of the area suitable for upland reclamation from the 6.6 million hectares of forest vary from 135,000 to 220,000 hectares (table 7.8). Part of this area is so steep that it is suitable only for orchard development, but about 110,000 hectares have slopes of less than 19 percent and most undeveloped areas have slopes of about 15 percent. If precautions were taken to avoid soil erosion, these areas could be developed for upland crop production. The chief constraints on upland development are the characteristics of the soil. Its poor fertility and high acidity need to be rectified with heavy dressings of fertilizer, compost, and lime. Its shallowness, coarse texture, and low content of organic matter and clay reduce its capacity to hold water. These factors, com- bined with low and irregular rainfall from mid-April to late June, make crop yields variable and uncertain. Nevertheless economic analysis of typical upland reclamation investments indicates that such investments can generate an economic rate of return of about 20 percent and be financially attractive to small-scale farmers, particularly if credit is avail- 3. This discussion relies heavily on recent experience with land and water development projects financed by the World Bank, as well as on work in Korea of the Agricultural Development Corporation and the Korea Agricultural Plan- ning Project. AGRICULTURE 215 Table 7.8. Parameters of Additional Land Development, 1977 Cost, Area (dollars Type of development (hectares) per hectare) Creating additional cropland Tidal land reclamation 410,000b 10,000' Forest land conversion 135,000-220,000 2,000 Improving land base Irrigation 240,000 6,000 Consolidation 314,000 1,200 Source: World Bank calculations. a. In 1977 prices; includes physical contingencies. b. To produce 410,000 hectares of cropland requires reclaiming 600,000 bectares of tidal land. c. Based on an appraisal of the Yong San Gang Stage II irrigation project, it is probable that costs would range from $6,000 to $14,000 per hectare, depending on the site. able to establish apple orchards.4 Although determining the amount of upland that could be developed must await more detailed investigation, enough land of high potential is available to justify a program of at least 50,000 hectares over the 1977-86 period. If it turns out that only the lower estimate of 135,000 hectares should ultimately be developed, the pressure to expand the land base by re- claiming tidal flats might increase. Because tidal reclamation for agri- cultural use is expensive and has a long gestation period, government should consider the alternative of using reclaimed tidal land for indus- trial purposes. Development costs would be lower, and the additional land would relieve the pressure to convert prime agricultural land for industrial use. Nevertheless experience with tidal land reclamation in recently completed irrigation projects indicates that the cost per incre- mental ton of rice produced compares favorably with that of other modes of land and water development because incremental production and cost per hectare are both high. The estimated economic rate of return for such development is about 13 percent.5 Although less attractive than 4. This analysis was conducted during appraisal of the Rural Infrastructure Project financed in part by the World Bank. 5. This tidal reclamation was a component of the Pyongtaek-Kumgang and Yong San Gang Stage II projects financed in part by the World Bank. 216 PART THREE: MAJOR SECTORS upland reclamation, tidal land reclamation appears to be technically feasible and economically viable. At full development the incremental financial return to the farmer is expected to be almost 50 percent greater per hectare than that from upland development, but the time lag is greater. Having determined that different types of land and water develop- ment projects appear to be individually justified, the next step is to estimate the area of each type of improvement that could be developed for three alternative levels of investment: * no further investment beyond completion of ongoing projects6; * an annual investment of $60 million in addition to completing on- going projects; or * an annual investment of $120 million in addition to completing ongoing projects. The pattern of land development corresponding to these investment levels was the result of two judgments. First, estimates of the area that could be developed each year were based on the skilled manpower available, the managerial capacity of the responsible institutions, and the current state of knowledge about converting steep forest land and reclaiming tidal land. Second, the results of a study to determine the optimum pattern of land development through 1990, with different levels of investment and the object of maximizing grain production in 2000, were modified to accommodate the first judgment.7 The resulting program shows the area of land that would be improved, starting from a 1975 base, under the three investment strategies for land and water development (table 7.9). CROPPING INTENSITY. Double-cropping creates peaks in labor demand, especially for areas sown to rice and barley. Technological improve- ments, particularly the development of earlier maturing rice varieties and the rearrangement of paddy fields to permit better access and water control, have substantially increased cropping on paddy land during the winter. The cropping intensity on paddy land rose from 146 percent in 1966 to 162 percent in 1976. Nevertheless cropping intensity on all 6. It is estimated that completion of ongoing projects for land and water development during the fourth plan period will cost $600 million, or $120 million a year on average. 7. Richard D. Duvick, Jeong Bae Kim, and Jeong Boo Kim, "Analysis of Long-Term Impacts of Land and Water Development on Self-Sufficiency of Food Grains, Korea, 1976-2000" (Seoul: n.p., 1976; processed). AGRICULTURE ZI7 Table 7.9. Cumulative Area of Improved Land to be Available for Cropping under Three Levels of Investment, 1977-86 (thousands of hectares) 1981 1986a Additional investment Additional investment $60 $120 $60 $120 million million million million Type of development None a year a year None a year a year Creating additional cropland 17 38 46 17 63 92 Tidal land reclamation 5 10 10 5 15 32 Forest land conversion 12 28 36 12 48 60 Improving land base 78 110 148 78 155 240 Irrigationb 38 50 64 38 80 120 Consolidation 40 60 84 40 75 120 Total 95 148 194 95 218 332 Note: The investment levels are in addition to investment committed to projects under way in 1976. Source: World Bank calculations. a. All improved land would be under the first crop by 1986. The full yield potential would be attained by 1990. b. It is assumed that all of the irrigated area developed from 1976 onward will be consolidated. cultivated land, including upland, fell slightly during the same period, from 141 to 138 percent. This implies a decline in upland cropping intensity from 135 percent to 105 percent. Much more needs to be known about the reasons for changes in cropping intensity, particularly in upland areas. The importance of cropping intensity is clear. A rise in overall in- tensity from 138 percent to 160 percent, associated with an increase of 500,000 hectares in double-cropped upland area, would add more than the entire potential cropland that could be reclaimed from tidal flats and two to three times the potential area that could be converted from forest. Several ways to increase cropping intensity further appear pos- sible: increasing the domestic price of winter crops relative to rice; im- proving the reliability of irrigation water in paddy areas, thereby releas- ing water now saved for the rice crop; raising the yields of winter crops and permitting earlier planting of barley; developing wheat varieties that mature faster; mechanizing farm operations selectively in the peak seasons. 218 PART THREE: MAJOR SECTORS YIELDS. After Japan, Korea has the highest average rice yield in Asia. The average yield of polished rice was 3.9 tons a hectare in 1975 and 4.3 tons a hectare in 1976. These may have been abnormally good years. To establish the range of rice production likely to result from the three investment alternatives, low and high sets of yield assumptions were made. The low base yield was taken as 3.7 tons a hectare, equal to the average yield for 1972-75; the high base yield as 4 tons a hectare. For the period 1976-90 it is expected that the new Yushin and Milyeng varieties will replace the lower yielding Tongil variety and that the area under improved varieties will continue to expand beyond the present area planted with Tongil and Yushin. Average rice yields are assumed to grow at 1.5 percent a year and in 1990 reach 4.5 tons a hectare from the low base and 5 tons a hectare from the high base. Because of the effect of land improvement, average yields would tend to be higher with higher levels of investment. For barley and wheat, base yields of 2.4 and 2.2 tons a hectare were respectively assumed. These yields are equal to actual 1976 yields. Barley yields were assumed to increase at 0.9 percent a year and wheat yields at 1 percent a vear. These assumptions must be placed in perspective. One of every three Korean farmers obtained a yield of more than 5 tons of polished rice a hectare in 1976. Average Korean yields are still below those attained by Japanese farmers and 70 percent of experimental farm yields. The effects on rice and barley production of attaining average Japanese yields or 90 percent of experimental farm yields are substantial (table 7.10). Table 7.10. Increase in Annual Grain Production under Alternative Yields (thousands of tons) Average Ninety percent Net Japanese of experimental imports Grain yields yield?s in 1974 Polished rice 290 880 206 Polished barley 130 780 299 Note: The increases are based on comparisons with average Korean yields in 1975 of 3.9 tons a hectare for rice and 2.4 tons a hectare for barley. Source: Sung Hoon Kim and Dong Min Kim, "Population and Food in Korea," (paper prepared for PAO/IAEE/UNEPA cosponsored seminar on Population, Food, and Agricultural Development, Rome, December 1-5, 1975; processed), p. 48. a. Experimental yields are those obtained in Korean experi- mental stations. AGRICULTURE 219 With average Japanese yields Korea would have been self-sufficient in rice in 1974 and would have reduced barley imports by more than one- third. With 90 percent of experimental yields Korea would have pro- duced a surplus of both crops. Because of greater plot size, generally lower technological and managerial skills, and the untimeliness of such operations as planting and spraying, it is not expected that farm yields will reach experimental yields. Because of the intensive farming condi- tions in Korea, yields are nevertheless expected to reach 90 percent of experimental yields. Reinforcing this expectation is the fact that Korean farmers, despite their high levels of achievement, still obtain yields that are well below potential. One possible reason for this pattern is that large areas of land do not have assured supplies of irrigation water or adequate water control, both of which should improve. Rice yields in 1990 should therefore be closer to the high estimate than to the low estimate. RECOMMENDED INVESTMENT. By combining the foregoing assump- tions of investment levels, cropping intensities, and yields, it was possible to compute the corresponding production levels for rice, wheat, and barley. Expected rice production for 1981 and 1990 under three levels of investment and two yield assumptions is presented in table 7.11. The estimated production of the three grains, based on additional assump- tions about cropping intensity, is compared with demand for 1990 in table 7.12. If government makes no further investment beyond the completion of existing projects, and if actual yield experience falls between the high and low assumptions, an average annual import requirement of about 300,000 tons of rice can be expected. If weather is poor in any given year, imports of rice might be as much as 800,000 tons. Even if prices were to rise from $400 to $800 a ton in 1977 dollars and the foreign exchange requirement from $120 million to $640 million, the higher figure would still be less than 2 percent of total imports projected for 1990. The additional investment of $120 million a year would significantly improve Korea's prospects for self-sufficiency in rice. It could even gen- erate a small surplus by 1990, which could be exported or used to augment stocks. Beyond this level of investment, the returns are likely to diminish. Incremental physical production would probably decline and become less important if Korea were to become a net exporter of rice.8 It therefore is reasonable to recommend that the Korean govern- 8. Duvick, Kim, and Kim, "Land and Water Development." 220 PART THREE: MAJOR SECTORS Table 7.11. Rice Production under Three Levels of Investment, Projections for 1981 and 1990 (thousands of tons) 1981 1990 Additional investment Additional investment $60 $120 $60 $120 mil- mil- mil- mil- lion lion lion lion Item None a year a year None a year a year Base production Low yield assumption 4,876 4,876 4,876 5,486 5,486 5,486 High yield assumption 5,364 5,364 5,364 6,095 6,095 6,095 Minus production from annual land loss of 5,000 hectares Low yield assumption 100 100 100 315 315 315 High vield assumption 110 110 l 11 350 350 350 Plus additional produc- tion over 1975 Low yield assumption 101 141 161 116 303 487 High yield assumption 109 149 174 126 328 523 Projected rice production Low yield assumption 4,877 4,917 4,937 5,286 5,474 5,658 High yield assumption 5,363 5,403 5,428 5,871 6,073 6,268 Note: The low base yield was assumed to be 3.7 tons a hectare; the high base yield 4 tons a bectare. Average rice vields are assumed to increase by 1.5 percent a year and reach 4.5 tons a hectare from the low base in 1990 and 5 tons from the high base. Source: World Bank calculations. ment invest about $120 million annually in land and water develop- ment, in addition to completing ongoing projects, during the fourth plan period. By the early 1980s experience will have flushed out the problems of forest conversion and tidal land reclamation. Evaluation of the costs and returns associated with reclaiming the remaining in- accessible areas should then be undertaken. The most striking feature of table 7.12 is the impact of increased cropping intensity on wheat and barley production. If overall cropping intensity stays permanently at an average level of 140 percent, even the high level of investment will generate only about 65 percent of the AGRICULTURE 221 Table 7.12. Foodgrain Production, Demand, and Irport Requirements, Projections for 1990 (thousands of tons) Additional investment $60 $120 million million Item None a year a year Rice production Low yield assumption 5,286 5,474 5,658 High yield assumption 5,871 6,073 6,268 Wheat and barley production Forty percent double-cropping 1,568 1,646 1,714 Sixty percent double-cropping 2,030 2,780 2,862 Rice demand 5,857 5,857 5,857 Wheat and barley demand 4,834 4,834 4,834 Rice surplus or deficit Low yield assumption -571 -383 -199 High yield assumption 14 216 411 Wheat and barley surplus or deficit Forty percent double-cropping -3,266 -3,188 -3,120 Sixty percent double-cropping -2,804 -2,054 -1,972 Note: Import requirements are equal to deficits. As described in the text and in notes to tables 7.7, 7.9, and 7.11, alternative investment levels are those in addition to investment already committed to cropland development in 1976; demand is based on a 1990 population of 46.5 million persons; and yield assumptions for 1990 are 4.5 tons a hectare from the low 1975 base and 5 tons a hectare from the bigh 1975 base. Source: World Bank calculations. requirement for wheat and barley. If self-sufficiency in foodgrains is the objective, upland production, particularly double-cropping cereals, de- serves more attention. A 20 percentage point increase in double-cropping would add I million tons of grain. In sum, self-sufficiency in rice should be relatively easy to attain by 1990. Attaining self-sufficiency in barley, and producing enough wheat to reduce dependence on imports significantly, will be much more difficult. LIVESTOCK. The fourth plan proposes a 20 percent increase in the number of beef and draft cattle from the 1976 level of about 1.8 million 222 PART THREE: MAJOR SECTORS head. This increase is to be achieved in small units of two to five animals per household. Domestic waste, natural pastures, and rice straw, as opposed to imported feedgrains, will be used for feed. Purchases of animals are to be financed by special credit funds. Biological constraints will not hinder achievement of the target, but demand for animals will depend on the interest of farmers in caring for the animals, their desire for cattle as an asset, and the value of cattle as draft animals in the face of the planned spread of power tillers. The fourth plan also proposes a threefold increase in the number of milk cows and the volume of milk production. Demand in 1990 is projected to be seven times current consumption (table 7.7). The plan's assumptions for production and herd growth, given the herd charac- teristics of countries from which the base stock is to be imported, appear to be biologically ambitious. An additional limiting factor is that few Korean farmers have mastered the art of pasture management. Many pastures developed in recent years have been expensive to establish and short-lived; suitable pasture areas often are several kilometers apart. Consequently collection of milk from the scattered herds would be ex- pensive and, because roads are inadequate and collection times are slow, would require refrigerated tankers to reduce bacterial contamination and spoilage in the hotter periods of the year. The net benefits of de- veloping scattered, pasture-based dairy holdings, including effects on rural income and employment, should be weighed against the alterna- tives of importing powdered milk or lot-feeding animals with imported feedgrains. Demand in 1990 for pork and poultry is projected to be only double the 1976 production (table 7.7). Consequently the planned 50 percent increase in the output of these meats during the fourth plan period would, particularly if continued in the 1980s, lead to a large exportable surplus. Biological constraints will not impede achievement of this in- creased output, but export markets will have to be developed. Potential markets exist in Japan and Hong Kong, but their exploitation may re- quire great attention to quality. Because most production is likely to be on a large scale, specialized production units could provide the requisite quality control. Although increased pig and poultry production would depend heavily on imported high-protein feed, the foreign exchange cost would be offset by the export of meat. Further investigation of export markets, likely fluctuations in market demands, and the need for higher quality control are warranted. If all of Korea's meat and milk requirements in 1990 are to be met from domestic production, an estimated 70 percent of the nutrients AGRICULTURE 223 required, or more than 3.5 million tons of feedgrain, would have to be imported. Because of the high import content, government should weigh the benefits and costs of domestic production of livestock products against those of directly importing meat and dairy products before em- barking on a major production program. Preliminary analyses indicate that a beef industry based on imported feedgrains is unlikely to be economically viable. The import cost of beef that is equivalent in quality to Korean beef is projected at roughly $1.40 a kilogram in 1976 dollars. It takes about 7 kilograms of grain to produce 1 kilogram of beef sold. Making a conservative allowance for other production costs, the ciF cost of imported grain would have to be less than $150 a ton if domestic production is to be economically justifiable." Large-scale do- mestic production of beef therefore appears to be of marginal economic benefit. Further analysis of the economic benefits and costs of beef pro- duction should be carried out, as should separate analyses for each of the other animal protein products, and any expected improvement in productivity should be taken into account. Sustaining einployment and equity Of Korea's 1975 population of 35.3 million, 13.2 million lived in farm households. In the farm sector there were 8.5 million persons over 14 years, of whom 5.7 million were classified as economically active and 5.1 million were engaged in agriculture and forestry. Despite widespread migration from rural areas, the work force and the number of hours worked per week in agriculture have increased since 1968. Although it is difficult to interpret the statistics for hours worked because of the seasonal nature of agricultural production, the upward trend is clear. The explanation for this trend, in the face of fairly constant grain output, increasing mechanization, and declining cropping intensity, is not clear. Mechanization may have had little impact on the number of workers employed in agriculture. Although percentage increases in the number of power tillers and power threshers have been spectacular, these in- creases have been from a small base. Only one household in forty has a power tiller. Power tillers seem, moreover, to be used mostly for trans- port. When they are used for tilling they require, because of their 9. Other production costs include health costs, meat preparation costs, and 5 percent mortality, but not labor or maintenance costs. 224 PART THREE: MAJOR SECTORS weight, two or three men to operate them over a twelve-hour day. One man can work all day with a draft animal. Thus, although the power tiller is somewhat faster, it actually increases the demand for labor and its net effect on labor input may be small. A second factor affecting employment in agriculture may be the participation of women in the agricultural labor force. The proportion of women employed in the rural labor force is greater than that in the urban labor force. In addition, the proportion of women in the agricultural labor force has been increas- ing; it rose from 38 percent in 1963 to 43 percent in 1974. If women replace men in such physically demanding tasks as cultivation and manual transport, more women, more hours of wvork, or a combination of both is needed to produce a given unit of output. A third factor may be the considerable expansion of fruits, vegetables, and other cash crops since the mid-1960s. All these crops are labor intensive and not readily mechanized. They require, and will continue to require, substantial labor inputs. OFF-FARM EMPLOYMENT. Under the New Community movement, two types of off-farm employment opportunities, in addition to work on village improvement, are being developed for rural households.'0 The first of these is the encouragement of small-scale craft and other industry at the village level-for example, the manufacture of a wide range of articles from rice straw. The second is the creation of a New Community factory in each district. By 1975 some 420 factories had been built. These factories produce such goods as textiles and electronic products and employ from 50 to more than 300 persons; average employment in 1975 was 80 persons a factory. The success of these factories has been varied, partly because of underfinancing and poor management. In some instances the management deficiency has been overcome by creating an association with a well-managed urban company. The growth of New Community factories, along with programs to reduce urban growth by decentralizing industry, should maintain the number of persons living in rural households and increase the proportion of household income from nonfarm sources. If anything, however, the proportion of eamings from off-farm employment is falling (table 7.13). 10. In Korea off-farm or nonbusiness employment means any work for salary or wages not performed on one's own farm. No distinction is made between casual work for another farmer and working in a factory in the rural environment, even though employment in a factory might effectively remove a person from the agricultural sector. AGRICULTURE 225 Table 7.13. Composition of Farm Household Receipts, by Source, 1969-75 (percent) Source of income 1969 1971 1973 1975 Agriculture 76.7 81.9 81.2 81.9 Side business 3.7 3.1 3.2 2.5 Nonbusiness 19.6 15.0 15.6 15.6 Source: MAF, Report on the Results of Farm Household Economy Survey (Seoul. 1975), p. 68. It dropped from about 23 percent of total farm household receipts in 1969 to 18 percent in 1975. This apparent anomaly may be explained in part by the high price of rice after 1968 and its effect on the value of household grain inventories. LABOR SUPPLY AND DEMAND. Long-range forecasts of the supply of agricultural labor, based on high growth rates for industry and manufac- turing, indicate that the supply will increase slowly until the mid- 1980s. Toward the end of the 1980s, the supply of labor will decline. In order to examine whether labor availability will pose a serious con- straint on agricultural output during the 1980s, labor requirements have been estimated by taking into account the seasonal needs of each crop and the changing pattern of crop plantings (table 7.14). The estimates allow for an increase in vegetable production, which is relatively labor intensive, and for a constant base demand of about 2 million man-years in 1975 and 2.5 million man-years in 1990. Table 7.14. Agricultural Labor Requirements, by Month, Estimates for 1975-90 (millions of man-years) Month 1975 1990 May 3.56 4.02 June 6.64 6.05 July 4.39 3.10 August 3.43 3.83 September 3.47 3.83 October 4.41 4.71 November 5.17 5.17 Source: World Bank calculations. 226 PART THREE: MAJOR SaCTORS The demand in 1990 is based on the assumption that the use of light power tillers, contract threshing, chemical weedicides, and small har- vesters will be widespread. The current labor force can and does supply the needs of peak demand, but it might be preferable to regard the July and October requirements of about 4.4 million workers in 1975 as the normal labor supply. On this basis the labor demand in June, October, and November in 1990 would exceed the normal supply of labor now available. If these peaks can be met now, it would seem that they can be met in the future, especially because the supply of agricultural labor is expected to increase somewhat. The distribution of labor demand projected for 1990 is more even than that estimated for 1975. In addition, the demand for labor would fall substantially if land preparation, transplanting, and harvesting were to be further mechanized. Under these circumstances, a much smaller work force would be needed in 1990 to produce Korea's agricultural output. Furthermore, if beef and milk were imported or if beef and milk production were concentrated under feedlot conditions, the demand for labor would be even less. Care must therefore be taken to ensure that employment opportunities outside agriculture will be available, in both urban and rural areas, before undertaking major programs to re- duce agricultural labor requirements outside the peak seasons. PRODUCTIVITY AND INCOME TRENDS. Government has placed consid- erable emphasis on raising rural incomes to the level of those in urban areas, in part as an incentive for agricultural workers to remain in the rural sector. The Ministry of Agriculture and Fisheries (MAF) and the Economic Planning Board (EPB) have for many years conducted studies of rural and urban household expenditure. Because rural household incomes fell far below urban bousehold incomes in 1967, government stepped up investment in rural areas to raise agricultural labor produc- tivity and took steps to improve the agricultural terms of trade. It may be argued, however, that the high purchase price of rice does not help the 34 percent of farm households with less than one-half hectare of land, because such households consume most of their rice production. In fact, because high prices are a factor in increasing the cost of labor and consequently the prices of other goods, high rice prices may ad- versely affect the poorest farm families. Government comparisons are usually made on a household basis, but comparing per capita or per worker income with consumption might be more appropriate (table 7.15). Rural households are slightly larger, 5.6 persons compared with 5.2, and they have an average of 2.9 working AGRICULrTURP 227 Table 7.15. Annual Income of Farm and Urban Households, 1963-75 (thousands of won) Item 1963 1967 1971 1975 Farm households Household income 93 149 356 873 Per capita income 15 24 61 155 Per worker income 29 48 122 305 Urban householdsa Household income 80 248 452 856 Per capita income 14 45 86 166 Per worker income 67 192 337 639 Ratio of farm income to urban income (percent) Household income 116 60 79 102 Per capita income 107 53 71 93 Per worker income 43 25 36 48 Sources: MAF, Farm Household Survey, 1975; iPB, Monthly Statistics of Korea, January 1976. a. Households of salary and wage earners only, in all cities. members compared with 1.3 for urban households. It should also be noted, however, that the statistics underlying table 7.15 suffer serious limitations. There are shortcomings in the definition of incomes and the coverage of the two surveys. The urban survey is confined to wage and salary earners, not extended to all income earners, and it excludes those who earn more than a specified ceiling. The rural survey appears to define income inclusive of changes in the book value of inventories and thus to overstate rural incomes in periods of rising inventories and grain prices. These shortcomings tend to overstate the increase in rural incomes in relation to urban incomes. Such a bias is confirmed by cross- checks using national accounts data and comparing consumption, not income."' These studies lead to a conclusion very different from that of official comparisons and indicate that per capita consumption in urban 11. See Bertrand Renaud, "Economic Growth and Income Inequality in Korea," World Bank Staff Working Paper, no. 240 (Washington, D.C.: World Bank, February 1976). 228 PART THREE: MAJOR SECTORS areas still is about double that in rural areas. Probably the relative position of urban and rural households has not changed much in recent years. Because of expected productivity gains in manufacturing, con- tinuing efforts will be needed to increase farm labor productivity, create nonfarm rural jobs, and expand rural services if the position of rural households relative to urban households is not to deteriorate. Mechanization The power tiller can increase the rate of tillage and theoretically re- duce labor demand during peak periods. But if power tillers are to reduce labor requirements in peak seasons and increase the rate of tillage, lighter machines must be developed and marketed. Moreover, if lighter machines cannot be used for transport, they must be much cheaper than current models because the hours of field work would be insufficient to make their use financially attractive to farmers. Current models have the capacity to farm 3.5 to 5 hectares. The average farm size is only 0.9 hectare, and the legal limit on paddy land is 3 hectares. A smaller, even slower, tiller might be more suitable for the conditions in Korea. Four-wheel tractors are uncommon in Korea-there were only 790 in 1976-and unlikely to be economic on terraces because most paddy fields are small. Reclaimed tidal land could be laid out to facilitate the operation of tractors. But even on the plains, it would be uneconomic for a farmer to own a tractor for use on a farm legally limited to 3 hectares. Studies should therefore be conducted to determine appro- priate means of leasing or contracting tractor services on a cooperative basis. Because of the need to reduce seasonal labor peaks while maintaining a substantial rural work force during the other ten months of the year, mechanization should focus on machines that reduce the requirements for labor during peak seasons. Examples are harvesters, transplanters, driers, and machines that speed land preparation and handle straw left by the previous crop. These machines would greatly increase labor pro- ductivity during peak periods, but one notable problem is that they can be used for only 10 or 20 days a year. Consequently, if a single farmer is to purchase them, they must be low in cost and capacity. If groups of farmers are to lease, contract, or own such machines, their capacity would have to be high and the timing of their use would be critical. Current research on mechanization is directed toward solving these problems. Nevertheless power tillers, which are too large for cultivating most paddy fields, are still being intensely promoted. Despite the capac- AGRICULTURE 229 ity of these tillers to work 40 percent faster than draft animals, few farmers obtain these rates, and then only by using several operators. If the objective of mechanization is to reduce peak labor demand, gov- ernment should seriously consider shifting its emphasis and resources to the development of machines that are better suited to the require- ments of Korean farmers. Prices and subsidies Government established the Fertilizer Fund and the Grain Manage- ment Fund to stimulate agricultural production, raise rural incomes, and cushion the impact of inflation on urban consumers. As the cost of imports increased in 1973 and 1974, deficits in both funds grew rapidly; the deficits in 1974 and 1975 were equal to about 2 percent of GNP. Government subsequently acted to contain these deficits. In December 1975 it increased the selling price of fertilizer by about 80 percent. This increase was sufficient to eliminate the deficit in the Fertilizer Fund, and current policy is to adjust selling prices to reflect changes in costs. The decline in the wvorld market price of wheat enabled govemment to eliminate the consumer subsidy on wheat. Domestic purchase and release prices for rice and barley are also being modified. The objective is to eliminate the deficit in the Grain Management Fund by 1981. FERTILIZER PRICE INCENTIVES. The introduction of Tongil and other improved crop varieties, the reduced supply of labor for weeding, and the price subsidv before December 1975 all had a substantial impact on the use of fertilizers and agrochemicals. Between 1965 and 1975 the consumption of nitrogen increased from 218,000 to 482,000 nutrient tons, phosphorus from 124,000 to 238,000 nutrient tons, and potassium from 52,000 to 167,000 nutrient tons. Insecticide application over the same period increased from 3,200 kilograms to 81,000 kilograms; herbi- cide application from a mere 25 kilograms to more than 25,000 kilo- grams.12 Demand for fertilizer at the subsidized price consistently ex- ceeded supply, and this made it necessary for MAF to allocate supplies to each region on the basis of expected crop plantings and application rates recommended by MAF's Office of Rural Development. Farmers re- ceived premixed fertilizer and were unable to select their own rates of 12. Ministry of Agriculture and Fisheries, 1977 Yearbook of Agriculture and Forestry Statistics (Seoul, 1977). 230 PART THREE: MAJOR SECTORS application for nitrogen, phosphorus, and potassium. Officially they could not switch between crops. In fact, farmers bought fertilizer from each other or diverted it to other crops on their own farms. Rice pro- duction does not seem to have been affected. The average yield of 4.3 tons a hectare for the 1976 rice harvest was the highest ever attained in Korea despite difficult seasonal conditions and the elimination of the fertilizer subsidy. As a result of the fertilizer price increase, consump- tion fell from the record 1975 levels to 361,000 nutrient tons of nitrogen, 142,000 nutrient tons of phosphorus, and 140,000 nutrient tons of potas- sium in 1976. Although the price increase has eliminated excess demand for the time being, demand for fertilizer is expected to continue to grow. The irrigated area under higher yielding varieties of rice is still increas- ing. Farmers have been getting higher incomes. Labor costs have been rising, and this may cause farmers to substitute chemical fertilizer for labor-intensive compost. Consequently it is probable that fertilizer use has not peaked, but will continue to increase, although possibly at a slower rate than in the past. GRAIN PRICE INCENTIVES. . The Grain Management Fund (GMr) has two basic purposes: to buy and sell foodgrains, both domestic and im- ported, in such a way as to provide adequate production incentives for farmers; to hold down consumer prices, particularly in urban areas. The prices set for the purchase of grain from farmers, together with the prices farmers pay for inputs, largely determine rural household income. The reasons are that the portion of income derived from grain is large and private market prices tend to follow the GMF price. Deficits in GMp occur in two ways: when the release price of a unit of domestically produced or imported grain is less than its cost, including the costs of transport, handling and administration, storage, spoilage, and interest; when the purchase cost of a unit of inventorv exceeds sales receipts from a unit of inventory. The book value of inventory changes can vary markedly depending on the accounting method used for inventory valuation. Deficits and surpluses are thus determined by the loss or profit per unit and the volume traded. With the exception of the CIF price of imports, the purchase and release prices and the volume traded through GMF are variables subject to government policy. The following discussion, which is based on rice but is essentially applicable to other grains, considers some of the effects of changes in these variables. As long as government requires GMF to purchase rice from Korean farmers for more than the release price to consumers, GMIF can reduce its deficit only by minimizing domestic purchases. But if the amount AGRICULTURE 231 purchased at the support price is too small, the private farmgate price might fall below the support price, which would increase the pressure on GMF to increase its share in the market. Because the official release price for traditional varieties in 1976 was 20 percent below the private market price to consumers, this subsidy tended to reduce the average consumer price of rice, lead to more consumption, and increase demand for imports. Thus, if GMF tries to minimize losses by reducing domestic procurement, the result could be lower rural incomes, increased rice consumption, and more imports. The dilemma is clear, but it is impos- sible on the basis of present information to estimate the minimum level of GMF domestic purchases needed to support the farmgate price. The second way to reduce the deficit is to reduce the government purchase price from farmers. Although farmers retain much of their own production for family consumption, the effect of a price reduction of W5,000 per bag would amount to about a 6 percent decline in farm household income on an average farm of 0.9 hectare. The associated loss of cash income might also cause farmers to reduce their use of inputs and possibly reduce household income by as much as 8 or 9 percent. Thus, for each W5,000 decrease in the farmgate support price, average income of farm households could drop by $115 to $170 below the 1975 level of $1,900. For the country this loss of income could amount to betwveen $270 million and $400 million. In 1976 GMF ac- quired about 10 million bags. A price reduction of W5,000 per bag would have saved GMF only $103 million, or about 20 percent of the GMF deficit for all grains. Using this variable alone is thus not a viable policy option for eliminating the GMF deficit. It would probably cost the private sector more than it would save government. The third way to reduce the deficit is to raise the release price of rice from government stocks. Such an increase would have little effect on the rural poor, but would cause consumers to substitute barley and wheat for rice. If consumers did this, imports of those grains and the stocks of rice held by GMF would increase. A substantial rise in price would also lead to pressure to increase wages so that real standards of living could be maintained. Such wage increases would be inflationary, but so are the GMF deficits. A reduction in the deficit would either re- lease funds for use by other sectors or reduce inflationary pressure, thereby offsetting some of the inflationary effects of higher wages. A related issue is the need to adjust the cMF release price seasonally. The current policy of uniform government release prices makes no provision for storage costs. Private traders are now able to sell rice at a maroin over GMF release prices sufficient to cover storage costs, either 0) 232 PART THREE: MAJOR SECTORS because they fine-mill to nine degrees, not the prescribed seven degrees, or because they sell more palatable varieties. If Tongil is replaced by high-yielding varieties that are more palatable, some marginal private traders might be forced out of business because they could no longer afford to pay farmers a premium over the government purchase price. If this happened, GMF would be forced to purchase a much higher pro- portion of traded rice to maintain the support price. Therefore, if the financial difficulties of GMF are not to be worsened, seasonal variations in prices will be essential to enabling recovery of storage and interest charges. Although this analysis of changes in government policy variables is highly qualitative, one thing is clear. If government actions to eliminate the GMr deficit are not to create more problems than they solve, more information is needed-about price elasticities of supply and demand for the commodities included in GMF and about the effects of price and quantity changes on the real incomes of producers and consumers-to provide definite answers to the questions raised by the foregoing dis- cussion. Chapter 8 Larry E. Westphal Manufacturing KOREA HAS ATTAINED SEMI-INDUSTRIAL STATUS by following an outward- looking strategy favoring two fundamental objectives: vigorous export expansion in highly labor-intensive products, and selective import sub- stitution in capital-intensive intcrmcdiate products and consumer dur- ables. Because of the successful pursuit of these objectives, the industrial sector is reasonably well diversified. But the emphasis given to exports of light manufactures has resulted in a rather high degree of dependence upon imports of many vital raw materials and intermediate inputs. The domestic value-added content of exports in 1976 was only about 50 percent. Much electronics production for export still is largely an assembly operation. Korea also imports a large share of the machinery and equipment required for investment. In some subsectors, however, backward linkages from production for the export and domestic markets have been developed: for example, from textiles through synthetic fibers to the production of basic petro- chemicals; from television assembly to the production of basic com- ponents; and from automobile assembly to the production of such com- ponents as engines and axles. Nor has industrial shallowness, where observed, necessarily been undesirable. Market size and comparative advantage limit the efficient exploitation of backward linkages. In this respect, industrialization in Korea has generally been consistent with its evolving comparative advantage in labor-intensive processing activi- ties. In addition, Korea's poor natural resource endowment has required that it remain dependent upon imports for many raw materials, includ- ing petroleum. Nevertheless there are areas where a deepening of Korea's industrial structure appears warranted. Opportunities have emerged as indus- trialization progressed, and Korea's comparative advantage has changed. The internal market for many industrial intermediate products has grown and will continue to grow, progressively enabling economies of scale to be realized to a greater extent. For example, the rapid growth of 233 234 PART THREE: MAJOR SECTORS domestic demand for petrochemicals is such that the second petro- chemical complex, which is to be built during the fourth plan period, will be roughly three-and-one-half times larger than the first, which was built to satisfy domestic demand toward the end of the second plan period. Whereas the first complex is not internationally competitive because of its size, the second complex will be. The same trend is evident in steel production. The capacity of the first integrated steel mill, completed in the middle of the third plan period, is to expand from 1 to 8.5 million tons during the fourth plan period. Furthermore the experience accumulated across the entire industrial sector can be transferred to lines of production not yet undertaken. The knowledge gained in the construction and shipbuilding sectors can be transferred to the fabrication of plants for the process industries; that gained in the expansion of mechanical engineering firms, to the production of deck and other equipment for ships. Korean policymakers have been concerned about these issues for some time. Their awareness of the desirability of deepening Korea's industrial structure was present in the formulation of both the second and third plans, particularly in the decisions to build the first integrated steel mill and petrochemical complex and to import-substitute selectively in other areas. These issues received even greater attention during the early 1970s when discussions to determine the appropriate industrialization strategy for the 1970s and 1980s were extensive. A long-term perspective plan covering the decade to 1981 was published by the Economic Planning Board (EPB) in 1973. While not abandoning export growth as a major source of industrial development, this plan promoted the accelerated development of the heavv and chemical industries as a critical element in Korea's future industrialization. Among the sectors given priority were basic metals and chemicals, electric and nonelectric machinery, ship- buildino, and electronics, all of which EPB classifies among the heavy and chemical industries. An ambitious investment plan was outlined for these sectors with the expectation that each would export a signifi- cant proportion of its total output by 1981. The long-term perspective plan was soon recognized to have been overly ambitious. Certainly, subsequent developments in the world economy made it so. The investment finance required to implement the plan would have been difficult to raise in 1974 and 1975. Furthermore the plan called for large-scale exports of basic metal and chemical prod- ucts, which are now recognized to be not in accord with Korea's com- parative advantage. MANUFACTURING 235 Korea's comparative advantage is shifting away from unskilled labor- intensive products toward skill-intensive products. If not a critical factor now, Korea will increasingly face competition from less industrially advanced countries where unskilled labor is cheaper. Korea is also be- ginning to experience limits on the expansion of certain traditional manufactured exports. Its dominant position in some markets, such as for plywood, makes it difficult to expand its share further. In other markets, such as for textiles and footwear, the developed countries are erecting trade barriers.' The preparation of the fourth plan thus involved a reassessment of the shift in emphasis of the industrialization strategy called for in the long-term perspective plan. Many large-scale, capital- intensive industrial projects identified in the perspective plan were dropped or postponed beyond 1981. Exports of skill-intensive manufac- tures were given increased importance. Because the fourth plan is the product of several years of intensive appraisal and discussion by govemment of past and future industrializa- tion strategy, it is particularly important that the plan be seen in its historical context. This is desirable in any event, the more so because government finally decided to continue following the same strategy on the basis of its remarkable past success. Vigorous export expansion in line with Korea's comparative advantage is the engine of growth under this strategy; import substitution is selectively used to develop individual subsectors supplying basic producer goods. An historical perspective is especially required on two distinct but interacting dimensions. The first concerns Korea's industrial structure: its efficiency and contribution to basic development objectives, and its internal balance, particularly with respect to backward integration. This dimension is the main subject of this chapter. The second concerns the structure of Korea's exports: the prospects for their continued rapid growth and their vulnerability to cyclical swings in the world economy. This dimension is treated in chapters 2 and 3 and in appendix B. The remainder of this chapter is divided into five main sections. The first three sections provide the historical perspective: they examine policies used to promote industrialization, delineate the role of the industrial sector in Korea's development, and introduce questions about the depth 1. For an elaboration of these points see Bela Balassa, "Korea's Development Strategy for the Fourth Five-Year Plan Period 1977-81" (Washington, D.C.: World Bank, 1975; processed). 236 PART THREE: MAJOR SECTORS of industrial structure. The fourth section summarizes the salient fea- tures of Korea's fourth plan for the industrial sector. The fifth and concluding, section evaluates the plan and highlights some potential un- resolved policy problems. Incentives to Industrial Expansion An incentive system that generally favored exports over import sub- stitution in manufacturing gradually came into being in the first half of the 1960s.2 Industrial incentive policies during the last half of the 1950s were those typically associated with an import-substitution strat- egy. By the mid-1960s, however, policyrnakers came firmly to accept that rapid economic development depended upon an export-oriented industrialization strategy. This view was predicated on the understand- ing that Korea's natural resource bdse was poor and the realization that further opportunities for import substitution were to be found only in intermediate and durable goods. The limited domestic market for these goods durincg the mid-1960s could not justifv establishing plants large enough to realize technological economies of scale. A major associated policy change was the lasting establishment of a uniform exchanoe rate in 1964, whcn the official won-dollar exchange rate was nearlv doubled for the second time in three years. As in the carlier devaluation in 1961, the 1964 devaluation was primarily intended to simplify the exchange regime and offset domestic inflation, not to increase the level of export incentives. In fact, exporters began to benefit from an expanded range of explicit export-incentive mechanisms in 1959, when government first granted tariff exemptions on imports of raw materials used in export production. Other price incentive mechanisms were gradually added, so that by 1967 exporters operated under a virtual free trade regime. They benefited from free access to imported inputs, indirect tax and tariff exemptions, and re- duced charges on overhead inputs as well as credit and direct tax pref- erences. The most important incentives to exporters in 1967 included: 2. Much of the discussion of the historical perspective is a substantially modi- fied and abbreviated version of Larry E. Westphal, "The Republic of Korea's Experience with Export-Led Industrial Development," World Development, vol. 6, no. 3 (1978), pp. 347-82. MANUFACTURING 237 * unrestricted access to, and tariff exemptions on, imported inter- mediate and capital goods; * exemption from payment of indirect taxes on major intermediate inputs, whether imported or domestically purchased, and on export sales; * generous wastage allowances in determining raw material imports exempted from duty and indirect tax, which permitted the use of some of these imports in production for the domestic market; * reduced prices for several overhead inputs, including electricity and railroad transport, which were intended to compensate in part for payment of indirect taxes included in normal charges for these inputs; * a 50 percent reduction in direct taxes on income earned from ex- porting, and provisions for accelerated depreciation; and * immediate access to subsidized short- and medium-term credit to finance working capital and fixed investment. In addition, the so-called exportimport link system entitled selected exporters to import certain popular items not otherwise approved for import. This system was used to subsidize exports during the late 1950s and much of the 1960s. It has since been used only intermittently on a more or less ad hoc basis to offset exporters' temporary losses arising from market fluctuations and the entry into new markets. Some incentives are not genuine subsidies. Except for instances of overrebating, the exemption of export sales from indirect taxes levied when the same products are sold domestically merely establishes tax neutrality. Under the destination principle, exports are exempt from indirect taxes, but such taxes are imposed on imports. In turn, exemp- tion from tariffs on imported inputs and from indirect taxes on all in- puts, regardless of the source of purchase, does favor exports over pro- duction for domestic sale. Permitting exporters to have free access to imported inputs and restricting this access in production for domestic sale does the same. In the context of world market prices, however, the effect of these measures is neutral: exporters purchase their tradable inputs and sell their exports at world market prices. To varying degree, depending upon the time period, currency over- valution added an implicit export tax that acted to offset genuine export subsidies. Overvaluation was the result of protection given to production for domestic sale, which increased the relative price of competing im- ports and hence reduced the demand for them. 238 PART THREE: MAJOR SECTORS A comprehensive study is needed to determine the net impact of the entire incentive system. Such a study has been made only for 1968.3 The degree of currency overvaluation in that year was found to be about 9 percent. Taking this into account, the study concluded that value added on the given volume of exports was nearly as large as it would have been under a free trade regime. This near equivalence indicated, at least for 1968, that genuine export subsidies were slightly more than offset by overvaluation. Indeed, the evidence suggests that genuine sub- sidies were varied in almost direct proportion to the degree of currency overvaluation between 1964 and the early 1970s. The study for 1968 also concluded, for all primary and manufacturing sectors taken together, that incentives provided to domestic sales and to exports were nearly equal. Nevertheless incentives were not uniform across industries. There was some discrimination against primary exports; in manufacturing the variation benefited less efficient exports. The less efficient, typically import-substituting sectors within manufacturing re- ceived higher-than-average export incentives. But incentives to domestic sales in these industries were even higher, and this pattern gave rise to an apparent bias of incentives against exporting. In contrast, low protec- tion in the domestic market led to negative incentives for domestic sales and gave rise to an incentive bias in favor of exports in the more efficient sectors within manufacturing. Although import restrictions have since been gradually liberalized, quotas and high tariffs continued to be in force to protect a number of small and inefficient industries. Moreover, through the instruments of its planning apparatus, government has selectively promoted import sub- stitution in a few nontraditional product lines. The most important of these lines have been producer goods and, of late, consumer durables and automobiles. Thus many explicit subsidy mechanisms used to pro- mote exports have also been applied to engender import substitution. In short, the government's strategy has not been characterized by purely neutral free trade. Incentive policies, particularly the instruments of protection, have discriminated in favor of agriculture and, within manufacturing, in favor of those sectors in which opportunities remained for substantial import substitution. Nevertheless protection in the do- mestic market has generally been quite low by international standards. 3. For the analysis underlying this and the following related statements see Larry E. Westphal and Kwang Suk Kim, "Industrial Policy and Development in Korea," World Bank Staff Working Paper, no. 263 (Washington, D.C.: World Bank, 1977). MANUFACTURING 239 By maintaining the exchange rate near the free trade level and granting exporters free access to imported inputs, government has been able to provide, on average, roughly equal incentives to production for domestic sale and for export. Although subsidies have not been provided on a completely uniform basis with respect to value added at world prices, traditional and more efliciently produced exports have not been dis- criminated against. In addition, by extending the complete package of export incentives to producers supplying intermediate inputs to ex- porters, government has not unduly stimulated the use of imported inputs by exporters. Instead, government has tended to promote efficient backward linkages from exports. This last point is discussed further in the next section. By avoiding an incentive bias against exports, government made an important contribution to the rapid expansion of exports. Between 1960 and 1975 the average annual growth of exports was nearly 30 percent in constant prices. In turn, Korean entrepreneurs consistently demon- strated their capacity to respond to profitable export opportunities. No- table examples of this capacity are the rapid rise of wig exports in the 1960s, the later expansion of exports of electronics and footwear, and the continuing dynamism of miscellaneous manufactures, the composi- tion of which is constantly changing. To offset inflation at higher rates than experienced in Korea's export markets, frequent devaluations and compensatory changes in export subsidies were used to maintain the real effective exchange rate faced by exporters at a relatively constant level between 1964 and the early 1970s.4 Between 1970 and 1973, however, the real effective exchange rate for exports increased nearly 30 percent, principally as a result of the appreciation of the Japanese yen. Exports in 1973 were more than two-and-one-half times their real value in 1970. In response to these and other, less transitory indications that export incentives were perhaps yielding excessive profits, government started to abolish a number of incentive mechanisms in 1973. The benefits of lower direct tax rates and automatic tariff exemptions on imported capital goods were with- drawn; wastage allowances and interest subsidies declined. Nevertheless the real effective exchange rate in 1974 and 1975 remained slightly higher than its average over the 1960s. Because of the world recession, not the change in incentives, the value of real exports fell by 2.3 percent 4. Effective exchange rates are used as an indicator of the total receipts from a dollar's worth of exports FoB or the total payments for a dollar's worth of im- ports cIF, with associated government subsidies or levies taken into account. 240 PART THREE: MAJOR SECTORS in 1974. Real export growth was nearly 15 percent in 1975; it regained its rapid pace in 1976, when exports in current dollars rose by 52 percent over their value in 1975. Trade Expansion and Industrial Development The atypically low share of exports in Korea's GNP in the 1950s sug- gests that the growth of exports might simply be interpreted as catching up with the international norm. If true, this would make the Korean case somewhat less interesting because it could not be claimed to typify an export-led industrialization strategy. What is needed is a comparison of Korea's economic structure with those of other countries of similar size and development level. It is not very useful to do this directly, however, for few countries in the world are similar to Korea in these respects. A more useful, albeit indirect, comparison can be made using the results of cross-country regression analysis aimed at determining the average, or norm, economic structure at different levels of development and according to country size. Tables 8.1 and 8.2 enable comparison of structural norms applicable to Korea for 1955 and 1972 with historically observed structural shares. Rather than a later year, 1972 is used to avoid the possible distorting effects of the world boom in 1973 and the subsequent world recession. Two sets of norm estimates are given. The first uses Korean values of per capita income, population, and the ratio of the current account deficit to total domestic resources to calculate the norms from cross-country regressions in which these variables appear as explanatory factors. The second also uses Korean values of per capita income and population, but arbitrarily assumes that the capital-inflow ratio is zero. Differences between these sets of norms reflect the effects of foreign capital inflows at the unusually high rates observed in Korea. In 1955 the share of industry (manufacturing plus construction) in GNP was somewhat below that which would be expected; the share of exports was very far below the average for a country of Korea's size and per capita income. By 1972 unusually rapid industrialization reversed the pattern. Industry's share was somewhat above the norm; the share of exports, when adjusted for the inflow of foreign capital, was nearly twice the norm. The figures shown in tables 8.1 and 8.2 suggest three conclusions. First, the share of primary production probably was atypi- cally low over the 1955-75 period. Second, the pace of industrialization was faster than in many other countries. Third, the unusually rapid MANUFACTURING 24 I Table 8.1. Observed Structural Shares in Korea, 1955 and 1972 (percent) Item 1955 1972 Per capita gross national product (dollars in 1965 prices) $79 $179 Capital inflow ratio 7.7 4.9 Share in gross domestic product Investment 12.0 20.8 Exports 1.7 21.0 Manufactured exports 0.4 17.8 Imports 10.0 26.1 Primary sectors 48.0 32.0 Industry 13.0 26.0 Utilities 3.5 7.5 Services 35.5 34.5 Source: Larry E. Westphal and Kwang Suk Kim, "In- dustrial Policy and Development in Korea," World Bank Staff Working Paper, no. 263 (Washington, D.C.: World Bank, 1977), table J. Table 8.2. Norm Structural Shares (percent) Actual capital inflow Zero capital inflow Per capita income Per capita income Item $79 $179 $79 $179 Share in gross domestic product Investment 14.4 20.2 12.8 19.2 Exports 9.8 10.8 16.0 14.8 Manufactured exports 1.4 2.9 0.5 2.3 Imports 17.6 15.8 16.1 14.8 Primary sectors 52.8 33.5 55.4 35.3 Industry 14.4 24.9 11.7 23.1 Utilities 5.2 7.1 5.6 7.4 Source: The norms are those estimated in the Chenery-Syrquin large country sample. Hollis B. Chenery and Moises Syrquin, Patterns of Development, 1950- 1970 (London: Oxford University Press, 1975). 242 PART THREE: MAJOR SECTORS growth of manufactured exports reflected that Korea was more than simply catching up to the norm after the dislocations caused by Japanese colonial policy and two wars. It also is interesting to obsetve that the growth of investment was too considerable and rapid to be attributed merely to high inflows of foreign capital.5 Sources of growth Input-output data can be used to calculate the contribution to changes in industrial output, sector by sector, of domestic demand ex- pansion, export expansion, and import substitution. Import substitution is defined in terms of the changing share, for each industrial sector, of imports in total supply. Seven input-output tables spanning the 1955-73 period at roughly three-year intervals are available for Korea. To insure that the analysis is robust, it is necessary to convert data for different vears to constant domestic or world prices. This has been done only for the tables covering 1955 to 1968. The following analysis is necessarily restricted to this period. In addition, because data are not available at the same level of detail for 1955, parts of the analysis cover only 1960 to 1968. In the top part of table 8.3, the sources of the growth of aggregate output are disaggregated to indicate the relative contributions of the primary, manufacturing, social overhead, and service sectors. Between 1955 and 1968 the growth of the manufacturing sector was responsible for more than one-half of the growth of aggregate gross output. The growth of manufactured exports directly accounted for 16 percent (8.4 divided by 53.2) of the increase in manufactured output. The estimates in the total column include the indirect contribution arising from in- duced changes in intermediate demand.0 Thus the expansion of exports, 5. In relative terms 1972 was a recession year in Korea, which explains the investment rate's proximity to the norm in that year. From 1969 to 1971 invest- ment was more than 28 percent of GDP. 6. Using the inverse Leontief matrix, the growth of demand for domestically produced intermediate inputs was separated into that caused by domestic final- demand expansion, export expansion, and import substitution. Each of these components was then added to the corresponding direct contribution to obtain the total contributions. (In the estimates of total contributions, the effect of input-output coefficient changes is included with that of domestic demand expan- sion.) All of the growth of demand for domestically produced intermediate inputs is counted under domestic demand expansion in the estimates of direct contri- butions. Thus the total contribution of domestic demand is less than its direct contribution. MANUFACTURING 243 Table 8.3. Sources of Growth of Aggregate Output, by Sector, 1955-68 (percent) Domestic demand Export Import expansion expansion substitution Sector Direct Total Direct Total Direct Total Total All goods and ser- vices 1955-68- Primary sectors 18.8 16.3 0.7 3.3 -2.3 -2.4 17.2 Manufacturinge 42.6 38.5 8.4 12.9 2.2 1.8 53.2 Social overhead 14.4 13.8 1.4 1.9 -0.0 0.1 15.8 Services 13.2 11.9 0.7 2.0 -0.2 -0.2 13.7 All sectors 89.0 80.5 11.2 20.2 -0.3 -0.6 100.0 Manufactured goods 1960-68' Exporting sectors 10.2 8.7 11.5 13.0 0.5 0.5 22.2 Import-competing sectors 23.6 22.4 .3 2.4 -1.8 -2.7 22.1 Nonimport- competing sectors 44.1 39.7 1.1 6.0 6.2 5.7 51.4 Other sectors 3.9 3.2 2.1 2.8 -1.7 -1.7 4.3 All manufactures 81.7 74.0 15.1 24.3 3.2 1.7 100.0 Note: The definitions chosen for assigning sectors to trade categories are as follows: in exporting sectors, more than 10 percent of output is exported; in import- competing sectors, more than 10 percent of domestic supply is imported; in non- import-competing sectors, neither the export nor import shares exceed 10 percent; in export-and-import-competing sectors (other) both shares exceed 10 percent. The classification is based on export and import shares in 1968. Figures may not rec- oncile because of rounding. Source: Westphal and Kim, "Industrial Policy and Development," table N. a. Based on data in constant domestic prices. b. Includes "other." c. Based on data in constant world prices. not just in manufacturing but throughout the economy, contributed 24 percent (12.9 divided by 53.2) to the growth of manufactured output. This comparison indicates that backward linkages from exports were strong. In contrast, import substitution, by which is meant a fall in the share of imports in total domestic supply, directly accounted for slightly more than 2 percent of the growth of manufacturing. Its total contribu- tion was even less, indicating that import substitution generated indirect 244 PART THREE: MAJOR SECTORS demands on sectors having higher-than-average requirements for im- ported intermediate imports. More than 50 percent of the growth of output in the exporting sectors from 1960 to 1968 was the result of export expansion; exporting sectors, in turn, were the source of more than 75 percent of the growth of man- ufactured exports and 20 percent of the growth of manufacturing (see the bottom part of table 8.3). The expansion of domestic demand was also important, as may be seen either in its contribution of nearly 75 percent of the growth of all manufactured output, or in the total magnitude of the contribution of the nonimport-competing sectors, in which exports accounted for less than 10 percent of output in 1968. Import substitution was the source of more than 10 percent of output growth only in sectors that had achieved nonimport-competing status by 1968. The important role of import substitution in some sectors is hidden in the aggregate figures presented in table 8.3. Import substitution in these sectors was offset in the aggregate bv negative import substitution in other sectors, that is, by a risino ratio of imports to domestic supply. Between 1960 and 1968 import substitution contributed more than 20 percent to the growth of twelve of the eighty manufacturing subsectors dis- tinguished in the analysis from which these estimates were drawn. Among these sectors, in order of the relative contribution of import substitution, were those producing fertilizers, petroleum products, sew- ing machines, electrical equipment and products, drugs, steel ingots, paper and paperboard, basic inorganic chemicals, and cast and forged steel. But import shares actually increased over the eight-year period, leading to negative import substitution, in thirty-nine of the manufactur- ing sectors and eight of twelve primary sectors. Export expansion was the source of more than 20 percent of output growth for twenty nmanu- facturing sectors. Included among these were sectors producing various textile products at different levels of fabrication, miscellaneous manufac- tured products, lumber and plywood, apparel and accessories, and elec- tronics and electrical equipment. At the same time the contribution of domestic demand expansion exceeded 80 percent in fifty-three of the manufacturing sectors. Thus the importance of domestic demand growth observed in the aggregate carried over to individual sectors as well. The pattern of manufactuLring growth from 1960 to 1968 featured widespread export expansion, concentrated in labor-intensive sectors, and selective import substitution, primarily in sectors producing basic inter- mediate products. A more recent study has analyzed direct growth con- MANUFACTURING 245 Table 8.4. Direct Contributions to Growth in Manufacturing, by Sector, 1960-73 (percent) Domestic demand Export Import Sector expansion expansion substitution Total 1960-68 Light industry 56.9 10.8 -1.0 66.7 Heavy industry 25.8 1.2 6.3 33.3 Total industry 82.7 12.0 5.3 100.0 1968-73 Light industry 40.1 21.8 -0.8 61.1 Heavy industry 30.3 9.1 -0.6 38.9 Total industry 70.4 30.9 -1.4 100.0 1960-73 Total industry 73.1 26.8 0.1 100.0 Note: Based on data in current prices. Totals may not reconcile be- cause of rounding. Source: Suk Tai Suh, "Import Substitution and Economic Develop- ment in Korea," Korea Development Institute Working Paper, no. 7519 (Seoul, 1975), table 5.5. tributions through 1973 (table 8.4).7 The results of this study unfor- tunately are not comparable with the estimates presented for the period through 1968. The sector classification and level of aggregation are different, and all estimates are based on data in current prices. They nevertheless document the continuation of the same pattern: widespread export expansion and selective import substitution. Moreover it is ap- parent that export expansion continued to increase in importance as a source of industrial growth. The results also indicate an increase since 1968 in the relative importance of heavy industrial sectors in the growth of manufacturing.8 7. Suk Tai Suh, "Import Substitution and Economic Development in Korea," Korea Development Institute Working Paper, no. 7519 (Seoul, 1975). 8. In Korean parlance the heavy industrial sectors comprise electrical and nonelectrical machinery as well as transport equipment and basic intermediate products. They thus include such products as precision instruments and electronics. 246 PART THREE: MAJOR SECTORS Sectors experiencing relatively substantial import substitution be- tween 1968 and 1973 included fiber spinning and textile fabrics, rubber products, chemicals, iron and steel, finished metal products, and non- electrical machinery. With the exception of the last two industries, import substitution was concentrated in the production of intermediate goods and took place largely between 1968 and 1970. Much of this import substitution was the result of the production of basic steel prod- ucts and petrochemical derivatives in medium-scale plants. Among the sectors classified as heavy industry, those contributing most to the expan- sion of exports between 1968 and 1973 were chemicals, steel products, finished metal products, and electrical machinery and appliances. The foregoing estimates understate the full contribution of export growth because the only indirect contribution included in the estimates in table 8.3, but not in table 8.4, is that arising from the derived demand for domestically produced intermediate inputs. These estimates under- state the full contribution of import substitution for the same reason. Two additional macroeconomic indirect effects may be distinguished: a multiplier effect arising from expenditure out of the additional income generated by the rise in production; a foreign exchange effect arising from increased production made possible by a rise in foreign exchange. receipts. These effects are most dramatic when factors of production would not otherwise have been employed. They also operate by increas- ing allocative efficiency whenever the domestic resource cost of exports, in shadow prices, is less than the shadow-priced value of foreign ex- change receipts. Note that either current or future consumption may be increased as a result, the latter because of the higher rate of in- vestment. Import substitution and backward linkages from exports The role of import substitution in Korea's industrial development deserves further comment. A useful starting point is the apparent anomaly in the relations between the shares of manufactured exports and industrial output in GDP. Returning to tables 8.1 and 8.2, it is found that Korea's share of manufactured exports in GDP in 1972 exceeded the norm by a far greater spread than did the share of industry in GDP. The difference is about 15 percentage points in the first case and only two or three percentage points in the second. This is largely explained by the absence of large-scale import substitution, which in tum is related to backward linkages from exports to the domestic production of inter- mediate inputs. MANUFACTURING 247 The share of domestic value added (direct and indirect) in Korea's manufactured exports was roughly 50 percent over most of the 1965-75 period.9 This pattern was not a consequence of the system of export incentives: the system has not discriminated between imported and do- mestically produced intermediate inputs (of the same quality) in the production of exports, except insofar as wastage allowances on imported inputs provide a bias in favor of domestically produced inputs. With protection in the domestic market and the overvaluation of currency, unrestricted access to and tariff exemptions on imported inputs would have led to a bias in favor of using imported inputs, unless this bias were somehow offset. But protection and currency overvaluation have both been low in Korea. In addition, government has used the domestic letter of credit (L/C), which gives the full range of export incentives to producers of intermediate goods supplied to exporters, to negate the remaining bias. Suppliers of exporters have thus been permitted access to intermediate inputs at world market prices, the same as exporters. By avoiding excessive subsidies to backward integration from exports, this system appears to have induced only those backward linkages that permitted efficient production.10 The high import content of Korean exports stems in part from the types of product being exported. Korea has the requisite natural re- source base for only a few manufactured exports, including silk textiles, cement, and ceramics. Most others, such as cotton and wool textile products, plywood, and steel products, require raw materials-cotton, wool, roundwood, iron ore, anthracite coal-the local supplies of which are either insufficient or prohibitively expensive. In several of these sectors, import substitution for intermediate inputs has been carried back to the stage where only the unprocessed natural resource product is imported. For example, natural-fiber yams and steel are domestically produced. But many exports, such as petrochemical-based products until 9. Charles R. Frank, Kwang Suk Kim, and Larry E. Westphal, Foreign Trade Regiimes and Economic Development: South Korea (New York: National Bureau of Economic Research, 1975), pp. 81-83. Data limitations make it virtually impossible to assess trends in the domestic value-added content of exports; on this point see David C. Cole and Larry E. Westphal, "The Contribution of Exports to Employment in Korea," in Trade and Development in Korea, eds. Wontak Hong and Anne 0. Krueger (Seoul: Korea Development Institute, 1975). 10. In some instances, say for petrochemicals, a case can be made that inefficient production has in part been established to supply exporters. But in such instances additional import-substitution incentives have been granted. Z48 PART THREE: MAJOR SECTORS recently and electronic assembly even now, require intemnediate inputs for which the Korean market has been too small for efficient scale pro- duction or for which sophisticated and costly technological know-how is needed. For most export products, Korea has exported the output of the labor-intensive processing stages. As noted above, however, the degree of backward integration in a number of product lines has also been considerable. Import substitution for intermediate products has, by and large, been carried back as far as is economical. The large contribution made by export growth to industrialization and the highly selective exploitation of backward linkages from exports help explain why import substitution has played such a subordinate part. Although some backward linkages from exports have been exploited through import substitution, the rapid change in the composition of exports has offset the import substitution that occurred. The result of this pattern is that the share of domestic value added in aggregate ex- ports has remained roughly constant. Government policy regarding imports of consumer durables and other luxury consumption goods has also influenced the minuscule contribu- tion of import substitution to overall growth. Imports of such goods as automobiles, refrigerators, and television sets have for the most part been prohibited. Nearly all of the latent domestic demand for these goods has been satisfied only after the start of domestic production. Thus government managed to delay the purchase of such goods in significant quantities until they were produced locally. This policy has given rise to a pattem to which the term, import substitution, does not strictly apply. The mathematics of calculating growth contributions implicitly recognizes this distinction by assigning a low weight to these sectors in the aggregate import-substitution estimates as a result of their initially small share in domestic demand. The rapid growth of foreign exchange receipts from exports has also allowed import substitution to be selective. This selectivity is reflected in the relatively small contribution of import substitution to aggregate growth. Not to be overlooked is the fact that pursuing an economy's comparative advantage is a matter not only of the composition of exports and imports, but also of their size relative to domestic production and consumption. The shares of exports and imports in GDP were abnormally low at the end of the Korean War. To pursue its comparative advantage, Korea had to increase the shares of both to atypically high levels. Import substitution naturally is small in aggregate when the share of imports is rising. Moreover selective import substitution enabled the concentra- MANUFACTURING 249 tion of scarce investment resources in a few sectors at a time. This con- centration, in turn, enabled greater exploitation of economies of scale and the linkages among closely interrelated activities.1l Thus, in most areas, import substitution was delayed until demand was sufficient to support plants of efficient scale. This is not to deny, however, that im- port substitution in other areas, notably in petrochemicals and automo- biles, probably was premature. It certainly appears to have been if learning-by-doing and other external-economy phenomena are neglected. Comparative advantage The determinants of comparative advantage are many and complex. They include the effects of natural resource endowment, the levels of labor and management skills, and the accumulation of these skills through learning-by-doing, risk and uncertainty, and a host of other factors which either are difficult to quantify or have not been quantified with sufficient precision in Korea. Nevertheless, to the degree that Korea's comparative advantage during the 1960s may be said to have been in labor-intensive activities, as opposed to capital-intensive activi- ties, a partial assessment is possible. This assessment is based on a simple two-factor model of static comparative advantage, which is insufficient to illuminate the full range of issues bearing on the country's compara- tive advantage. Thus merely finding that Korea exported labor-intensive commodities and imported capital-intensive commodities does not prove that resource allocation was generally efficient. But it does place the burden of proof on those who would argue otherwise. Examination of labor-capital ratios for exports and imports between 1960 and 1968 reveals that manufactured exports were substantially more labor intensive than imports of manufactures in every year.12 On the other hand, total exports were more capital intensive than the bundle of total imports. The contrast between the manufacturing and other sectors is explained by the complementarity of capital and natural re- sources, where the latter are excluded as a factor of production in the 11. In the context of a planning model, Westphal documents the potential gains from following such an investment pattern. Larry E. Westphal, "An Inter- temporal Planning Model Featuring Economies of Scale," in Studies in Develop- ment Planning, Hollis B. Chenery and others (Cambridge, Mass.: Harvard Uni- versity Press, 1971), pp. 60-86. 12. Westphal and Kim, "Industrial Policy and Development," pp. 4-31, 32, 46. 250 PART THREE: MAJOR SECTORS analysis. The effect of Korea's natural resource endowment on its com- position of trade in primary products has led it to export relatively capital-intensive minerals and to import highly labor-intensive agricul- tural products. Comparing estimates of direct factor intensity across years, it may be concluded that Korea's manufactured exports became more labor intensive over time, while its manufactured imports tended to become more capital intensive.'3 In fact, whereas Korea's manufac- tured exports were less labor intensive than the average for the entire manufacturing sector in 1960, they were more labor intensive by 1968. Manufactured imports were more capital intensive than the average for the entire manufacturing sector throughout the period. Trends in factor use Among the outstanding features of Korea's overall development per- formance is the increase in the rate of factor use. Korea has been rela- tively successful in finding employment for its labor force, which has grown at an average rate of 3.2 percent a year since 1960. This success is broadly indicated by the fall in the unemployment rate from a peak of 8.3 percent in 1962 to 4.1 percent in 1975. It is likely that unemploy- ment would have been serious, certainly in urban areas, if exports of light, labor-intensive manufactures had not grown so rapidly. This con- clusion can reasonably be inferred from estimates of the share of em- ployment attributable to exports presented in a recent study based on input-output data.14 Exports accounted for more than 25 percent of manufacturing employment and close to 10 percent of total employment in 1970. The contribution of export expansion to employment is even more impressive. The same study estimates that the growth of exports between 1960 and 1970 was responsible, directly and indirectly, for 38 percent of the growth of employment in manufacturing and 33 percent of the growth of total employment. Korea has also been successful in increasing the degree of capacity use in the manufacturing sector. The most reliable statistics for rates of capacity use are based on electricity consumption data that define 100 13. The labor-capital ratios on which these estimates are based are for a single year-1968. Thus changes over time in the estimated factor intensity of a particular aggregate are the result solely of changes in its composition. 14. Cole and Westphal, "Contribution of Exports to Employment," table 1. MANUFACTURING 251 percent capacity use as being equal to plant operation 24 hours a day and 365 days a year.15 On this basis, the aggregate rate of capacity use in manufacturing is estimated to have increased at an annual compound rate of 7.2 percent, from 18 percent in 1962 to 32 percent in 1971, the last year analyzed in this survey. For the measure used, 32 percent capacity use, which on average is roughly the same as single-shift opera- tion throughout all industry, is high by international standards, even when including developed economies in the comparison. Although it is not possible to state the degree to which the expansion of exports contributed to the increase in capacity use, there is little doubt that it was significant. Trends in the aggregate labor-capital ratio and factor productivity in manufacturing are also noteworthy.16 The labor-capital ratio in manu- facturing rose almost continuously during the first half of the 1960s, then fell almost continuously through 1972, and rose again in 1973. By 1973 the labor-capital ratio exceeded that in 1960 by more than 15 percent. Both the output-capital and output-labor ratios, in which output is measured by real value added, were rising continuously during this entire period. The output-capital ratio rose by 1973 to more than twice its value in 1960; the output-labor ratio increased by nearly 100 percent. Total factor productivity thus roughly doubled over thirteen years, giving an annual increase of more than 5 percent. Depth of Industrial Structure Is Korea's industrial structure undesirably shallow? Much is often made of the import intensiveness of Korea's exports, and further sig- nificance is attached to the predominance of light manufactures in exports. On the first point, it has already been noted that, for most 15. Young Chin Kim and Jene K. Kwon, "The Utilization of Capital and the Growth of Output in a Developing Economy: The Case of South Korean Manu- facturing," Journal of Development Economics, vol. 4 (1977), pp. 265-78, table 2, U, series. 16. Hong has analyzed growth in value added, employment, and capital stock in manufacturing for the 1960-73 period. Wontak Hong, Factor Supply and Factor Intensity of Trade in Korea (Seoul: Korea Development Institute, 1976), tables 7.6, A22, A27. 252 PART THREE: MAJOR SECTORS of Korea's principal exports, import substitution for intermediate inputs has been carried back as far as possible, given Korea's poor natural resource endowment. On the second point, the gains from exporting light manufactures should, insofar as light manufactures are labor- intensive and Korea's comparative advantage has been in labor-intensive products, be sufficiently clear. The import content of exports is the result of their composition, which in turn stems from Korea's compara- tive advantage in light manufacturing and processing. The concern for the shallowness of Korea's industrial structure that derives from the high import content of exports is thus misplaced. Even so, certain sectors of Korean industrv do exhibit shallowness. Electronics epitomizes what is found to varying degrees in other sectors. Electronics production is largely an assembly operation, and little infu- sion of the basic technological know-how has taken place. Korea exhibits the typical characteristics of an LrDC producer within the product cycle of an industry for which technological advance is rapid.17 In shipbuild- ing, too, Korea has yet to realize the full advantages of having a local design capability and of producing associated deck equipment, elements of the drive train, and so on. In addition, most of the machinery- producing sector is technologically immature. The basic metal sector epitomizes another kind of shallowness: high-volume, standardized products dominate the output mix. Specialty steels of various kinds, such as those required to make precision machinery, are not now pro- duced; Korean steel mills concentrate on output of standard specifica- tions and shapes. Nor is high-quality grey iron, of the type required in quality castings, produced in sufficient quantitv. The idea here is not to document comprehensively the shallowness of the industrial struc- ture where it exists, but to demonstrate its characteristics. In some sectors it is due to technological backwardness; in others it is the result of in- sufficient product differentiation toward specialtv products in the mix of outputs. Neither characteristic can be captured in aggregate statistics, even at the sectoral level. To say that parts of Korean industry are shallow in these dimensions is not to imply that industrial development should have taken a different course. Indeed the pattern in Korea is found in the industrial sectors of many countries that are more advanced, at least when measured by per capita income. This pattern nevertheless defines what perhaps is 17. Sang Chul Suh, "Development of a New Industry through Exports: The Electronics Industry in Korea," in Trade and Development in Korea, pp. 103-22. MANUFACTURING 2 53 the principal challenge of industrial development during the period of the fourth plan and points to some changes that may be expected in the character of industrial development. Manufacturing under the Fourth Plan The average growth rate of GNP during 1976-81 is projected to be about 9 percent, or slightly less than the average figure of about 10 percent for 1965-75.18 This forecast reflects anticipation of a relative tightening of constraints imposed by the availability of investment re- sources. The annual growth of value added in manufacturing is expected to decline to about 14 percent from 20 percent. Commodity exports are also forecast to grow less rapidly than in the past. Their annual growth rate is expected to decline to 16 percent during the fourth plan period from 34 percent during the third plan period. But, insofar as export-led manufacturing growth continues to be the fundamental element of Korea's development strategy in the fourth plan, there really is no break with the past. Indeed, compared with the performance over the 1965-75 period, the relative importance of both manufacturing activity and manufactured exports is expected to increase (table 8.5). The share of manufacturing value added in GNP is projected to rise from 28 percent to 40 percent. Manufacturing sectors are to provide 54 percent of the increment in GNP over the fourth plan period, compared with 43 percent during 1965-75. Similarly manufacturing is to generate more than 55 percent of additional employment, and the share of man- ufacturing employment in total employment is expected to rise to 26 percent. In addition, exports of manufactures are to continue to rise more rapidly than exports of primary products. The contribution of net manufactured exports to the growth of GNP is projected to more than double, from 17 percent to 36 percent. For imports, estimates in the plan indicate that the ratio of commodity imports to GNP is to rise from 36 percent to 41 percent in 1975 prices. To go from these figures to esti- mates of the contribution of import substitution requires further infor- mation on the sectoral composition of demand and supply. These pro- 18. All figures cited in this section are from the preliminary draft of the fourth plan. They do not differ substantially from figures in the plan's final version, which is itself subject to constant revision. EPB, The Fourth Five-Year Economic Development Plan 1977-81 (Seoul, 1976). 254 PART THREE: MAJOR SECTORS Table 8.5. Structural Indicators, 196541 (percent) From historical dataa Share of Indicator 1965 1975 increment Share of manufacturing value added in GNP 13.9 32.1 42.8 Share of manufacturing in total employment 8.4 19.1 42.8 Share of exports in GNP 5.2 28.4 42.1 Share of net manufactured exports in ONP' 1.1 10.9 16.7 Share of commodities in total exports 70.0 93.1 93.1 Share of manufactures in commodity exports 58.9 84.8 86.3 Ratio of imports to GNP 9.7 27.5 37.9 Share of commodities in total imports 93.3 91.8 91.6 Sources: Bank of Korea (Box), National Income in Korea, 1975; Box, Economic Statistics Yearbook, 1976; EPa, The Fourth Five-Year Economic Development Plan, 1977-81 (Seoul, 1976). a. In 1970 prices. jections were not available in the required format in the plan documents. According to World Bank estimates, the contribution of import substi- tution to the growth of manufacturing output implicit in the plan is less than 10 percent. Investment composition The share of total investment allocated to the manufacturing sectors is to rise to 27 percent in the fourth plan period from an estimated actual share of 25 percent in the third plan period.19 In comparison with the planned composition of investment among major manufacturing sectors under the second and third plans, the proportion of investment planned for the heavy industrial sectors-basic metals, chemicals, and machinery, electronics, and shipbuilding (MES)-is to increase further in the fourth 19. Investment figures for the fourth plan are on an outlay basis; those for the second and third plans are not. The share of manufacturing in total investment for the fourth plan is thus overstated by roughly 2 percentage points when com- pared with figures for the earlier plans. See table 8.10 for the comparison. MANUFACTURING 2555 From the fourth planb Share of 1975 1981 increment Indicator 28.4 39.7 54.1 Share of manufacturing value added in GNP 19.1 26.1 55.4 Share of manufacturing in total employment 31.4 52.3 78.9 Share of exports in GNP 11.4 22.0 35.5 Share of net manufactured exports in GNIP 85.0 80.9 78.8 Share of commodities in total exports 84.9 92.0 95.9 Share of manufactures in commodity exports 42.6 52.2 64.4 Ratio of imports to GNP 83.4 79.1 75.4 Share of commodities in total imports b. In 1975 prices. c. The historical figures for manufactured exports were obtained by multiplying the ratio of manufactured to total commodity exports in current prices by the constant price figures for commodity exports. Manufactured export figures from the fourth plan exclude processed food. Net manufactured exports are equal to the domestic content of manufactured exports. plan period (table 8.6). This increase will be offset by a reduction in the share of investment planned for the light manufacturing sectors other than textiles. The actual composition of investment does not, of course, have to be the same as the planned composition. At a somewhat lesser level of aggregation, the estimated actual allocation of investment during the third plan period can be compared with the fourth plan's targets (table 8.7). Here it is observed that investment is to shift awav from the basic metal and chemical sectors in the fourth plan period, not toward them. Investment during the third plan period was more heavily concentrated in both of these sectors than was originally expected. Using the same basis for comparison, the relative magnitude of investment in the textile sector was more than that planned in the third plan, and a reduction in the share of investment in textiles is planned for the 1977-81 pe- riod.20 The share of investment for the other light manufacturing sectors is to remain approximately constant. The shares of machinery and 20. Throughout this section the figures for textiles include finished clothing and accessories. 256 PART THREE: MAJOR SECTORS Table 8.6. Composition of Planned Fixed Investment in Manufacturing, by Sector, 1967-81 (percent) Second plan Third plan Fourth plan Sector 1967-71 1972-76 1977-81 Basic metals 12.5 14.3 16.2 Chemicals' 30.7 17.4 22.5 Machinery, electronics, and shipbuilding 11.6 16.8 25.4 Textiles and clothing 23.6 16.5 19.9 Other light manufacturing 21.6 35.1 16.0 Note: The figures for the second, third, and fourth plans respectively are in 1965, 1970, and 1975 prices. Sources: EPB, Second Plan; EPB, Third Plan; ErB, Fourth Plan. a. Includes cement and other glass, clay, and stone products. Table 8.7. Composition of Actual and Planned Investment in Manufacturing, by Sector, 1972-81 (percent) Actual Planned investment investment Sector 1972-76 1977-81 Basic metals 19.4 16.2 Chemicals' 23.6 22.5 Machinery 9.6 16.9 Electronics 4.5 6.9 Shipbuilding 4.3 1.6 Textiles and clothingb 22.4 19.9 Other light manufacturing 16.2 16.0 Note: Data are in 1975 prices and include only fixed in- vestment. Source: EPB, Fourth Plan. a. Includes cement and other glass, clay, and stone products. b. Includes footwear. electronics are to increase by more than 75 percent and 50 percent re- spectively; that of shipbuilding is to fall. Thus the most pronounced changes in the direction of industrial investment are the rise in the shares allocated to machinery and electronics and the fall in those allo- cated to basic metals, shipbuilding, and textiles. MANUFACTURING 257 Output structure The anticipated evolution of the structure of manufacturing activity naturally reflects the altered composition of investment. Basic metals and chemicals together will account for a somewhat smaller share of manufacturing value added in 1981 than they did in 1975 (table 8.8). Table 8.8. Composition of Manufacturing Value Added, by Sector, 1960-81 (percent) From the From historical dataa fourth plan' Sector 1960 1965 1970 1975 1975 1981 Basic metals 3.5 4.3 3.7 3.9 2.6 3.7 Chemicals' 7.8 15.3 22.4 17.6 25.0 21.7 Machinery, electronics, and shipbuildingd 6.2 9.8 10.7 15.8 14.8 24.2 Textiles and clothinge 27.8 22.4 20.7 28.7 25.4 24.1 Other light manufacturing 54.7 48.2 42.5 34.0 32.2 26.4 Sources: BOK, National Income in Korea, 1975; EPB, Fourth Plan. a. In 1970 prices. b. In 1975 prices. c. Includes cement and other glass, clay, and stone products. d. Historical data exclude optical and precision instruments. e. Includes footwear. The share of MES sectors is to increase from 15 percent to 24 percent at the expense of textiles and at the even greater expense of the other light manufacturing sectors. EXPORTS. One remarkable feature of Korea's industrialization strategy is the linkage between changes in industrial structure and the expan- sion of exports. In the past most import-substituting industries developed exports shortly after they were established. The pattern is to continue under the fourth plan and become more pronounced if the plan's pro- jections are realized. The share of MES sectors in total manufacturing output is to rise from 13 percent to 21 percent; their share in manufac- tured exports is to rise from 19 percent to 33 percent, with the machinery sector playing the leading role (table 8.9). The share of basic metals in 258 PART THREE: MAJOR SECTORS Table 8.9. Composition and Growth of Manufactured Output and Exports, by Sector, 197541 (percent) 1975-81 1975 1981 annual growth rate Sector Output Exports Output Exports Output Exqports Basic metals 4.6 8.5 6.2 8.0 23.4 19.0 Chemicalsa 24.5 6.7 22.2 9.3 15.4 26.9 Machinery 6.6 6.7 10.1 10.9 26.0 30.1 Electronics 5.0 9.5 8.4 14.9 27.9 29.6 Shipbuilding 1.7 3.2 2.1 7.0 21.4 36.9 Textiles and clothing' 25.7 46.6 24.2 33.7 16.2 13.9 Other light manufacturing 31.9 18.8 26.8 16.2 14.0 17.5 Notes: Data are in 1975 prices and exclude processed food. Source: EPE, Fourth Plan. a. Includes cement and other glass, clay, and stone products. b. Includes footwear. exports will fall slightly, but its share in output will increase. This difference is attributable to the anticipated rapid growth of domestic demand for basic metals to be used as intermediate inputs in the ma- chinery and shipbuilding industries. In contrast, the export share of the chemical sector is to increase by nearly 39 percent as its output share declines. Exports of cement, flat glass, fertilizer, and various petrochemicals are responsible for this result. For fertilizer and petrochemicals it appears that temporary exports are to be used to overbuild capacity in relation to the size of the domestic market when plants are brought on line. Exports of basic metals, at least in part, may have the same rationale. As a strategy to achieve greater economies of scale, this has much to recommend it. But the efficiency of continuing to export these products is open to serious question. The share of light industries in manufacturing output is to fall from 58 percent to 51 percent. The decline in their share of manufactured exports from 65 percent to 50 percent is projected to be even more precipitous. The adjustment on the export side is largely concentrated in textiles, as well as in footwear and plywood in other light manufac- turing. There is some question whether the projected concentration of the realignment of output in the other light manufacturing sectors is MANUFACTURISG 259 realistic. Any shortfall in MBS exports is likely to be compensated for by expanded exports of other light manufactures, a heterogeneous mix of products containing a number of Korea's most rapidly growing exports. Despite the change in the composition of exports, the ratio of exports to gross output is expected to rise in every major manufacturing group, but by much more in MEs sectors than in the others. The plan's projec- tions indicate that the heavy sectors will catch up with the light sectors in their dependence upon the external market by 1981. The light sectors have heretofore exported a much larger share of output. The result, as previously indicated, is that export expansion is to contribute propor- tionately more to the growth of manufacturing output under the fourth plan than under the third plan. This contribution reflects Korea's in- creasing dependence upon exports as a source of growth. If the plan's targets are met, the growth of exports will contribute 37 percent of the growth in manufacturing output. The highest contribution of exports to output growth is in MES sectors, for which it is forecast to be nearly 57 percent. The growth of exports from these sectors is to be responsible for somewhat more than 40 percent of the total absolute growth of manufactured exports between 1975 and 1981. The plan thus calls for concentrating the growth of exports in sectors that potentially are highly intensive in skilled labor. iMPORTS. The plan projects roughly a 50 percent fall in the rate of growth of commodity imports-from 22 percent a year over the 1965-75 period to 12 percent a year. The projections do not distinguish between primary and manufactured imports. In judging the feasibility of reducing the growth rate of imports, it is important to recognize that direct imports constitute nearly one-half of the content of gross exports. Consequently the reduced growth of exports, coupled with a projected decline in their direct import content, will contribute substantially to the decline in import growth. Grain imports are also expected to grow slowly. To- gether, imports for export and grain imports accounted for nearly 40 percent of total imports in 1975. The proportion of domestic demand supplied by imports is expected to fall for several commodities, including crude iron and steel, smelted zinc and copper, fertilizer, and a number of petrochemical products and their synthetic fiber and resin derivatives. Thus backward integration is to be continued over a modestly wide front in the basic metal and chemical industries. In most cases, however, this backward integration is to be the result of an expansion of existing production, including the 260 PART THREE: MAJOR SECTORS production of such complementary outputs as specialty steels, not of the start-up of wholly new production lines. The exception to this pattern is in machinery and electronics, where emphasis is more on initiating new types of production activities. The ratio of imports to domestic pro- duction is expected to rise for a number of key industrial intermediate products. It may thus be concluded that import substitution will not be a major source of overall industrial development, even though selec- tive import substitution in basic metals, chemicals, and MES sectors is to be continued. Sumnmary The fourth plan's strategy for industrial development is much the same in its basic thrust as that pursued under the second and third plans. The industrial structure will continue to deepen through the expansion of basic metal and chemical products. The focus of concerted import substitution will shift to the machinery and electronics sectors. This focus is appropriate. The activities to be expanded in the machinery and electronics sectors are intensive in skilled labor, and the founda- tions for their development have been laid. Furthermore import substi- tution in these sectors is not being pursued at the expense of export expansion. As in the past, export expansion and import substitution will proceed concurrently in the leading industrial sectors. Major Issues of Industrial Policy If the future were known with certainty and if plan targets were fulfilled by mere command, it would be sufficient to focus evaluation of the fourth plan on the consistency and efficiency of the targets. Because they are not, it is far more important to assess the flexibility permitted by the plan under changing conditions and to consider whether intended policies will be sufficient to achieve the plan's general objectives. Except for identified projects in selected key sectors, industrial planning in Korea is indicative. This planning is most usefully appraised by deter- mining the robustness of broad outlines established and policies to be followed in an uncertain world. Investment allocation In comparison with past performance, the plan projects a relatively low growth rate for the manufacturing sector. In addition, the level of MANUFACTURING 261 investment in manufacturing considered necessary to realize the plan's conservative forecast is a rock-bottom estimate. In both respects the approach underlying the fourth plan differs from that underlying the two previous plans. Table 8.10 compares the composition of investment and Table 8.10. Investment and Value Added in Manufacturing, 1967-81 (billions of won) Fourth Second plan Third plan plan (1967-71) (1972-76) (1977-81) Item Planned Actual Planned Actuala Planned Total investment 980.1 1,984.2 4,524.5 3,774.1 19,028.0 Manufacturing investment 283.5 424.9 1,235.3 769.9 5,087.8 Share of manufac- turing (percent) 28.9 21.4 27.3 20.4 26.7 Total value added 335.7 648.1 1,438.9 1,280.9 5,771.0 Manufacturing value added 122.3 282.0 528.5 652.0 3,138.8 Share of manufac- turing (percent) 36.4 43.5 36.7 50.9 54.4 Manufacturing icoR (unlagged) 2.3 1.5 2.3 1.2 1.6 Note: Figures for the second, third, and fourth plans respectively are in 1965, 1970, and 1975 prices. In addition, investment figures for the fourth plan are on an outlay basis; those for the second and third plans are not. Sources: EPB, Second Plan; EPB, Third Plan; EPB, Fourth Plan; EPB, Korea Statistical Yearbook, 1972; aBoK, Economic Statistics Yearbook, 1976. a. Covers only 1972-75. the growth of GNP in the three plans. Under the second and third plans roughlv 28 percent of total investment was to be allocated to the manu- facturing sector, which was expected to contribute about 36 percent of the growth of GNP. In contrast, the fourth plan allocates about 27 per- cent of total investment to the manufacturing sector, which now is expected to contribute 54 percent of the growth of GNP. A dramatic fall in the incremental capital-output ratio (ICOR) pro- jected for the manufacturing sector is responsible for the apparent paradox of a decrease in the sector's planned share of investment and an 262 PART THREE: MAJOR SECTORS increase in its planned contribution to GNP growth.21 The manufacturing ICOR was assumed to be approximately 2.3 in the second and third plan periods; it is assumed to be roughly 1.6 in the fourth plan period. For five-year intervals over the 1955-75 period, the manufacturing ICOR ranged between 1.4 and 1.8. Thus, whereas previous plans underesti- mated the efficiency of investment in manufacturing, the fourth plan takes an extremely optimistic, though not necessarily unrealistic, view. Over the 1965-75 period the manufacturing icoR was 1.3 in 1970 prices. The ICOR of 1.6 assumed for the fourth plan period is in 1975 prices and must be considered to be roughly equivalent to the figure of 1.3 in 1970 prices. The reason for this is the increase that has taken place in the relative prices of capital goods. In addition, the composi- tion of manufacturing investment anticipated for the fourth plan period is not the same as that experienced over the preceding decade. If any- thing, the shift in the allocation of investment in manufacturing, as measured by capital-output ratios, is toward activities that are at least slightly more capital-intensive. The effect this shift will have on the overall manufacturing icoR is difficult to project. It depends upon the precise composition of the projects to be undertaken in MFs sectors, about which there continues to be uncertainty. It was possible to examine projected investment costs for certain identified projects. These projections appear to be lower bounds on probable investment require- ments. In addition, the plan assumes that the ICOR in the light manu- facturing sectors will remain roughly at the historically low value achieved between 1972 and 1975. By overestimating the manufacturing ICOR, the previous plans im- plicitly included a margin that permitted the manufacturing sector to grow more rapidly than planned. This margin enabled the sector to contribute more to the growth of ONP and employment than expected. Under the second plan the actual level of investment, both in aggregate and in manufacturing, was far greater than planned. The result was that the actual share of manufacturing in investment was less than anticipated and its proportionate share of GNP growth was larger. It appears, however, that actual investment in the third plan period was 21. The ICOR is conventionally measured by dividing one year's investment by the change in output in that or the succeeding year. Periods longer than one year are also used. The icoR serves as an indicator of the capital cost of additional out- put and is inversely related to the productivity of capital. MANUFACTURING 263 much closer to that planned. (The actual figures for the third plan in table 8.10 omit the last year of the plan period, 1976.) Similarly it is considered unlikely that planned total investment over the fourth plan period will be exceeded. The successful development of Korea's manufacturing sector has been a result of the vigor with which entrepreneurs in the unplanned sectors seized emerging opportunities for profitable investment. Thus the danger is considerable in simultaneously making a conservative forecast of the manufacturing sector's growth and an optimistic forecast of the efficiency of investment. This danger is particularly apposite for the large-scale and capital-intensive projects in the planned sectors. To the extent that the plan's target for manufacturing investment becomes the basis for credit allocations, foreign-capital licensing, and other con- trols over investment, the latitude for aggressive responses to unforeseen opportunities by entrepreneurs in the manufacturing sector will decline. The dynamism of the economy's leading sector would thus be under- mined. Furthermore attempts to contain manufacturing investment within the target, in the face of possible cost overruns for planned projects, would lead to manufacturing growth that is lower than that projected. Policymakers will therefore need to monitor investments in the planned and unplanned sectors to determine the possible need for an increase in the investment allocated to manufacturing. On the one hand, it may turn out that opportunities for socially profitable investment in unplanned sectors will be more than those foreseen. Flexibility in the control of investment in these sectors is thus required even at the outset of the plan period, and the conservatism of the plan's target for total growth of manufactured exports should be kept prominently in view. On the other hand, investment cost overruns in planned sectors would require an increase in the total amount of investment allocated to man- ufacturing if unplanned sectors continue to grow at the projected rate. By increasing total investment requirements during the first half of the plan period, the bunching of investment in basic metals and chemi- cals will make it more difficult to respond flexibly in the short term to opportunities greater than those foreseen for the expansion of light industries. Even so, if the growth of manufactured exports is greater than that planned, it is likely that increased inflows of foreign capital would finance additional manufacturing investment. Indeed the possi- bility is not remote that a synergistic interaction will develop between the growth of manufactured exports, investment, and foreign capital 264 PART THREE: MAJOR SECTORS inflows. If it does, the growth of manufacturing could be greater than is planned, particularly in the light industries. This possibility bears close monitoring, not least because of its favorable impact on employ- ment. Employment generation The possible need for an increase in the amount of investment allo- cated to light manufacturing is reinforced by the likelihood that the plan's employment objectives will not be met with the investment targets that it has set. In part this is a question of the ICORS assumed. But it also involves the plan's assumptions regarding sectoral employment elas- ticities and capital-labor ratios. The employment issue is discussed in detail in chapter 6; only a few considerations are noted here. The in- vestment cost of employment generation is lowest in the light manu- facturing and MES sectors. It is to these sectors that attention is likely to be directed if policymakers attempt to increase employment gen- eration. Figures given in the fourth plan indicate that sixty-four additional jobs are created by the investment of WI billion in the basic metal and chemical sectors; 393.5 in MES sectors; and 410.6 in light industry.22 Because the last two groups of sectors have nearly identical capital-labor ratios, the employment-investment tradeoff between them is nearly one for one. But it is the tradeoff in growth rates that is most relevant to assessing the comparative difficultv of generating the same amount of employment in the two sectoral groupings. The reason is that the capac- ity to absorb investment is more a function of an industry's growth rate than of the absolute amount of investment directed to it. Total value added in the light industries will be more than four times that in MES sectors at the start of the plan period. Thus a I percentage point fall from the planned growth rate of the light industries requires a nearly 6 percentage point increase from the planned growth rate of MEs sectors in order to maintain the same level of employment. Leaving aside de- mand considerations, it would be easier to increase investment in the light industries than in MES sectors. Furthermore a shortfall of invest- ment in M9IEs sectors could more easily be offset by an increase in invest- ment in the light industries than the converse. 22. The approximate cost per job is $31,888, $5,186, and $4,970 respectively. Investment here is on an outlay basis in 1975 prices. MANUFACTURING 265 Light manufacturing, including textiles and clothing, is expected to contribute 63 percent of the additional jobs to be created in the manu- facturing sectors between 1975 and 1981. A 10 percent increase in the amount of investment allocated to light manufacturing would increase employment generation in manufacturing by roughly 6 percent, man- ufacturing investment by only 3 percent, total investment by less than 1 percent, and total employment in manufacturing in 1981 by nearly 2 percent. Thus increasing investment in the light industries to generate additional employment would neither increase total investment beyond reasonable limits nor, if total investment were held constant, greatly decrease investment elsewhere. Lioht industry is expected to provide 60 percent of the increment of manufactured exports and to be the foundation for continuing high rates of manufacturing growth. The main constraint on the growth of this sector is export demand. Never- theless it is quite possible that export demand may, as it has so often in the past, exceed the expectations of planners. If it does, adjustments would be called for in the plan's allocations. Industrial incentives The amount of investment targeted for the manufacturing sector in previous plans, whether by design or happenstance, enabled entrepre- neurs to expand output, particularly in export activities, without reduc- ing output elsewhere in the economy to a level below that planned. Thus exports of manufactures were able to grow far more rapidly than expected without creating shortages on the domestic market, except for such isolated instances as during the export boom in 1973. In contrast, government intends to hold a far tighter rein on the industrial sector during the fourth plan period. A great deal rests on whether exports of manufactures will be per- mitted to grow faster than planned, if this becomes possible, and on whether this possibility can effectively be signaled to policymakers, when and if it materializes. For the second question, the growth of ex- ports is itself the clearest signal. The only effective constraint on the growth of manufactured exports in the short run is the production capacity of existing plants. Although rates of capacity use already are high, plants can nevertheless achieve higher levels of output. In the longer run, investment in export activities can be increased as long as government imposes no constraints on investment through capacity licensing and other means. Such constraints have not been imposed in the past. There seems to be little likelihood that they would be imposed 266 PART THREE: MAJOR SECTORS in the future. The thrust of the foregoing assessment of the fourth plan is that it would be a fundamental mistake to impose constraints on export-related investments. Such constraints would seriously compromise the plan's employment objectives and further defer the date at which Korea will achieve self-sufficiency, as measured by the current account of the balance of payments. This question may, however, be moot. Korea's export performance in 1976 and 1977 greatly exceeded initial expectations, and this performance may already have led to a revision in policymakers' forecasts for the plan period. The maintenance of adequate incentives to manufactured exports is most important in Korea. As the historical record makes clear, rapid increases in Korean manufactured exports were achieved in the frame- work of an incentive system that did not discriminate against export activities. This system and the reliance of the fourth plan on continued rapid export growth make it worth while to examine recent changes in incentive policy in some detail. As noted earlier, the international com- petitiveness of Korean industry improved substantially between 1970 and 1973 because of several devaluations of the won and the international currency realignments during this period. Subsequent reductions in export incentives were motivated by two considerations. First, some argued that export incentives afforded needlessly large profits on export activity and consequently were too great. Second, policymakers wanted to stimulate the use of domestically produced intermediate and capital goods. Incentives to import substitution were modified to reflect the higher priority placed on expanding the domestic production of basic metal and chemical products as well as mechanical engineering products. In the following discussion, changes in the system of incentives are first reviewed for export production and then for import-substituting production. CHANGES IN EXPORT PROMOTION. Government began to reduce export incentives in 1973. The 50 percent reduction in direct taxes on profits earned in export activity was abolished, but tax credits continued for expenditure on the development of export markets. The preferential interest rate on short-term, export-related credit was increased.23 In late 1976 credit against accounts receivable from exports was limited to 23. The interest rate reform of 1972 may also have reduced the value of credit subsidies. MANUFACTURING 267 85 percent of the export value for up to 135 days and carried an interest rate of 8 percent a year. The nonpreferential, commercial bank lending rate was 15.5 percent. Exporters could also obtain loans for 90 percent of the value of imported inputs at 2.5 percentage points above the Eurodollar rate for up to 120 days and for 93 percent of the value of domestically purchased inputs at an interest rate of 8 percent for up to 135 days. The regulations thus favored the purchase of domestically produced inputs through lower interest rates and a greater proportion and longer period of financing. Government also took steps to reduce the incentives afforded by the terms of access to imports. First, wastage allowances were reduced on most commodities without being fully offset by the rates established for new export products. The simple average of wastage allowance rates fell from 16.6 percent in 1973 to 10.5 percent in 1976. Second, an advance deposit requirement was introduced, under which an amount equal to 15 percent of the value of imported inputs used in export pro- duction had to be paid 90 days in advance. This amount was small in comparison with the 100-110 percent advance deposit required on essential imports and that of 200 percent on luxury imports for domestic use. In July 1975 a switch was made to a tariff rebate or drawback system, under which tariffs on imported inputs for export production are paid at the time of import and rebated at the time of export. Under the previous system, tariff exemptions were granted when imports for export production cleared customs. For more than one-half of exports, draw- back rates are individually determined on the basis of documentary evidence of tariffs and indirect taxes paid. For the remainder, drawback rates are periodically fixed on a product-by-product basis. In the second case, the rebate due on a particular input is a weighted average of the levies applied to domestic purchases and imports; the weights are re- spectively proportional to the industry's domestic and imported purchases of the input. The drawback may thus overcompensate or undercompen- sate for duties and indirect taxes paid on inputs used in export produc- tion by an individual producer, depending upon whether that producer purchases relatively more or less of his input from domestic producers than the industry as a whole. Although exporters can apply for individ- ual drawbacks when the levies actually paid exceed by 20 percent or more the established drawback for the export in question, there is no provision for adjustment for overpayment. This system encourages the use of domestically produced inputs because the link between raw 268 PART THREE: MAJOR SECTORS material imports and tariff exemptions is no longer clear. It also negates the bias resulting from excessive wastage allowances.24 When introduced, the drawback system was provisionallv coupled with the deferrcd payment of customs duties on imports for export pro- duction. Until July 1976 payment was in most cases deferred for four months. At the choice of the exporter, this period is reduced by one month each year, or the amount of deferred payment is reduced by 25 perccnt each year. The drawback system without deferred payment will become fully operative in July 1979. It will then impose its full burden on exporters because thev will subsequently have to finance customs duties from the time of import until rebates are paid. There has also been a changc in the system of tariff exemptions for imported machinery and equipment and for materials used in fabricated metal structures. Since 1974 exporters have not received duty exemp- tions on their imports of capital goods. The incentive effects of this change are somewhat softcned by the fact that payment of customs duties on imported capital goods for use in exporting is stretched out over six equal installments made at six-month intervals. Tariff exemp- tions are now granted to selected importers in particular industries, irrespective of whether they produce for foreign or domestic markets. Exemptions are subject to three conditions: the imported capital goods must be essential to the manufacturing process; they must embody the latest technology; and they must not be domestically produced. As pre- viously, however, complete and automatic exemptions from tariffs con- tinue to be granted for imports of capital goods financed by foreign capital. Stimulated at least in part by government incentives, one important institutional change is rapidly occurring in the export sector. Since 1975 government has provided additional incentives for establishing large trading companies which produce and export on their own account or act as exporting agents for smaller enterprises. Large trading companies benefit from immediate access to short-term export credit for various purposes without L/C backing, but within a ceiling determined on the basis of their past export performance. They also benefit from relaxed controls on stockpiling inventories of imported raw materials and in- 24. If the incentive is effective in increasing the proportional use of domesticallv produced inputs, the drawback rate would fall over time, eventually reaching the value of indirect taxes paid at the time that all inputs are locally purchased. The incentive to use indigenous inputs is thus self-destroving when domestically pro- duced inputs are fully competitive with imports. MANUFACTURING 269 creased allowances of foreign exchange to conduct overseas marketing. In 1976 there were six large trading firms, each with an annual export target of between $100 and $350 million. Applications from other traders were pending. Government encouragement to establishing large trading houses apparently reflects a move to decentralize the administration of at least some export incentives. Export incentives approximately returned to their pre-1971 level in 1974, and they appear to have remained at this level in 1975 and 1976. Nevertheless exporters interviewed by the members of a World Bank mission expressed none of the complaints frequently heard in the past about the low level of export incentives. Government officials also re- marked about the virtual absence of pressure by exporters to have export incentives increased. The conclusion to be drawn is that export incen- tives in 1971 through 1973 may indeed have led to excess profits beyond those necessary to maintain the same level, or an only slightly dimin- ished level, of exports. About the appropriateness of the current level of export incentives, the full effect of most changes in export incentives has not yet been felt because the changes are phased over several years. Furthermore it is not the absolute level of export incentives that matters, but their relative level in comparison with incentives granted to produc- tion for domestic sale. This relative level presumably governs the move- ment of resources between production for the export and domestic markets. CHANGES IN IMPORT PROTECTION. In contrast with production for sale abroad, production for domestic sale has generally been subject to cus- toms duties and indirect taxes on inputs, higher interest charges, higher income tax rates, and higher prices for electricity and railroad transport. As noted earlier, however, key import-substituting industries have been granted many of the preferences given to exporters. Production for domestic sale also benefits from tariffs and restrictions on competing imports. More important than the recent modest decline in actual tariff rates has been the apparent trend toward increased reliance on import con- trols as a means of affording protection to domestic producers, partic- ularly for capital goods. Under the prevailing legislation for import con- trol, lists of prohibited and restricted items for import are published semiannually. Any item not appearing on these lists may be imported without restriction. When applied, import restrictions operate in one of two ways. For items on which there is an import quota, usually inter- mediate goods, the associations of producers requiring the item as an 270 PART THREE: MAJOR SECTORS intermediate input allocate the quota among individual producers. For items for which import licenses are handled on a case-by-case basis, the approval of the association of the producers of the domestic substitute is generally required to obtain the license. The Ministry of Commerce and Industry (MCI) acts as final arbiter in disputes among parties and has final authority over import licensing. Data on tariff rates, import prohibitions, and import restrictions do not show a consistent pattern. Although average tariffs and the number of prohibited imports have decreased since 1972, import restrictions now apply to a greater number of products than before. Thus the ques- tion is how lower tariffs and increased import restrictions translate into nominal protection rates. An investigator at the Korea Development Institute (KDI) used whole- sale price indexes to update the price information of the study of effec- tive rates of protection and subsidy in 1968. He concluded that the nominal protection of Korean production for the domestic market was negative in 1973. Although this result may reflect the special situation existing in 1973, it is confirmed by the results of a partial and informal survey conducted by EPB in 1976. This survey compared prices in the domestic markets of Japan, the United States, and the Federal Republic of Germany with those in Korea. For the fifty-one commodities included in the survey, Korean prices were on average 16 percent below those in Japan, 13 percent below those in the United States, and 1 percent below those in the Federal Republic of Germany. If the survey is assumed to be reasonably reliable, the low and in most cases negative nominal protection that it reveals is striking. None of the commodities covered can be described as traditional or uniquely Korean; all are outputs of a modern industrial sector. Nevertheless quality differences probably distort the comparison in some product areas, especially machinery. Despite the limitations of a price survey of restricted scope and un- certain reliability, the results appear to show, at the least, that average domestic prices are not higher in Korea than abroad. This pattern sug- gests that nominal protection rates for the domestic market have fallen since 1968. Considering the apparent trends in other determinants of the effective incentive rates to production for the domestic market, it appears that the aggregate effective protection and subsidy rates on domestic sales have remained close to their 1968 levels or fallen some- what. This performance does not mean that the spread between effective incentive rates on exports and on domestic sales, which strongly favored exports in 1968, has necessarily widened even further. Indeed, recalling that the real effective exchange rate on exports in 1976 was approxi- MANUFACTURING 27I mately the same as that in the late 1960s, it would have to be concluded that the spread most likely has narrowed somewhat, if not remained constant.25 Thus the evidence indicates that the current level of export incentives is sufficient to sustain rapid export growth to the extent that the same relative level of export incentives in the past contributed to Korea's remarkable export perfonnance. But this conclusion relates to the level of overall incentives; it appears not to hold true for some important import sectors. The flexible tariff system has been used to increase the protection granted to some producers in the heavy industrial sector, mostly pro- ducers of capital goods. In addition, there appears to have been a ten- dency, at least until 1975, to impose controls on competing imports of intermediate and capital goods immediately upon the initiation of their domestic production. Although some capital-goods imports were licensed in the 1960s and early 1970s, the licensing practiced during this period was pro forma and unrestrictive. The introduction of binding controls on capital-goods imports is thus a recent phenomenon. It is accomplished either by setting effective quotas on capital goods already on the re- stricted import list or by adding new items to the list. On the basis of interviews conducted by the World Bank staff on three visits to Korea during 1975 and 1976, it appears that competing imports of heavy industrial products have been effectively prohibited in numerous cases, even when the prospective importer is engaged in export activity. The products include basic chemicals and metals as well as mechanical engineering products. For the last group, the machinerv producers' association determines eligibility for an import license, subject to arbitration by MCI, for items on the restricted list. The criterion used is that equipment with domestically produced similars cannot be im- ported unless the terms on which it can be purchased, including price, quality, and delivery date, are not competitive with those for the im- ported item. Furthermore competing imports of machinery, when per- mitted, are now subject to the full legal tariff rate unless a reduction or exemption is permitted under other criteria. In the 1960s the economic cost of the system of import controls may not have been large in Korea because controlled items were mostly nondurable consumer goods and their inputs. Although the variety of consumer goods domestically available was limited by import control, domestic competition coupled with export competition in foreign mar- 25. All indications are that the spread was much greater from 1971 to 1973. 272 PART THREE: MAJOR SECTORS kets assured efficient operations in most of these products. The next stage of development began in the late 1960s, when industries producing petrochemicals, steel, and other heavy industrial intermediate products were established. Competition could not be relied upon to eliminate monopoly profits by reason of large-scale economies. Thus import control was made conditional on the maintenance of prices at levels considered to be acceptable. Because the products affected were largely standardized, price comparisons could be made, and the control of prices enforced, without much difficulty.26 Today import licensing and control are more complex. Simply stated, policies to encourage backward linkages from exports can, if not care- fully orchestrated, compromise the international competitiveness of Korea's exports. As Korea advances to the production of higher level intermediate products and machinery, the difficulties of administering controls on similar competing imports, without concurrently sacrificing too much economic efficiency, become vastly greater. The substantial differentiation, specialization, and quality differences found in such products make it difficult simultaneously to control imports and main- tain prices in a desired relation to world mnarket levels. In addition, the denial of import licenses to producers may lead to wrong choices, which would increase production costs or reduce quality at the next level of fabrication. Adverse effects are particularly likely to arise from the con- trol of capital-goods imports because differences in equipment specifica- tions can often be appraised only by the user. The domestic market for machinerv rarely is sufficient for production on an efficient scale. If operations are to be efficient, firms will have to relv on exports for part of their machinery output, which will be thwarted if high-cost plants are established under the cover of import controls. Furthermore subcontracting among small- and medium-sized firms is vital, particu- larly to the establishment of an efficient mechanical engineering sector. Yet small- and medium-sized firms, as well as firms located away from Seoul, are discriminated against in the administration of import con- trols, a fact established in the course of an McI survev of small- and medium-sized firms. Special promotional measures are needed to publicize the availability and technical characteristics of higher level intermediate products and to induce their use as they begin to be domestically produced. Recently, 26. This paragraph and the next draw from Bela Balassa, Policy Reform in Developing Countries (London: Pergamon Press, 1977), pp. 139-64. MANUFACTURING 273 moreover, various government agencies administering industrial incen- tives, and even the machinery producers' association, have recognized the potentially adverse effects of using import controls to achieve effi- cient import substitution. As a result, MCI began in about 1975 to reject applications by machinery producers to add new items to the restricted list of imports. Nevertheless items already on the restricted list have apparently continued to be subject to import controls. This practice indicates the absence of consensus on whether and how to use import controls to stimulate further import substitution in the heavy industrial sector. Widespread recognition of their adverse consequences for efficient production may lead to a trend away from their use. The govemment policy of easy access to imported capital goods appears certain to have retarded the development of the domestic capital- goods industry in the 1960s and first years of the 1970s. Particularly for exporters, the slight overvaluation of the won, tariff exemptions on imported capital goods, and the financing of purchases by supplier credits that carried low interest rates relative to those on the domestic market-all acted to increase the attractiveness of imported capital goods.27 The same factors affected choices in the key import-substituting industries. Domestic capital-goods producers received virtually no special incentives. Medium- and long-term credit for financing the purchase of indigenous capital goods became available under a special fund only after 1967, and then in relatively limited amounts.28 As mentioned earlier, various measures have been taken to redress the balance: ex- porters are no longer exempt from tariffs on capital-goods imports, and at least some capital-goods imports are restricted. The implications of these measures, especially those of import restrictions, for the efficiency of manufacturing production give rise to considerable concern. Govern- ment's policy in using these instruments has not yet coalesced. INVESTMENT INCENTIVES. Government has introduced other measures wvhich have the potential to distort efficiency to less degree. A number of specialized medium- and long-term credit funds have been established to provide won-financing for construction and for purchases of domesti- cally produced machinery. The National Investment Fund (NIF), which provides subsidized medium-term credit to various industries, is the 27. Yung W. Rhee and Larry E. Westphal, "A Micro Econometric Investiga- tion of the Impact of Industrial Policy on Technology Choice," Journal of Develop- ment Economics, vol. 4 (1977), pp. 205-38. 28. Frank, Kim, and Westphal, Foreign Trade Regimes, pp. 117-19. 274 PART THREE: MAJOR SECTORS largest. Over the fourth plan period, it is expected that NIF will finance nearly 18 percent of investment in the machinery sector and roughly 13 percent of total fixed-capital formation in manufacturing. The NIF interest rate is 12 percent, equal to that on subsidized medium- term won-lending to the export industries, and 3.5 percentage points below the unsubsidized commercial short-term rate in 1976. The maxi- mum period of an NIF loan is eight years. A committee including repre- sentatives from EPB, AICI, MOF, BOK, and KDB determines the guidelines under which this fund is allocated among specific industries; KDB ad- ministers most lending out of the fund. Other special funds, including those for small- and medium-sized enterprises and for exporters, are administered in a similar fashion, but funded on an annual basis by special appropriations. Several factors must be recognized in appraising the effects of these funds on resource allocation. Their size relative to total investment is limited, and they embody only a relatively small interest subsidy. Through foreign capital inflows and limited government financing, credit has been available at relatively low interest rates in the past to finance selected import-substituting investments. But low interest credit has heretofore been largely linked to the purchase of imported capital goods. The establishment of these funds means that subsidized credit on comparable terms is now available to purchase domestically produced capital goods. Thus the primary effect on resource allocation is the incentive provided to the domestic capital-goods industries by removing the bias that interest rate differences caused in favor of purchasing imported capital goods. The establishment of these funds is not a break with past practice of allocating subsidized credit to favored infant industries. But the mag- nitude of a possible secondary effect of resource allocation on the dis- tribution of investment among industries cannot easily be determined. The reason is that it would require an extensive and detailed analysis of the demand for investment funds for various uses and of the rationing of credit under various terms. It has not been possible to undertake this kind of analysis. In view of the small interest rate difference and the administered access to finance from these funds, the probable effect of the interest rate difference is simply to yield windfall gains to the favored producers, not to influence the direction of investment. It was noted earlier that tariff exemptions on capital-goods imports are granted to some industries, subject to the condition that the items are not produced domestically. The industries benefiting from duty-free treatment include: agriculture and livestock; fishing; mining; chemicals; MANUFACTURING 275 basic metals; electrical machinery; nonelectrical machinery; transport equipment; scientific apparatus; defense industries; machinery parts pro- duction; electronics and parts; electricity generation; and airline and railroad transport. In addition, domestic investors in the heavy indus- tries and infrastructure-that is, all but the first three industries just listed-are granted further investment incentives. Investors have a choice among three options: total exemption from the corporate income tax for three years and a 50 percent reduction in this tax for an additional two years; a reduction in income taxes equal to 8 percent of the total value of the investment; a doubling of normal depreciation allowances. There also is a tax reduction of 8 percent on the proportion of undis- tributed profits used for investment expenditure in new technology, or up to 10 percent if domestic machinery is used, as well as a tax exemp- tion on a reserve fund of up to 10 percent of profits. Investment incen- tives are somewhat greater for foreign direct investments in the heavy industries. In addition, producers in selected branches of the heavy in- dustrial sector will benefit from the proposed establishment of special technological research institutes concerned with foundry products, ma- chine tools, electronics, and shipbuilding. SUMMARY. So long as market prices are reasonably close to shadow prices and investment is not artificially constrained, entrepreneurial decisions will indicate where the opportunities for profitable investment lie and in what volume. The only possible exceptions are in large-scale industrial projects and in sectors, such as MES sectors, where leaming- by-doing plays an important part in achieving high levels of productivity. The stringency of the fourth plan makes two things important: domes- tic prices must continue to be maintained close to world market prices; policymakers must continue to be sensitive to unfolding developments in the industrial sector and to respond flexibly, lest opportunities for socially beneficial investment be missed. These objectives require that incentives to exports be maintained at least on a par with incentives to sales on the domestic market. If having proper policies for industrial incentives was important in the past, it is more important now. Although the evidence is partial and inconclusive, it appears that export incentives are on a par with, or only slightly exceed, incentives to domestic sales in those industries where exporting already is well established. Because exports from these industries are continuing to grow rapidly and exporters in these industries apparently feel that export incentives are adequate, this undoubtedly is the correct policy for these exports. 276 PART THREE: MAJOR SECTORS There is some question, however, whether maintaining equal effec- tive incentives to exports and domestic sales is the appropriate policy, particularly for the heavy industrial sectors in which exporting is not well established. Initially the costs of marketing in international markets are higher than those of marketing in the domestic market, if only because information about the domestic market is more readily available than that about overseas markets. Somewhat higher incentives to exports than domestic sales may thus be appropriate, at least temporarily. In addition, the relative incentive to producers in the basic intermediate and MES sectors has recently been increased to promote their development. In- centives are being given through protection in the domestic market and through subsidies to production. Unless protection is offset through special export incentives, its use to promote these industries will retard their export growth, except insofar as higher prices on the domestic market may be used to cover fixed costs not fully charged on export sales. Because the outputs of these industries are producer goods, their protection may detrimentally affect exports from other industries and retard further backward integration from current exports. These con- siderations define an important set of policy issues now being debated within the Korean government. It is not clear what set of policies will ultimatelv be chosen. Whatever course is chosen, policies to promote the heavy industrial sectors should be designed to provide adequate incentives to exports relative to domestic sales. But they should not raise the cost of produc- tion in well-established export activity. FurtheTmore the level of effec- tive incentives to the heavy industrial sectors, particularly the more capital-intensive activities within them, will have to be balanced against the need to provide adequate incentives to continue expanding exports of light manufacturing products and maintain relatively full employ- ment. Changed character of industrial deepening As a final major point, a subtle change in the character of industrial development should be emphasized because of its implications for the type of direct government interventions needed to promote efficient, rapid growth. Past government activities to promote industrial develop- ment have been highly differentiated. The relationship between gov- ernment and private enterprise has been quite different in different sectors. In the planned sectors government identifies specific projects and then promotes them to ensurc their implementation, sometimes even to MANUIFACTURING 277 the point of initiating them through the agency of a public enterprise. Explicit project identification plays a lesser role in the unplanned sec- tors, where incentive policies are used to direct resource allocation in desired directions. The line between planned and unplanned is not sharp. Steel is in- cluded among the planned sectors, but government has been actively involved in project identification and implementation only for the in- tegrated iron and steel mill, its expansion, and the special steel plant. Nevertheless numerous small private steel mills play an important part in the sector and constitute a more or less unplanned fringe. On the other hand, government involvement in such major unplanned sectors as textiles includes active participation, more by MCI than EPB, in estab- lishing production and capacity-expansion programs at the enterprise level. The major distinction is thus in the balance of power between decentralized, most often private, interests and the central authorities (EPB and MCI) in reaching and implementing economic decisions. Major sectors for import substitution have nearly always been in- cluded among the planned sectors and have therefore received dispro- portionate attention in past five-year planning exercises. This attention focused on large-scale projects to produce undifferentiated, standardized products in the basic metal and chemical industries. The focus in the fourth plan has shifted to machinery and electronics. These sectors produce highly differentiated, nonstandardized products and have a far larger range of distinct production processes. Moreover these sectors are to be simultaneously called upon to increase their exports at a quite rapid pace. This too is a new element in the character of industrial development because leading import-substituting sectors in previous plans were not expected to play a major part in export expansion. Furthermore, in such sectors as steel, investment will increasingly be directed toward the production of specialized products required by sec- tors using increasingly differentiated and sophisticated technology. The growing complexity of the industrial sector requires new modes of government promotion. Building an integrated steel mill or petro- chemical complex is far less intricate an operation than establishing an efficient mechanical engineering industry. The major challenge of the fourth plan is to achieve efficiently integrated development across a broad set of highly fnterdependent activities. Technology transfer, quality improvement, and internal and external market development will be most important in deternining progress toward achieving the structural changes contemplated. The most significant problems are likely to be encountered in industrial machinery and electronics because 278 PART THREE: MAJOR SECTORS a complete set of assistance measures has not been developed. For example, the plan generally stresses expenditure on research and devel- opment, but no major steps have been taken in the electronics industry to secure backward integration from the assembly stage. As described earlier, a package of financial incentives, including tax and credit preferences, is available to investors in heavy industrial proj- ects. In addition, protection from imports through tariffs and trade con- trols has been granted to some producers in MES sectors. But these incentives have been in force for some time for investors in the Chang- won Machinery Complex, where the number of plants in operation or under construction lags well behind government's original expectations. Furthermore the differences are large between the projects tentatively identified for the complex and those actually locating there. These differences suggest that much of government's project promotion in the machinery sector has been rather far removed from the reality perceived by individual entrepreneurs. Nevertheless the evidence is equally con- sistent with the view that entrepreneurs are waiting for government to adopt a clear strategy and set of supportive promotional measures. Of the promotional measures adopted, two appear particularly prom- ising. The first is for medium-term financing available through NIF for purchases of domestically produced capital goods. Some of this financing is to be used for export credits to support the overseas sale of ships and machinery; some is to be used to finance sales of capital goods to local purchasers. The incentive effects of the preferential terms are less im- portant than the increased availability of finance for the purchase of indigenous capital goods. In the past the absence of domestic medium- term credit has been a major disincentive to the development of the machinery sector. It may be desirable to increase the allocation of do- mestic medium-term credit to machinery purchasers in the manufactur- ing sector because the amount of NIF financing allocated to this sector is only about 13 percent of total fixed-capital formation anticipated in manufacturing. The second notable measure is the designation by MCI of fifty-two small- and medium-scale producers manufacturing a variety of machinery components and parts. Each of them is to receive fiscal incentives, assistance to establish technical cooperation with foreign firms, and a loan of $1 million, repayable over five years at an interest rate of 13 percent following a three-year grace period. These firms, which are to be increased in number to 500 by 1981, were selected from more than 1,000 producers surveyed. This step marks an important departure from the past tendency to concentrate on establishing new enterprises, mainly MANUIACTURTNG 279 in the Changwon Complex. It may begin a trend toward the increased involvement of existing producers in achieving greater specialization and efficiency of production. In the long run of, say, ten years, a fully in- tegrated machinery complex will play a dominant role in developing exports and meeting home demand. But in the medium-run of five years or so, the already established small- and medium-sized firms, most of them located away from Changwon in the Seoul, Busan, and Daegu metropolitan areas, must be relied upon to meet production and export targets. It may further be expected that the development of such firms will lay the foundation for establishing the vital network of suppliers of machinery parts and components. Technology transfer and export strategy are the areas in which gov- ernment promotional measures are least well articulated. There is not yet a consensus in government about measures to promote new product lines and production processes involving more sophisticated technology. Some policymakers hold the view that domestic machinery design capa- bilities could be developed with minimal reliance on foreign technical assistance; others argue that Korea must rely upon foreign design capa- bilities. The truth lies between these extremes. Korean capability is sufficient in some areas and deficient in others. A clear delineation of the criteria for government's support of technological development, along with measures to be adopted under different circumstances, has not yet been established. Nevertheless a start has been made. For example, several special technological research institutes are being established for foundry products, machine tools, electronics, and shipbuilding. Korea's MES exports fall into two categories: labor-intensive, techno- logically unsophisticated products, such as black-and-white television sets, which Korean entrepreneurs can export without the day-to-day involve- ment of foreign collaborators in production and marketing management; subcontracting for the labor-intensive, relativelv unsophisticated pro- cesses involved in the production of high technology products which require foreign collaborators for much of the production and marketing management, as well as for the teclnology. The clearest exception to this pattern is the shipbuilding industry. If Korea is to expand its exports of machinery and electronics at the rate planned, and at the same time increase the domestic value added in these exports, it must develop a widespread capability to manage the production and marketing of technologically sophisticated products. This obviously will take time. The fourth plan period is best conceived as laying the foundation for rapid export expansion along these lines in the fifth plan and subsequent periods. What remains to be developed is a 280 PART THREE: MAJOR SECTORS clear vision of an export strategy that will carry Korea's industrial development beyond the 1970s. Reliance upon exports intensive in labor, even skilled labor, without a supporting domestic technological infra- structure will not be sufficient to maintain the momentum that has so far been achieved. This conclusion obviously unleashes a difficult set of issues for policymakers and industrialists to confront, but it is a critical challenge that must be faced during the fourth plan period. Chapter 9 Philippe Bourcier Energy ANTHRACITE is Korea's main domestic energy resource. Total reserves are estimated by the Korea Mining Promotion Corporation (KMPC) at about 1,500 million tons. Of these reserves more than 600 million tons, the equivalent of thirty years of production at 1975 rates, are recover- able. Additional reserves of 60 million tons were discovered in 1975, and government is stepping up the exploration program.' Anthracite deposits are located in mountainous areas which require drift-type and shaft-type underground mines and labor-intensive techniques. Produc- tivity in Korean mines consequently is low-about 1.1 tons a man-shift, compared with a maximum of 3 tons a man-shift in the United States. The quality of coal produced is also low-about 3,500 to 5,000 kilo- calories a kilogram-and is not suitable for coking. Most of it is briquetted and used in the domestic sector. Korea's total hydroelectric potential is estimated at about 2,000 mega- watts, of which about 620 megawatts have already been developed. A further 410 megawatts are under construction or planned for completion before 1981. Hydroelectric resources are concentrated on four main river systems: Han, Naktong, Kum, and Sumjin. The sites generally are of small size and low head and require the construction of costly dams to regulate the river flows. Most of the existing and planned dams are multipurpose projects. Hydropower contributes about 9 percent of total gross power and about 12.5 percent of peak output. It is not expected that hydropower will contribute substantially in the future to the total development of the power subsector. Korea uses relatively large quantities of noncommercial energy-that 1. In this chapter, as elsewhere in this book, all tons are metric tons equal to 1,000 kilograms. 281 282 PART THREE: MAJOR SECTORS is, firewood and animal and agricultural waste. Total consumption in 1974 was estimated at about 3 million tons of oil equivalent (TOE). As would be expected, most production and consumption of noncommercial energy takes place in rural areas. With the exception of charcoal, there is no transport of noncommercial energy. Because Korean forest re- sources are limited, the production of noncommercial energy from fire- wood has been strongly discouraged by government. Nevertheless gov- ernment is considering the creation of tree farms for the purpose of charcoal production. Other potential resources include hydrocarbons, tidal power, and uranium. Limited drilling for hydrocarbons has been under way since the early 1970s, but no productive wells have been discovered. Korea's tidal potential is estimated at about 1,600 megawatts, but because of the technical and financial problems associated with the exploitation of tidal power, it is unlikely that this potential will be developed in the near future. Although uranium exploration has been going on since 1970, no substantial quantities of radioactive fuel have been discovered. With its energy resources limited to coal and hydropower, and with the likelihood that potential oil resources offshore will not be discovered and developed before 1985, Korea will be dependent on foreign imports of petroleum and nuclear fuel to meet its internal requirements in the short and medium terms. For the longer term government has already launched a comprehensive research and development program as part of its policy to promote the maximum development of domestic re- sources. This program is based on the adoption of techniques developed abroad and their modification to suit local conditions. It will examine the technical and economic feasibility of nonconventional sources, such as wind, solar, and geothermal sources. Growth and Pattern of Energy Consumption During 1966-74 total internal energy consumption in Korea grew at an average rate of 7.2 percent a year from about 10.4 million TOE to about 18.1 million TOE. Per capita consumption over the same period grew at an average rate of 5.2 percent a year from 0.36 TOE to 0.54 TOE. Since 1966 the share of noncommercial energy in total demand has been declining considerably, from 3.9 million TOE to 2.9 million ToE. In 1974 it accounted for only 16.2 percent of total internal final con- ENERGY 283 Table 9.1. Consumption of Commercial Energy, by Sector, 1966-74 (thousands of tons of oil equivalent) Residential- Item Industry Transport Others commercial Total 1966 Coal 417 172 357 3,640 4,586 Petroleum 640 583 340 75 1,638 Electricity 214 .. 3 43 260 Total 1,271 755 700 3,758 6,484 1974 Coal 251 ... 124 5,874 6,249 Petroleum 4,146 2,051 594 935 7,726 Electricity 1,015 .. 5 173 1,193 Total 5,412 2,051 723 6,982 15,168 Change in consump- tion, 1966-74 Coal -166 -172 -233 2,234 1,663 Petroleum 3,506 1,468 254 860 6,088 Electricity 802 ... 2 129 933 Total 4,142 1,296 23 3,233 8,684 Share in total increase in con- sumption (percent) 47.7 14.9 0.2 37.2 100.0 ... Zero or negligible. Source: World Bank estimates based on data supplied by the Ministry of Com- merce and Industry (MCI). sumption, compared with 37.5 percent in 1966. Final consumption of commercial energy grew at 11.2 percent a year.2 The share of the principal commercial fuels, by source of final con- sumption of energy, changed considerably during 1966-74 (table 9.1). 2. All figures cited in this chapter have been taken or adapted from data presented in the background papers used in preparing the preliminary draft of Korea's fourth plan. In general, figures for coal were supplied by the Korean Mining Promotion Corporation (KMPc), those for power by Korea Electric Company (xEco), and those for petroleum by the petroleum division of the Ministry of Commercy and Industry (Mci) and by the Economic Planning Board (EPa). 284 PART THREE: MAJOR SECTORS With the exception of coal, the consumption of fuels rose over the period. The increase in the amount of electricity and petroleum directly used for energy was more than fourfold. This increase was mainly the result of the rapid growth of demand in the industrial and transport sectors and of the substitution of petroleum products for coal in the residential-commercial sector. The share of coal in total demand fell from about 71 percent in 1966 to 41 percent in 1974. The structure of demand bv sector also changed markedly over the 1966-74 period. The share of the industrial sector increased from 19.5 percent in 1966 to 35.7 percent in 1974; that of the transport sector from 11.5 percent to 13.5 percent. The share of the residential-commercial sector declined from 58 percent to 46 percent. These changes were the result of the strong policy emphasis on industrial development and the restraint on private consumption. Over the 1966-74 period total energy grew at a somewhat lower rate than GNP, 7.2 percent compared with 10.5 percent; commercial energy consumption grew at a higher rate than GNP, 11.2 percent compared with 10.5 percent (table 9.2). Over the same period the energy inten- Table 9.2. Energy Consuviption, Energy Intensity, and Gross National Product, 1966-74 Average annual growth rate Item 1966 1974 (percent) Gross national product (billions of won in 1970 prices) 1,270 3,810 10.5 Total energy consumption' (thousands of tons of oil equivalent) 10,372 18,102 7.2 Commercial energy consumption (thousands of TOE) 6,484 15,168 11.4 Total energy intensity (TOE per million won of GNP) 6.23 4.75 - Commercial energy intensity (TOE per million won of GNP) 3.77 3.98 0.7 (ToE per thousand dollars of GNP) 1.83 2.00 1.1 - Not applicable. Sources: Table 9.1; World Bank, "Current Economic Position and Prospects of the Republic of Korea," appraisal report no. 768-KO (a restricted circulation docu- ment) (Washington, D.C., 1975; processed). a. Includes noncommercial energy. ENERGY 285 sity of the economy, measured by the consumption of commercial energy per unit of GNP, rose from 3.77 TOE per million won to 3.98 TOE per million won. These figures indicate that, although less energy is re- quired to produce one url.t of GNP, Korea's economr.is becoming more energy intensive. This trend is likely to continue in the future as changes occur in the structure of demand by fuel and sector. Unless conserva- tion measures are applied, it is likely that substitution of commercial energy for noncommercial energy in the residential sector and an increase in the energy demand in the transport sector will cause the consump- tion of energy per unit of GNP to rise above the 1974 levels. Primary energy requirements During 1966-74 primary energy requirements grew at an average rate of 8 percent a year from 11.2 million TOE to about 20.3 million TOE. If noncommercial energy is excluded, the annual rate of growth of com- mercial primary energy would be about 11.5 percent a year. This gives an income elasticity of about 1.1, which is comparable with elasticities observed in Japan, the Republic of China, France, and the Federal Re- public of Germany. The share of coal, which accounted for 71 percent of total demand in 1966, declined to about 38.6 percent in 1974; the share of petroleum increased from 24.7 percent to 59.6 percent. Most of the coal was used in the residential sector; its relative contribution to industry and power generation declined considerably. In 1966 coal accounted for 78 percent of thermal power generation and 39.5 percent of industrial requirements; in 1974 its respective shares were only 5.2 percent and 5.7 percent. This decline was the result of four major factors: the increase in demand for coal in the residential sector; the relative stagnation of domestic coal production; the comparative techni- cal and economic advantages of petroleum for power generation and in- dustrial use; and the availability of low-cost petroleum supplies until the end of 1973. Primary energy requirements were met by domestic production of hydropower and coal and by imports of petroleum. As time passed, how- ever, the ratio of domestic production to total demand declined-from 75 percent in 1966 to 40 percent in 1974. If noncommercial energy is included, the ratios would be 84 percent in 1966 and 48 percent in 1974. Imports of fuels accounted for 16.2 percent of total imports in 1974, compared with 7 percent in 1966. Korea followed the same pat- tern as most nonoil-producing countries by replacing domestic solid fuels with imported oil because of its convenience and relatively low cost. 286 PART THREE: MAJOR SECTORS Development of domestic resources Because Korea is largely dependent on imports, it was hit hard by the increase in the world prices of fuels in 1973 and 1974. Government promptly launched an interim energy program, calling for increased de- velopment of domestic resources and nationwide energy conservation, and began preparing longer-term plans aimed at rationalizing and mini- mizing the cost of energy supplies. The main objective of the program was to increase production of domestic coal. For this purpose government increased the price of coal by 51 percent in 1974 and 25.5 percent in 1975. As a result, coal production increased from about 14 million tons in 1973 to more than 17 million tons in 1975. Current coal production is adequate to meet the demand of the tradi- tional and low-cost residential housing sector. In the medium-term it is not expected that enough coal can be developed to allow its economi- cal substitution for petroleum in the industrial and power generation sectors. It therefore is likely that the main result of the increased avail- ability of coal will be to slow down the substitution of petroleum prod- ucts for coal in the residential and commercial sectors. This is desirable for two main reasons. First, the current supply-demand mix for petro- leum products is well balanced. An increase in the demand for middle distillates, such as kerosene and light fuel oil, could create imbalances which would require additional refinery investments or imports of high- priced refined petroleum products. Second, there are no short-term practical alternatives to the use of coal in the residential sector. It therefore is advisable that coal production be allocated to space heating and other residential uses. Development of the Fuel and Power Industry The performance of the energy sector and of the various subsectors over the second half of the 1960s and early 1970s generally was satis- factory. Rapid growth in energy supply was achieved without obvious misallocation of resources, and the quality of service was good. But some institutional and financial problems emerged partly as a result of the rapid expansion of the sector and partly as a result of government policies for pricing and subsector development. In the early 1970s it appeared that future expansion of the energy sector would be based primarily on imported oil and nuclear power. Domestic coal was to be progressively phased out. The increase in the prices of fuels, particularly ENERGY z87 of petroleum in 1973 and 1974, drastically altered this pattern and compelled government to reconsider its energy policy. Government's response to the oil crisis was impressive: prices of petro- leum products were increased in line with international prices; a com- prehensive energy conservation program was launched; policies were laid down to promote the development of domestic energy sources. These actions brought about some significant results. For example, the increase in final energy consumption in 1974 was limited to about 7 percent, compared with an average growth of 11.2 percent over the 1966-74 period. This initial energy program was subsequently consoli- dated in a more comprehensive energy development program outlined in the fourth plan. Institutional framework The energy sector comprises a mix of publicly owned, semipublicly owned, and private companies which are under the supervision of the Ministry of Commerce and Industry (McI). Although all companies are theoretically autonomous, each with its own board of directors, the in- fluence of government in their development and operations is strong, particularly in pricing, financing, and determining long-term sector ob- jectives. Energy prices are controlled by government at the producer and consumer levels. In the petroleum subsector, which is owned in joint venture with foreign oil companies, the government's pricing policy has generally been in line with international prices. But in the coal and power subsectors, pricing policy appears to have been directed more to providing relatively inexpensive energy than to resolving prob- lems of the adequacy of resource allocations and the financial viability of power and coal companies. Partly because of the inadequate level of prices and partly because of difficulties in mobilizing domestic and external resources for financing energy projects, government has been heavily involved in providing financial resources for the development of coal mines and power facili- ties. It has done this through a complex system of subsidies and budget allocations. Although the policies followed in the coal subsector appear generally satisfactory, the power subsector has suffered from unilateral decisions by government to construct new facilities, from insufficient coordination between MCI and the Economic Planning Board (EPB), and from inadequate resource allocations, particularly in the coal sector, where low prices appear to have prevented increases in production. 288 PART THREE: MAJOR SECTORS Coal Most domestic coal is produced in small anthracite mines having an average capacity of about 70,000 metric tons a year. It is used in the commercial-residential sector where it cannot be easily substituted for alternative fuels because of the nature of heating systems in traditional housing. Govemment is conscious of this problem and considers coal to be as essential a commodity as rice. It has taken measures to ensure that coal prices remain at an acceptable level for low-income groups and that sufficient resources are available to finance the modernization of existing mines and the development of new ones. Until 1973 coal prices on average were less than production costs. This led to the stagnation of coal production and consequently of coal demand. Price increases in 1974 and 1975 have shown that production is sensitive to a change in price, but that consumption is not. Produc- tion and consumption increased from 12 million tons in 1973 to about 17 million tons in 1975. Coal production costs have increased rapidlv, particularly because Korean mines are not mechanized. In the future, given the unavoidable increases in labor costs, govemment will have to review its coal-mining development policy to keep cost increases under control and ensure that the demand of the residential-commercial sector is adequately met. To help coal mines modernize and to promote the development of domestic production, government provides grants of up to 70 percent of the cost of getting a mine into production and concessionary loans of up to 15 percent of this cost. The remaining 15 percent has to be pro- vided by the mine ownership. This subsidy is administered by KMPC, which draws its funds from government and particularly from the pro- ceeds of a special tax levied on Bunker C. This policy, instrumental in the increase in production in 1974 and 1975, is a good example of government intervention that can effectively serve multiple purposes. By increasing the cost of Bunker C by 5 percent, it discourages con- sumption of refined oil products. It also provides resources for the de- velopment of domestic resources which can be substituted for imported oil in the residential sector. Because coal is an essential commodity in the residential sector, gov- ernment has restricted the nonresidential use of domestic coal. Although this policv appears justified in view of the importance of the residential sector and the social implications of coal shortages, it has created con- siderable problems in the operation of coal-fired power plants. These ENERGY 289 plants were designed to use a mixture of coal of a given quality and of heavy fuel oil in the ratio of 90 to 10. Now using coal of lower grade or coal-oil mixtures not in accord with design specifications, they are faced with eficiency and maintenance problems. If this restrictive policy is continued, which seems likely, steps should be taken to modify boilers for more use of liquid fuels. Power Three companies operate in the power sector. The Korea Electric Company (KEco)-jointly owned by government and private interests -is the sole distributor of electric power in the country. KECO currently owns and operates the bulk of the countrv's total generating capacity and, except for multipurpose projects, is responsible for the planning and construction of all power projects. The Kyongin Energy Company, a private generating company having an installed capacity of 325 mega- watts, sells its entire output to xEco. The Industrial Sites and Water Resources Development Corporation (IswAco), which owns and oper- ates the 200-megawatt Soyanggang multipurpose project, also sells its power output to KECO. ISWACO is constructing the 90-megawatt Andong multipurpose project. Electrification is fairly well advanced in Korea. About 95 percent of all households had access to electricity supply by the end of 1976- about 99 percent were in urban areas and 91 percent in rural areas. The average household used 580 kilowatt-hours in 1973 and 564 kilowatt- hours in 1974. Total consumption during 1974, excluding consumption by self-producers, was about 14,048 gigawatt-hours. Per capita consump- tion was 420 kilowatt-hours, which is about 25 percent of the world average, 10 percent of the 1973 per capita consumption in Japan, and 70 percent of that in Brazil.3 More than 80 percent of consumption was used by industry. The average growth of electricity consumption was 21.3 percent a year over the 1966-74 period (table 9.3). Sales to large industrial consumers grew particularly rapidly, reflecting the gov- ernment policy of promoting industry. At year-end 1974 the total installed generating capacity in Korea was about 4,820 megawatts, of which 13 percent was hydroelectric, 79 percent steam, and 8 percent diesel and gas turbine. When comparing KECO's 4,524 megawatt capacity with its peak load requirements of 3. Average world consumption is total world consumption divided by total world population. See United Nations, World Energy Supplies (New York, 1976). 290 PART THREE: MAJOR SECTORS Table 9.3. Electricity Sales and Distribution, 1966 and 1974 1966 1974 1966-74 Average annual Giga- Per- Giga- Per- growth watt- centage watt- centage rate Sales category hours of total hours of total (percent) Lighting 502 i6.7 1,853 13.2 17.5 Small power 311 10.3 966 6.8 15.1 Large power 2,165 72.0 11,169 79.5 22.8 Agriculture 30 1.0 60 0.5 9.0 Total sales 3,008 100.0 14,048 100.0 21.3 Transmission and distribution losses 664 - 1,864 - 13.7 Auxiliary use 213 - 923 - 20.0 Gross generation 3,885 - 16,835 - 20.1 - Not applicable. Note: Sales and generation figures exclude self-producers. Source: Data supplied by Korea Electric Company (rEco). 2,922 megawatts at generator terminals in 1974, it is found that total generating capacity considerably exceeds what normally would be re- quired to ensure adequate supply. Part of this excess has already been absorbed, but it will not be until 1981, according to current development plans, that reserve margins are brought down to an acceptable level. In the early 1970s Korea decided to embark on a nuclear generating program aimed at achieving more than 50 percent nuclear generation by 1986. The first nuclear plant, Kori 1, with an installed capacity of 600 megawatts, was to go on stream by mid-1977. The second nuclear power plant, originally scheduled for 1979, has been delayed until 1982 because of problems in securing adequate financing and the decision to review KECO'S long-term development plans. The development of transmission and distribution facilities-together with the geographically balanced development of generating capacity- permitted a reduction in line losses from 15.7 percent in 1966 to 12.1 percent in 1974. Korea's 154-kilovolt transmission network has until recently been generally adequate as the backbone for bulk supply (table 9.4). It will progressively be superseded by a 345-kilovolt transmission system, which will interconnect the main generation plants and the first nuclear plant at Kori. Transmission and distribution systems were given second priority in the past and received only 23 percent of total power ENERGY 291 Table 9.4. Transmission and Distribution Facilities, 1974 Line length Substation capacity Type of facility (kilometers) (megavolt-amperes) Transmission 6,728.1 6,481.9 345 kilovolt, 189.3 ... 154 kilovolt 2,004.8 3,752.8 66 kilovolt 3,501.7 1,844.3 22 kilovolt 1,032.3 884.8 Distribution 6,730.5 3,164.8 ... Zero or negligible. Source: KFCo, Statistics of Electric Power in Korea, 1975. a. Under construction. sector investment during the 1962-73 period. Between 1973 and 1975, however, additions to transmission and distribution were substantial, particularly with the construction of the 345-kilovolt transmission system and the substantial additions to the number of transmission substations and distribution facilities. Petroleum Korea has no known hydrocarbon resources, but it does have a large petroleum industry engaged in refining, transport, and distribution. The industry is principally supplied by imports from the Middle East. Most production is for the domestic market, but refined oil products are exported to neighboring countries. As the petrochemical sector grows, exports of petroleum products will be gradually phased out. The first petroleum production facility, the Ulsan refinery, was con- structed in 1964. Since then its capacity has been expanded to about 215,000 barrels a day (10.75 million tons a year) and two more refineries have been constructed. The two largest refineries are located on the south- ern coast. The third refinery is located at Incheon on the northwestern coast close to the Seoul market and the Kyongin power plant, which it was originally intended to serve. The total refining capacity in 1974 was about 435,000 barrels a day (21.75 millions tons a year). This remark- able increase of about 28.6 percent a year was needed to supply domestic demand and meet the growing requirements of the petrochemical in- dustry. Domestic demand grew at an average rate of 25 percent a year from 1.8 million tons in 1966 to about 10.6 million tons in 1974. The 292 PART THREE: MAJOR SECTORS requirements of the petrochemical industry grew at nearly 35 percent a year and accounted for about 14 percent of total demand in 1974. Ex- ports grew at 21.5 percent a year from about 350,000 tons in 1966 to about 1.7 million tons in 1974. Exports in 1974, however, were about 16 percent less than those in 1973. During 1966-74 demand in the power generation and residential- commercial sectors increased significantly faster than total demand (table 9.5). The main reasons for this were the substitution of heavy fuel oil for coal in the power generation sector and the development of modern multistoried buildings which use fuel oil for space heating. Contrary to the pattern in most countries, transport demand in Korea grew at a lower rate than total demand, mostly because private owner- ship of cars has been discouraged. Korea has only about 200,000 road transport vehicles, of which fewer than 50 percent are private cars, and gasoline prices have been kept high. Demand in the industrial sector has grown faster than sector output, measured in 1970 prices, reflecting the decline of coal and the development of energy-intensive industries. The most remarkable change in the structure of demand for individ- ual petroleum products is the large increase in demand for Bunker C, a heavy residual fuel oil. As already mentioned, this increase is ex- plained by the rapid growth of petroleum demand in the power genera- tion and industrial sectors and the relatively small increase in gasoline demand. So far, market requirements and refinery yields have been satisfactorily matched, and there have been no imbalances requiring imports of products. In the future, however, if the internal demand Table 9.5. Consumption of Petroleum, by End-use Sector, 1966-74 1966-74 1966 1974 average _ _ _ _ _ _ _ _ _ annual Thou- Per- Thou- Per- growth sands centage sands centage rate Sector of tons of total of tons of total (percent) Power generation 164 9.1 2,932 27.5 53.5 Industry 640 35.5 4,146 38.9 26.3 Transport 583 32.3 2,051 19.2 17.0 Residential-commercial 75 4.2 935 8.8 37.0 Other 340 18.9 594 5.6 7.2 Total 1,802 100.0 10,658 100.0 25.0 Source: Figures supplied by MCI. ENERGY 293 for light fuels and petrochemicals grows faster than that for heavy fuel oil, Korea may have to import light products or install reforming and cracking equipment in existing refineries. The impressive growth of total petroleum demand, including exports, slowed down somewhat after 1970. The average rate of increase between 1970 and 1974 wvas only 11.7 percent, compared with an average rate of about 25 percent for 1966-74. This slowdown was mainly the result of slower growth in industrial production and the increases in the price of petroleum. In 1973, however, when GNP and industrial production increased at rates even higher than those of the late 1960s, total petro- leum demand increased by 19 percent and total domestic demand by 17 percent. The October 1973 increase in petroleum prices and subse- quent government action brought this growth to an abrupt halt. In 1974 total demand increased by only 1.6 percent; domestic demand by about 4 percent. Korea's petroleum import bill increased from about $300 million in 1973 to $1,120 million in 1974; the share of petroleum in total imports, from about 7 percent in 1973 to 16.2 percent in 1974. Petro- leum imports in 1974 were equal to about 25 percent of all export earnings excluding fuels, a pattem repeated in subsequent years. On the internal market the increase in the cost of petroleum supplies was reflected by an increase in the price of domestic products, particularly such light products as gasoline and kerosene. As a result, consumption of gasoline and kerosene in 1974 dropped below the 1973 level. In 1975 gasoline consumption continued to decline by 6.4 percent; kerosene demand exceeded the 1973 level by 3.6 percent. The reaction of government, combined with its earlier policy for pricing petroleum products, was thus instrumental in cushioning the impact of the increase in the cost of petroleum supplies. It is apparent, however, that the ensuing world recession was also a factor in keeping consumption under control and that an upturn of economic growth would result in further growth in petroleum demand. The potential for interfuel substitution is limited, and the costs of fuels, even after the price increase, is a comparatively small component of the cost of indus- trial products. Government's pricing policy has been to promote indus- trial development by providing low-cost energy to the productive sectors and bv heavily taxing the nonproductive sectors. This policy is evident in the schedule of taxes on oil products: there is no tax on kerosene and only a 5 percent tax on heavy fuel oil; these rates are much more favorable than the tax rate of 20 percent on light fuel oil, 40 percent 294 PART THREE: MAJOR SECTORS on diesel fuel, and 300 percent on gasoline. It is questionable, however, whether such a policy can be continued. The main potential for energy savings is in the industrial sector. Energy under the Fourth Plan The preparation of the fourth plan provided government with the occasion to formulate a medium-term energy policy and develop com- prehensive projections for a period of five to ten years. The task was complicated by the limited experience of Korean authorities in energy planning and by the magnitude of choices that had to be made in a short period to avoid disruptions of supply, particularly in the power and coal subsectors. The fourth plan nevertheless contains the basic objectives of Korea's proposed energy policy and the main policy actions required to achieve them. *Stable energy supply. Because Korea's dependence on imports will continue to increase during the fourth plan period, a long-term pro- curement plan for imported energy is to be drawn up, mainly for petroleum and coal, and adequate storage facilities are to be de- veloped. At the same time, exploration for and development of domestic resources-primarily coal and hydroelectricity, but also tidal power-are to be accelerated. * Maximum exploitation of domestic energy resources. The develop- ment of domestic energy resources is to be supported by adequate pricing policies, particularly in the coal industry, and adequate measures are to be taken for financing coal mines, modernizing mining technology, and improving the working and living condi- tions of coal miners. * Rational energy use. The Korean economy is to become more oriented toward energy conservation, and steps are to be taken to improve energy efficiency at user plants, promote energy conserva- tion in the residential-commercial sector, and develop new and more efficient techniques of energy use based on the results of research and development studies carried out in developed countries. In addition to these efforts, specific targets and policies have been set for each of the energy industries (table 9.6). The consumption of coal is expected to grow at an annual rate of 6.2 percent during 1974-81, compared with 3.3 percent a year during ENERGY 295 Table 9.6. Consumption of Commercial Energy, by End-use Sector, 1974-81 (thousands of tons of oil equivalent) Residential- Item Industry Transport Others commercial Total 1974 Coal 251 124 5,874 6,249 Petroleum 4,146 2,051 594 935 7,726 Electricity 1,015 5 173 1,193 Total 5,412 2,051 723 6,982 15,168 1981 Coal 415 43 8,539 8,897 Petroleum 7,238 4,200 841 1,874 14,153 Electricity 2,843 13 373 3,229 Total 10,496 4,243 854 10,786 26,379 Change in consump- tion, 1974-81 Coal 164 43 -124 2,665 2,748 Petroleum 3,092 2,149 247 939 6,427 Electricity 1,828 8 200 2,036 Total 5,084 2,192 131 3,804 11,211 Share in total increase in con- sumption (percent) 45.3 19.6 1.2 33.9 100.0 Zero or negligible. Note: According to the final version of the fourth plan, total consumption is to be as follows: for electricity, 3,370 TOE, or 11.3 percent of total consumption; for coal, 8,954 TOE, or 30 percent of total consumption; for petroleum, 17,501 TOE, or 58.7 percent of total consumption. These projections, particularly that for petroleum, appear to be more realistic than those presented in the preliminary draft. These figures were not broken down by end use, however, and no detailed analysis was possible at the time of writing. Source: World Bank estimates based on data supplied by MCI and EPB. 1966-74, and reach about 24 million tons in 1981. The coal development plan of the Ministry of Commerce and Industry (MCI) outlines the main features of basic strategy during the plan period: increased ex- ploration and completion of the geological survey; maximization of coal production by modem mining techniques; expansion of existing coal transport facilities; establishment of a coherent support system for the coal industry. 296 PART THREE: MAJOR SECTORS For the power industry supply and demand projections indicate that gross power generation should grow from about 16,835 gigawatt-hours in 1974 to about 45,300 gigawatt-hours in 1981. The average annual rate of growth is about 15.2 percent, compared with 21 percent for the previous period. The structure of demand is not expected to change substantially. Demand for industrial and large commercial uses wvill continue to account for more than 85 percent of total power sales. Total generating capacity is expected to grow from 4,720 megawatts in 1975 to about 10,052 megawatts in 1981 at an average rate of 16.3 percent a year. Even though this growth rate is slightly lower than that for peak demand, generating capacity is still projected to exceed peak require- ments in 1981 by about 26 percent. The increase in generating capacity is to be achieved primarily bv the introduction of one nuclear power plant of 600 megawatts, the completion of three hydroelectric plants and two pumped storage power plants, each with a capacity of 400 mega- watts, and the expansion of conventional thermal power plants. It is also expected that the transmission system will expand upon completion of the 345-kilovolt transmission lines, which will reduce losses and improve the quality of service. For the petroleum industrv it is difficult to estimate accurately the future internal consumption of refined petroleum products, mainly be- cause this consumption will depend on what happens in other sectors. For example, slower growth in coal production would push up the demand for refined petroleum products. Another problem is that the difference between energy and nonenergy use is not always clear, and some double-counting may occur. Resolving these difficulties will de- pend on the quality and reliabilitv of time series data and on the knowl- edge of the evolution of demand by end use. For Korea, time series data are not very useful. The period in which petroleum products have been used in significant quantities is relatively short. In addition, most infor- mation on demand by end use is held by private and semiprivate oil companies which do not wish to divulge information they consider to be confidential. These difficulties notwithstanding, consumption of petroleum products is projected to grow at about 10.8 percent a year over the plan period. This growth rate is less than one-half of that for the 1966-74 period. Demand is to be met solely from imports of crude oil, which will be refined in the private sector. Refining capacity is to increase from 435,000 barrels a day to 675,000 barrels a dav. The total primary energy requirement, including noncommercial energy, is projected to increase at an average rate of 7.9 percent a year from 20.3 million TOE in 1974 to 34.6 million TOE in 1981. Commercial energy requirements are to increase at about 9.2 percent a year from ENERGY 297 Table 9.7. Investment Program for Energy, 1972-81 Third plan 1972-76 Fourth plan 1977-81 Billions Percentage Billions Percentage Item of won of total of won of total Electricity 770 81.0 2,299 84.2 Generation 408 42.9 1,350 49.4 Nuclear 256 26.9 629 21.6 Transmission and distribution 351 36.9 824 30.2 Other 11 1.2 125 4.6 Petroleum 73 7.7 161 5,9 Coal 79 8.3 189 6.9 Technology 8 0.8 44 1.6 Conservation 14 1.5 35 1.3 Other 7 0.7 3 0.1 Total 951 100.0 2,910 100.0 Note: All figures are in 1975 prices. Source: EPB, Fourth Plan. 17.4 million TOE to 32.3 million TOE. The income elasticity associated with this growth is about 1.07 to GNP, slightly lower than that for the preceding period. The share of coal in total primary energy requirements should remain almost constant; the share of petroleum will continue to increase. Hydropower is also expected to increase slightly and be com- plemented by nuclear power, which will contribute about 2.2 percent to the total energy supply. The ratio of domestic production to total demand will decline from 39 percent in 1974 to about 34 percent; if noncommercial energy is included, from 48 percent to 38 percent. The total cost of the energy program, including interest during con- struction, is estimated at about W2,740 billion in 1975 prices (table 9.7). Of this amount about 84 percent is for power facilities, 7 percent for coal, 6 percent for petroleum, and 3 percent for others. Investment in the energy sector is planned to be about 16.5 percent of total invest- ment; investment in the power industry alone, 13.8 percent. Major Issues of Energy Policy Despite Korea's relatively poor endowment of natural resources, the energy sector has managed extremely rapid growth at reasonable cost. 298 PART THREE: MAJOR SECTORS The planning emphasis has been on increasing production capacity, not on modernizing management procedures and policies. Many decisions have been based on short-term considerations, not on long-term analysis and planning. The consequences of this approach were not detrimental in the period before oil prices increased, but higher energy prices and the burden of oil imports on the balance of payments signaled the need for revision of past policies. One thing is clear: whatever policy is adopted will be based on a combination of oil imports and nuclear power generation. It is not obvious, however, that the current or projected mix of fuels corresponds to the optimum use of resources. Nor is it obvious that the current and projected policies form an adequate framework in which the development of the energy sector can be optimized. The success Korea has in developing the energy sector will depend on gov- ernment's ability to make improvements in three fundamental activities: sector planning and coordination; energy pricing and allocation; policy formulation and implementation. Sector planning and coordination Korean planners, in devising the beginnings of a comprehensive energy program in the fourth plan, do not appear to have followed the integrated approach required to develop proper coordination procedures among the various energy subsectors and the client sectors of industry, transport, and urban development. The energy policy guidelines of the plan, although presenting a catalog of reasonable recommendations, do not constitute the framework of an energy policy. They do not distin- guish among priorities. They do not analyze the relation of the energy sector to the entire economy. They do not integrate subsector policies in an effort to achieve common objectives. For example, it is stated in the plan that coal production is to be maximized. But this policy objec- tive is not reflected in the policy statements for other fuel industries or for the client sectors. In fact, although the plan sets objectives for the energy sector, it appears to assume that the sector is independent from the rest of the economy and that the development strategy in industry or the residential sector will not substantially affect the future supply and demand pattems for energy. Efforts should therefore be made to improve coordination, perhaps by creating a system in which one central agency provides the frame- work of macroeconomic indicative planning and all decentralized agen- cies have full operational responsibility under the supervision of the ENERGY 299 Ministry of Energy and Resources (MER).4 Instead of creating new institutions, as currently proposed for the power industry and for energy conservation, government should strengthen existing institu- tions. Greater attention should be given to intersectoral relationships and to tradeoffs between energy consumption and alternative invest- ments. Greater attention should also be given to the problems of sector and subsector development in the medium and longer terms, particu- larly for the coal and power industries, and to the evaluation of the relative costs of coal, petroleum, and alternative modes of power gener- ation. Pricing Energy prices have been used in Korea to promote industrial develop- ment and to subsidize the poorest segments of the population. With the exception of petroleum products, energy prices have not generally been related to the economic cost of developing and exploiting energy re- sources. This pattern may have to change. The current method for pricing can lead to inefficiencies and to inappropriate evaluation of the costs and benefits of alternative development projects. Nor have energy prices been applied to influence resource allocation. This, too, may have to change. Concern about such matters is particularly apposite for Korea at a time when new policies are being recommended for energy supply and a strong program of energy conservation is being promoted. Two basic considerations will affect the shape of these new policies: financing the development of the energy sector; and ensuring that rising living standards do not introduce serious constraints in supply, partic- ularly in the industrial, commercial, and transport sectors. In a pure market economy, the price mechanism should discourage potential users and provide incentive to potential producers so that the volume and structure of supply and demand are at all times optimum. Of course, pure market economies seldom exist, and there are ways in which the price mechanism may instead convey wrong or untimely signals to consumers and producers. Government intervention is there- fore seen as required to correct some deficiencies of the price system and to take into account the extemalities not perceived by consumers and producers, such as environmental costs or the costs associated with retraining. The main problem associated with government intervention, 4. The Ministry of Energy and Resources (mER) was established in 1977. 300 PART THREE: MAJOR SECTORS which implies interference with price and market decisions, is selecting the decisions to be interfered with. It is recommended here that govern- ment should concentrate on prices and that these prices should, to the extent feasible, be related to the long-term marginal costs of supply. If this were done, the increases in prices of coal and power would be substantial, which in turn might create some problems of sectoral adjust- ment. It is likely that government action would be required to cushion to the impact of higher energy prices on such critical sectors as export- oriented industries and residential and commercial users. Such consid- erations should not, however, delay revision of the structure of energy prices. In addition, alternative means of subsidizing consumers should be devised, if it turns out that such subsidies are in fact required. In the short term, prices based on long-term marginal costs may not be sufficient to ensure the financial viabilitv of energy enterprises, as in the powver subsector, or to provide adequate incentives to attract new producers, as in the coal subsector. Producer prices should cover the full cost of production and provide a reasonable return on the capital invested. Thev should also enable utilities to finance a reasonable part -sav, 30 percent-of their future capital requirements. The level of prices that is adequate in each subsector to ensure financial viability would, of course, have to be determined in relation to each subsector's investment plan and to existing and forecast financial commitments. It seems likely that price increases can be imposed without serious dis- ruption to the industrial sector, which is the main consumer of energy. The indications are that steps are already being taken in this direction. Policy formizulation and implementation The formulation of a national energy policy requircs insight into almost all facets of the economic environment and presents planners and de- cisionmakers with a series of choices requiring understanding of the sector and its relation to the entire economy. It therefore is not a straightforward undertaking. In addition, the energy scene is rapidly changing and will continue to change. This makes it necessary for government and the agencies dealing with energv to make choices among alternative energy-development strategies and to decide on the construction of facilities that may come on stream five or ten years after investment decisions are first made. It is essential that energy stratecies be formulated rationally and that a system be established to enable evaluating and monitoring the consequences of these decisions and accumulating knowledge of the sector to facilitate responsible choices ENERGY 301 in the future. The process is continuous. To launch this process, how- ever, considerable political motivation and willpower are required, par- ticularly if a rational approach to energy development is to emerge in the long run. For the coal industry the problems associated with increasing produc- tion in the second half of the 1970s and the first half of the 1980s- labor, technology, mine size, and so on-should be assessed to devise appropriate policies for pricing, safety, working conditions, and technical and financial assistance. This assessment should focus on the creation of new capacity and on the assistance that KMPC could extend to private mines, especially during the initial stages of operation when heavy equipment is required for creating mine infrastructure. The success of the development plan for coal mining will depend largely on the supply of skilled coal miners. For example, 25,000 additional miners will be required during the fourth plan period. It therefore is urgent that KMPC launch a training program to upgrade the skills of miners and take actions to improve the safety and working conditions in underground mines and the housing and other services in mining communities. Ac- tions should also be taken to ensure coordination of transport and dis- tribution facilities, perhaps by creating a coordinating mechanism to link the efforts of KmPC and the Korean railways, and to ensure rehabili- tation of coal-fired boilers of power plants so that these can accommodate lo-w-grade coal or a mixture of coal and oil. For the power industry the most important actions to be taken are the following: review of the long-range power-development plan on the basis of projected demand; review of assumptions about the timing and cost of developing nuclear facilities as opposed to conventional plants; review of the rate structure for electricity and the financial viability of companies operating in the sector; review of the level of investment necessary to optimize the operation of the transmission and distribution system. Government has already taken steps in these directions and is considering changes in the organization of KECO. One proposal is for the creation of a separate company to be responsible for developing and operating nuclear power plants; another proposal is for splitting KECO into twvo or more companies. Final decision on these matters should, however, await the results of studies currently under way and the for- mation of a single entity responsible for the planning of power systems in Korea. For the petroleum industry the major problems are the adjustment of supply and demand and the impact of shortfalls in other energy sub- sectors on petroleum imports and the balance of payments. It is recom- 302 PART THREE: MAJOR SECTORS mended that a special group be created in MER to study the energy market and, given the results achieved in the coal and power subsectors, to maintain a continuous forecast of future petroleum requirements. It also is recommended that MER study the feasibility of public interven- tion in the transport and storage of oil products, with a view to promot- ing joint ventures and avoiding duplication of facilities constructed by the private oil companies. For energy conservation it is recommended that all energy agencies dealing with conservation, research, and development be placed under the supervision of MER, which could then promote a coordinated pro- gram of conservation, supervise its implementation, and monitor per- fonnance. MER should organize a unified and comprehensive system for collecting and maintaining energy data. Recording and analysis should be performed according to internationally accepted standards for the analysis of demand by end use, the segregation of primary and second- ary energy, and the segregation of energy and nonenergy use. It is particularly important that MER create an information system to make available the data gathered by energy conservation agencies to other organizations, whether private or public. Chapter 10 Tillman Neuner Transport SIGNIFICANT CHANGFS in the transport sector accompanied fundamental changes in Korea's economy over the 1965-75 period. The country's export drive and increased reliance on foreign trade led to a sixfold in- crease in the volume of port traffic, an eightfold expansion of Korean- owned oceangoing vessels, and a twentyfold rise in the volume of inter- national air travel. As GNP more than doubled and the share of industry in GNP rose from 14 percent to 30 percent, the railway lost its prepon- derant position in transport and Korea established a diversified road transport system. The length of paved roads more than quadrupled over the decade. Although the ratio of private cars to population in 1975 was lower than that in such countries as Indonesia and the Philippines, Korea about matched these countries in the ratio of trucks and buses to population. The share of coastal transport in goods traffic also increased greatly, primarily because of the movement of oil products. As Korea's economy continues to grow and the pace of industrializa- tion intensifies, transport will experience a parallel expansion. The in- ternal transport system will have to accommodate much greater con- sumption of certain widely used bulk commodities. One study projected the following increases in domestic demand between 1973-74 and 1991: anthracite from 14.8 to 33 million tons; petroleum products from 14.6 to 87.6 million tons; cement from 8.3 to 36.4 million tons; fertilizer from 1.6 to 3.7 million tons; timber from 8.6 to 17.3 million cubic meters.' Industrial complexes growing up on Korea's south and south- east coasts will result in the import of large amounts of industrial raw 1. All tons are metric tons equal to 1,000 kilograms. King and Gavaris Con- sulting Engineers, Inc., and Trans-Asia Engineering Associates, Inc., a joint venture in association with Arthur D. Little International, Inc., "Korea Ports Phase Two Development Study: Final Report," 10 vols. (Seoul, 1976; processed), vol. 7. 303 304 PART THREE: MAJOR SECTORS materials and the shipment of intermediate products between plants. Some of these plants may also become large volume exporters. In addi- tion, a more diffuse pattern of growth in nonbulk traffic and passenger traffic will follow upon the general increase in GNP. Sectoral Structure Traffic statistics demonstrate the changing structure of the transport sector (table 10.1). The share of rail transport in passenger traffic de- clined from 46 percent of total passenger-kilometers to 24 percent over the 1965-75 period; the share of road transport increased from 53 per- cent to 74 percent. The share of rail transport in goods traffic declined from 86 percent of total ton-kilometers to 52 percent over the same period; the share of road transport increased from 9 percent to 21 per- cent; the share of coastal transport increased from 5 percent to 26 per- cent. These figures are based on official returns of licensed carriers and do not take into account the substantial volume of persons and goods moving in privately owned vehicles. This omission causes an understate- ment of road traffic volume. In 1965 privately owned vehicles already represented 42 percent of sedan registrations and 24 percent of truck registrations; in 1974 these proportions had respectively risen to 58 per- cent and 46 percent. The modal balance has thus changed to reduce the once dominant position of the railways, which have nevertheless assumed new roles in Korea's increasingly complex economy. Three bulk items-anthracite, cement, and petroleum products-ac- counted for nearly 60 percent of all freight ton-kilometers in 1974. Bulk movements characterized both rail and coastal transport: seven com- modities made up about 80 percent of rail ton-kilometers; petroleum products alone constituted more than 70 percent of coastal shipments. Moreover increases in the three principal commodities accounted for more than 80 percent of the rise in rail ton-kilometers between 1965 and 1974. Road transport carries mostly general goods, which accounted for more than one-half of total traffic in this category in 1974. Korea's rail and road networks extend to all parts of the country, and well-developed ports are available at intervals of 200 kilometers or less along the coastline. A system of farm roads, aggregating nearly 40,000 kilometers by the end of 1974, was built up in recent years. It is not known, however, how many rural communities remain without road TRANSPORT 305 Table 10.1. Volume of Domestic Traffic, 1965-75 Category/mode 1965 1970 1975 Freight (millions of tons) 49.1 103.8 139.1 Railway 22.4 31.6 42.8 Highway 24.0 61.8 84.5 Coastal shipping 2.7 10.5 11.8 Freight (billions of ton-kilometers) 5.8 13.4 17.8 Railway 5.0 7.7 9.3 Highway 0.5 1.4 3.8 Coastal shipping 0.3 4.2 4.7 Passenger (millions of passengers) 1,308.4 2,881.5 4,804.1 Railway 107.2 131.0 197.8 Subway . 56.8 Highway 1,195.5 2,743.8 4,342.7 Coastal shipping 5.5 5.9 5.9 Air transport' 0.2 0.9 0.9 Passenger (billions of passenger-kilometers) 15.1 30.3 52.9 Railway 6.9 9.8 13.1 Subway 0.3 Highway 8.0 20.0 38.9 Coastal shipping 0.2 0.2 0.2 Air transport 0.1 0.3 0.3 Port' (millions of tons) 11.4 43.3 68.2 Exports 1.6 3.6 10.1 Imports 5.2 18.7 35.1 Coastal shipping' 4.6 21.0 23.1 ... Zero or negligible. Sources: Ministry of Transportation (MOT), Statistics Yearbook of Transport, 1975 and earlier years; figures supplied by MOT. a. There were 77,000 international air passengers in 1965, 392,000 in 1970, and 1,566,000 in 1975. b. Of total port traffic, Incheon handled 12.2 percent in 1965, 18.5 percent in 1970, and 17.4 percent in 1975; Busan handled 38.5 percent in 1965, 21.3 per- cent in 1970, and 24.5 percent in 1975. c. Includes entered and deared coastal cargo. 306 PART THREE: MAJOR SECTORS access. Few areas are completely isolated from Korea's transport network, but in many cases the only link is a tortuous and badly maintained road. Additional information on Korea's transport sector is presented in tables 10.2, 10.3, and 10.4. Only a few outstanding characteristics need be mentioned here. Virtually all railway traction now is diesel or electric; rail freight is moved in substantial loads, averaging more than 40 tons a wagon, but over short distances averaging less than 250 kilo- meters. The highway network is still 80 percent unpaved; 52 percent of road vehicles are buses and trucks. Tankers account for more than one- half of gross registered tons (GRTs) of coastal and oceangoing vessels. Transport facilities are concentrated along the Seoul-Busan axis and the corridors serving the country's northeast quadrant, which contains the major coal deposits and most cement plants. Busan and Incheon, the city Iying immediately west of Seoul, are Korea's most important ports. In 1975 Busan handled 25 percent of total traffic; Incheon 17 percent. The Seoul and Busan metropolitan areas also have Korea's largest concentrations of population and industry. The intervening cor- ridor includes Daegu, the third most important industrial center, and Daejeon, a city of about one-half million inhabitants. The corridor is served by a double-track rail line, two main highways including a limited access four-lane road, and roughly one-quarter of all domestic flights. Table 10.2. Railway Survey, 1965-75 (units) Item 1965 1970 1975 Steam locomotives 272 109 87 Diesel locomotives 125 277 375 Electric locomotives ... . 66 Diesel railcars 83 158 123 Suburban electric railcars ... 126 Passenger carsa 1,390 1,563 1,717 Freight cars 10,587 14,407 15,866 Length of line (kilometers) 2,980 3,193 3,144 Average haul per passenger (kilometers) 65 75 59 Average haul per ton of cargo (kilometers) 225 244 217 Operating ratio (percent) 89 84 98 ... Zero or negligible. Source: Figures supplied by Korean National Railroad. a. Excludes railcar trailers. TRANSPORT 307 Table 10.3. Highway Survey, 1965-75 Item 1965 1970 1975 Road network (kilometers) 28,145 40,245 44,905 National highways (kilometers) 5,999 8,659 9,375 Paved 1,042 2,461 4,762 Gravel 4,849 6,158 4,601 Unrepaired 8 40 12 Local roads (kilometers) 22,146 31,586 35,530 Paved 585 1,403 5,238 Gravel 17,874 26,803 26,694 Unrepaired 3,794 3,420 3,610 Length of highways per square kilometer of land area (kilometers) 0.29 0.38 0.46 Motor vehicle fleet (units) 41,511 129,371 200,521 Passenger cars 13,001 60,677 84,212 Passenger buses 9,316 15,831 21,818 Trucks 16,015 48,901 82,862 Other' 3,179 3,962 11,629 Population per vehicle (persons) 771 247 176 Sources: Figures supplied by Ministry of Construction and MoT. a. Includes motorcycles. The coal-cement region is traversed by an electrified rail line reaching from Seoul to the east coast port of Mugho; other rail lines radiate from this region to the south and southwest. A road for high-speed traffic between Seoul and Mugho, passing just north of the main coal-cement area, was opened in 1975. Road connections from this area southward have been upgraded or are being improved. Pricing and Finance Government fixes all prices in the transport industry. Prices for pub- licly owned transport enterprises, such as the railways, are proposed by enterprise management to the Minister of Transport, who in turn sub- mits the proposals to EPB. The Price Bureau of EPB analyzes these 308 PART THREE: MAJOR SECTORS Table 10.4. Shipping Survey, 1965-75 (gross registered tons) Type of vessel 1965 1970 1975 Oceangoing 163,301 717,680 1,320,085 Passenger 915 915 Cargo 156,844 467,878 742,116 Oil tanker 5,242 248,887 577,969 Coastal 66,000 188,036 204,579 Passenger 14,886 17,742 18,042 Cargo 42,498 96,073 78,549 Oil tanker 8,616 74,221 107,988 Other 156,485 298,948 552,503 Total 385,486 1,204,664 2,077,167 Zero or negligible. Source: Figures supplied by MOT. proposals and presents them to a government committee on utility prices. Recommendations of that committee are then presented to the cabinet. Prices of privately owned transport are subject to the same review, ex- cept that proposals originate in Ministry of Transport (MtOT) divisions, which act on the basis of representations made bv industrv organizations. In addition to explicit regulation of prices, government indirectly affects prices in the transport sector by tax policies, subsidies to various trans- port operators, freight equalization of basic commodities, and user charges for such government-furnished facilities as airports and roads. The system for regulating transport prices concentrates on standard line-haul charges, the easiest and most obvious point to applv regulation. The system gives onlv limited attention to the numerous special prices which are charged by Korea's transport agencies and which inevitably play a major role in any complex transport operation. Regulations are based on cost data generally characterizing the modes, not on detailed costs for point-to-point movements. Moreover none of the modes has a comprehensive costing system. For road transport and coastal shipping, govemment negotiates with industry representatives and bases its deci- sions on cost data supplied by them. The regulatorv svstem thus cannot make subtle distinctions. This inabilitv is not necessarilv a fault, but it is a limitation to be considered when evaluating the system. The extent of regulation varies among the modes. For some modes and TRANsPORT 309 services, such as railways and buses, the regulated rate is inflexible. For others, such as highway freight, the regulated rate constitutes a maxi- mum and may be negotiated downward. Truck rates, for example, may be as much as 40 percent less than official rates. Although coordination of pricing is one aim of regulation, limitations of existing institutions and available information preclude meeting this aim effectively. The Transport Coordination Office (rco) in MOT could play an active part in coordination, but has not yet done so. The regulatory process also fails to take adequate account of the many transport activities which are not subject to regulation, but which inevitably affect the regulated sector. These activities include the operation of cars, buses, trucks, and ships by nontransport companies for their own account. The flow of funds through Korea's public sector transport enterprises is substantial, but these enterprises do not appear to generate significant surpluses for use in other parts of the economy. In addition, the degree of financial self-sufficiency varies considerably among parts of the pub- liclv owned transport sector. The finances of Korean National Railroad (KNR) are the largest single element in the publicly owned part of Korea's transport system. In 1975 KNR revenue was nearlv W77 billion, which covered all costs including depreciation, but not interest charges. Government therefore paid a subsidy of more than W14 billion to KNR for 1975. KNR investment that year wvas about W45 billion, of which about W8 billion were derived from KNR's depreciation reserves. The remainder came from government loans and foreign credits. The Korea Highway Corporation (KHC) had revenue, mostlv from tolls, of about W8 billion in 1974. Its expenditure, primarily for maintenance of expressways, amounted to roughly W5 billion. No comprehensive data is available on port finances. One consultant estimated in 1971 that port revenue collected by fourteen main ports amounted to W820 million.2 This total has risen steeply since then; the port revenue of Busan alone was W2.2 billion in 1975. It has been esti- mated that revenue of Korea's main ports will exceed W22 billion by 1978.3 It is not known whether port charges are now sufficient to cover expenses; in the past revenue constituted only a fraction of expenses. Port investments are estimated at about W40 billion in 1975. 2. Lyon Associates, Inc., in association with Economist Intelligence Unit Ltd., "Korea Port Development Study," 9 vols. (Seoul, 1972; processed), vol. 5, p. 103. 3. Booz, Allen and Hamilton International, Inc., "Korean Port Authority Tariff Analysis Guidelines" (Seoul, 1975; processed). 310 PART THREE: MAJOR SECTORS Korea's airports levy user charges, including landing fees for aircraft and departure taxes for passengers, but it was not possible to determine the total amount collected for use of aviation infrastructure. In addition, accounts were not available to compare government revenue and ex- penditure for aviation. Because government investment in aviation infra- structure is small in comparison with other parts of the transport system, it is reasonable to assume that government revenue and expenditure related to aviation also constitute a small part of government finances in the transport sector. The annual flow of funds through Korea's publicly owned transport organizations was thus about W200 billion in 1975, including current and capital expenditure. No common basis exists for measuring the financial performance of these organizations. Road transport and coastal shipping are privately owned. A census of transport enterprises was conducted in 1964, but no comprehensive information has since been collected on this subsector. Nevertheless road transport appears to be self-supporting: no operating subsidies are paid by government; capital is provided to the industry through commer- cial sources. Although precise information is not available, highway user charges-that is, taxes on transport, vehicles, and fuel-probably more than cover expenditure on roads by various levels of government. In 1975 user charges exceeded W150 billion; the expenditure by central government for highways was about W55 billion; that by local authori- ties betveen W20 billion and W30 billion. But these user charges include taxes on gasoline and diesel fuel for all users, not for motor vehicles alone. Moreover, because about one-half of user revenue comes from the gasoline tax, gasoline-powered vehicles (mostly cars) may be subsidizing the diesel fleet (mostly buses and trucks). This distortion is reflected in the retail prices of the two fuels: gasoline sells for roughly W200 a liter; diesel oil for W70 a liter. The price of diesel fuel, at about 58 cents a gallon, is also low in absolute terms. This raises the question whether gasoline purchasers are paying a tax on top of the real price of gasoline to the Korean economy.4 4. The high relative price of gasoline might be viewed as a congestion charge for vehicles in urban areas, where cars are concentrated. But the price also tends to distort the choice of technology: many taxis now use liquefied petroleum gas; very light trucks are powered by diesel engines. This tendency will be further accentuated if government implements proposals for further increases of the gasoline tax, as is discussed in chapter 11. TRANSPORT 31I Coastal shipping receives significant direct and indirect aid from the Treasury. Shippers of coal by coastal vessel receive a government sub- sidy of about W1,300 a ton. Government also pays subsidies for vessels operating to remote islands. Information was not available on the total amount paid each year for these purposes. Coastal operators also receive indirect financial support through low interest loans for equipment from the Korea Development Bank (Kma). In addition, coastal shipping bene- fits from tax advantages extended to Korea's merchant marine and from port charges that may be set too low to cover the cost of port operation. Accounts are not available for the charging system for aviation infra- structure. Korean Air Lines (KAL) may benefit from the use of this infrastructure without paying for its full cost. But domestic operations account for only a limited part of KAL'S revenue-20 percent in 1973- and therefore have a limited effect on KAL'S finances. KAL does not receive significant financial assistance from government; most of its long- term debt comes from foreign lenders. Licensing System for Private Transport Government licenses private firms that operate services for hire in Korea's road and coastal transport systems. Road transport licenses are issued for scheduled bus and truck services along established routes and for nonscheduled taxi and truck operations in defined areas. There is no commodity licensing for trucks, but special licenses are issued for tractor trailers, tank trucks, and container trailers. In 1974 about 2,000 licenses were in force for the operation of motor vehicles for hire. In the three largest categories the average fleet of licensees ranged from 44 to 48 units. Express bus companies and scheduled trucking enter- prises tend to be larger on average than other operators. Gwangju High- way Company operates more than 200 buses in all but one of Korea's provinces. Korea Express Company has more than 1,700 trucks nation- wide. The place of registration for road transport vehicles indicates the concentration of vehicle registration in the two large metropolitan areas. Seoul and Busan accounted for 68 percent of cars and jeeps, 41 percent of all buses, 55 percent of larger buses, 47 percent of all trucks, and 52 percent of trucks under five tons. Only the larger trucks are spread more evenly throughout the country. MOT issues licenses for coastal shipping. In 1975 it licensed 64 com- panies operating about 140 ships with a total capacity of about 16,000 312 PART THREE: MAJOR SECTORS GRTS for passenger service and 241 companies operating about 1,300 ships with a total capacity of about 81,000 GRTS for cargo business ex- cluding petroleum transport. Most vessels are small. Since 1970 govern- ment has sought to increase the size of cargo ships in the coastal trade, but the average size-61 GRTS in 1970 and 62 GRTs in 1975-has re- mained almost the same. Small ships will probably continue to be used for services to offshore islands and remote mainland locations having limited port facilities. Only larger vessels can compete with the railways for the large volume of coal and other bulk commodities flowing from northeast to southwest and southern Korea. For coastal shipping the minimum requirements are that the licensee must control ships aggre- gating more than 100 GRTS for passenger vessels and 150 GRTs for cargo vessels, have capital of at least WI million per vessel, and have shore facilities to handle passengers or cargo. The authorities follow a similar procedure for licensing road operators. Depending on the type, scheduling, and location of service, the licensee must operate a minimum of from 5 to 50 taxis, 10 to 30 buses, and 5 to 30 trucks. The minimum fleet requirement for road transport does not mean that the licensee must own the requisite number of vehicles. Vehicles frequently are individually owned, and licenses are issued to a loose association of owners. Road transport licensees must also have certain minimal office and parking facilities and provide their own capital for roughly 30 percent of their assets. Taxi and truck licensing requirements were recently relaxed to allow the issuance of licenses to owner-operators of one vehicle, to allow the waiving of the five-unit fleet requirement in appropriate cases, and to exempt pickup trucks from licensing. Licensing authorities attempt to match licensed capacity to demand. For road transport, surveys are annually conducted to determine the number of passengers carried daily by taxis and buses and the number of kilometers traveled daily by trucks. If the use factor exceeds a set daily standard-29 persons for buses and 170 kilometers for trucks- additional capacitv is licensed. This system is based on fairly uniform vehicle sizes and makes no allowance for improvements in productivity, whether through acquisition of larger vehicles or better use of existing vehicles. Licensing authorities have tried to maintain a stable number of licensees and to increase capacity by authorizing proportionate incre- ments to the fleets of each licensee. During the first half of the 1970s the number of bus licensees fluctuated between 328 and 378; that of TRANSPORT 313 truck licensees between 614 and 769. For both types of license the total number of vehicles steadily increased. For licenses issued by routes- there were about 4,000 bus routes in 1975-capacity assessments are made for each route and additional vehicles are accordingly assigncd. Licensing authorities also control the number of daily runs offered by each bus licensee on a particular route. The exact time of runs is selected by the operator and is only reported to government. New routes are authorized when demand has been demonstrated for the service. Coastal passenger services are also organized by routes-there were 143 routes in 1975-and operators provide monthly reports on perfor- mance. If the reports show vessels to be operating at more than 70 percent of capacity, MOT authorizes additional ships for that route. New routes are authorized on the basis of demand. Coastal cargo vessels are licensed for operations anywhere along the coast. MOT, in consultation with other ministries, makes detailed annual forecasts of the principal commodities to be moved and the share to be assigned to coastal shipping. The fore- casts also include an assessment of likely peak periods. Coastal cargo capacity is licensed on the basis of these demand forecasts. Transport under the Fourth Plan One basic objective of government is to develop a transport system that meets future increases in traffic demand. Several policies set out in the fourth plan will influence the pursuit of this objective. * Rate adjustments and other regulatory means are to allocate medium- and long-distance traffic to the railways and generally increase transport efficiency. - Rate increases and operating improvements are to reduce the fiscal burden railways impose on government. * Greater weight is to be put on expenditure for maintenance than for new investment. - The organization and management of road transport enterprises are to improve. * The increased share of freight carried in Korean vessels is to en- hance the country's international payments position. * Transport systems are to promote balanced regional development. * The formulation of a long-term urban transport plan, including additional mass underground transit, is to reduce traffic congestion in the Seoul metropolitan area. 314 PART THREE: MAJOR SECTORS The plan addresses most major issues confronting the transport sector. Additional strategic objectives emerge from proposals for modal invest- ment programs, for example, the emphasis on paving main roads and restraining the growth of the fleet of private automobiles.5 Demnand forecasts Demand trends have been analyzed in detail only through 1981, the last year of the fourth plan period. An interministerial working group on transportation (WGT), with the assistance of several specialists from semigovernment research organizations, prepared detailed transport de- mand forecasts for this period, relying largely on contract work by the Korea Development Association (KDA), a private consulting group. The WGT's December 1975 forecasts were amended in final revisions of the fourth plan to allow for reduced consumption of petroleum products and revised annual export targets. An increase from 56.9 billion to 92.1 billion passenger-kilometers and from 19.6 to 30.6 billion ton-kilometers is foreseen over the period of the plan. Passenger transport is projected to grow more slowly in the fourth plan period than in the previous period, but will continue to increase at 10.1 percent a year, or slightly faster than GNP (table 10.5). Freight trans- port will rise by 9.3 percent a year, or more than twice the rate of ton-kilometer growth in the third plan period.6 Freight traffic is thus predicted to stay parallel with GNP during the fourth plan period; it grew more slowly than the economy as a whole during the third plan period. The absolute increases in the volume of services to be provided by 1981 are 35 billion passenger-kilometers and 11 billion ton-kilometers, roughly 40 percent more than the increase in passenger-kilometers and 190 percent more than the increase in ton-kilometers under the third plan. Investment program Transport investment under the fourth plan is projected to be W2,784 billion in 1975 prices, or 14.6 percent of total investment of W19,028 5. EPB, The Fourth Five-Year Economic Development Plan, 1977-81 (Seoul, 1976), pp. 53-57. 6. The government's formulations for the transport sector in the fourth plan used the base year 1972, when freight traffic decreased and therefore showed a higher rate of traffic growth for the third plan period than if 1971 had been used. The comparisons here and in table 10.3 use 1971 as the base year. TRANSPORT 315 Table 10.5. Volume of Domestic Traffic, 1971-81 Annual groieth rate 1971-76 1976-81 Category/mode 1971 1976 1981 (percent) Passenger (million passenger-kilometers) 32,237 56,868 92,089 12.0 10.1 Railway 8,750 14,185 18,930 10.2 5.9 Subway ... 377 609 ... 10.1 Highway 22,917 41,709 71,628 12.7 11.4 Coastal shipping 256 265 364 0.7 6.6 Air transport 314 332 558 1.1 10.9 Freight (million ton-kilometers) 15,796 19,647 30,643 4.4 9.3 Railway 7,841 9,962 13,638 4.9 6.5 Highway 3,302 4,333 7,649 5.6 12.0 Coastal shipping 4,653 5,349 9,352 2.8 11.8 Air transport ... 3 4 - 5.9 ... Zero or negligible. - Not applicable. Note: Figures for 1971 are actual, figures for 1976 are estimated, and figures for 1981 are planned. Sources: Figures for 1971 supplied by EPa; EPB, Fourth Plan. bi]lion. Corresponding investment under the third plan is estimated to have been about W1,860 billion in 1975 prices, or 16.3 percent of total investment. Thus, whereas W1,860 billion, in addition to renewals, enabled an increase of about 25 billion passenger-kilometers and 4 bil- lion ton-kilometers during 1972-76, W2,784 billion (or 50 percent more investment) is expected to provide an increase of about 35 billion passenger-kilometers and 11 billion ton-kilometers during 1977-81. The traffic mix changes over time, and the investment needed to produce a passenger- or ton-kilometer varies for this reason. Furthermore the cost of renewals rises as the system expands. Thus investments cannot be precisely related to traffic. Although the 40 percent increase in passenger- kilometers roughly corresponds to the increase in total investment from the third to the fourth plans, freight traffic is to grow much more rapidly than transport investment. Based on these measures, transport invest- ment in the fourth plan is modest in relation to demand and implies significant productivity gains in the transport sector. The transport in- 316 PART THREE: MAJOR SECTORS vestment originally planned for the third plan period was 22.5 percent of total investment; actual investment was 16.3 percent of total invest- ment. Similar 3Iortfalls could occur in the execution of the fourth plan: for example, in the road program. Greater effort must therefore be m-de to achieve efficiency in transport investment to avoid having the supply of transport services fall substantially below demand. If the economy grows more quickly than expected, however, certain factors of the transport system should promote a rapid response on the supply side without great risks of crippling bottlenecks. The fixed infra- structure of rail lines, roads, and ports in aggregate still has considerable unused capacity which could be put into service in a short time. The vehicle fleet for road and rail could be augmented, and Korea has already demonstrated its skill for efficiently organizing the intensive use of mass facilities. Such measures might be regarded as higher cost alternatives to meet unexpected demand increases until cheaper permanent solutions can be realized. Korea has also demonstrated its capacity for rapid execu- tion of large-scale public works, such as the Seoul-Busan expressway. This capacity suggests that interim measures would not last for an un- duly long period. The modal allocation of transport investment under the fourth plan is presented in table 10.6. About one-third of the W402.1 billion pro- posed for railways is for the acquisition of rolling stock; one-quarter is for increases in line capacity; one-sixth is for renewals and improvements to tracks and bridges; the remaining quarter is for miscellaneous invest- ments, mostly devoted to increasing capacity and avoiding major bottle- necks. Of investment proposed for road transport, slightly more than one-third is for highwavs. The remainder, to be made bv the private sector, is for replacements and additions to the commercial motor vehicle fleet. More than one-half of the proposed road construction program is devoted to paving about 4,100 kilometers of roads, including about 3,700 kilometers of national roads. The plan proposes that 91 percent of na- tional highways be paved by 1981. Roughly one-quarter of the invest- ment program for ports and shipping is allocated to ordinary ports, which are first-class ports constructed and operated bv the Korea Mari- time and Port Authority (KMPA), and industrial ports, which are being constructed at industrial sites. Private investment in shipping, projected at W733.3 billion, is targeted to increase the gross registered tonnage of Korea's ocean fleet by a factor of 2.2 and that of the domestic fleet by a factor of 1.7. The increase in the ocean fleet is intended to match the increase in Korea's foreign trade and give Korean flag vessels 50 per- cent participation in that trade by 1981, compared with 38 percent in TRANSPORT 317 Table 10.6. Investment Program for Transport, 1977-81 (billions of won in 1975 prices) Public Private Mode sector sector Total Railways 396.9 5.2 402.1 Road transport 460.8 764.8 1,225.6 Highways 457.8 . 457.8 Motor vehicles ... 741.2 741.2 Other 3.0 23.6 26.6 Ports and shipping 244.0 752.1 996.1 Ports 239.8 11.4 251.2 Shipping 733.3 733.3 Other 4.2 7.4 11.6 Air transport 54.7 4.2 58.9 Airports and navigational aids 54.7 54.7 Aircraft ... 4.2 4.2 Subway 101.1 ... 101.1 Total 1,257.5 1,562.3 2,783.8 . . Zero or negligible. Source: EPia, Fourth Plan, pp. 152-53. 1976. Airport investment is concentrated on facilities that handle the bulk of international traffic and the best-patronized domestic routes. Subwav investment is for expanding the subway system in metropolitan Seoul. Investment allocations under the third and fourth plans are compared in table 10.7. The substantial increase in the share of maritime transport is in part the result of the planned expansion of Korea's merchant marine and industrial ports. The reduction in the share of aviation is the result of the shift in the basis for fleet expansion from purchase to leasing. The shares of rail and road transport decline only slightly. Overall the transport investment program does not show any obvious internal imbalances. The emphasis on road transport reflects the facts that most of Korea's roads remain unpaved and that its level of motoriza- tion is low compared with countries of equivalent income. As already 318 PART THREE: MAJOR SECTORS Table 10.7. Composition of Transport Investment, by Mode, 1972-81 (percent) Third plan Fourth plan Mode 1972-76 1977-81 Railways 15 14 Road transport 47 44 Ports and shipping 28 36 Aviation 7 2 Subway 3 4 Total 100 100 Sources: Figures for 1972-76 supplied by EPB; EPB, Fourth Plan. indicated, modal changes are likely, but only in the longer term. They do not appear in the medium-term investment projections of the fourth plan. To meet the fourth plan's expenditure targets, average annual bud- getary allocations will have to rise by about 20 percent (table 10.8). With a gradual increase in annual budgets over the five-year period, the annual budget in 1981 would have to be about 35 percent more than the 1976 budget to achieve the targeted investment. If the economy grows as expected, increases of this magnitude should be within the revenue capacity of government and not go beyond trends observed in the third plan. The proposed investment in roads increases most, but actual expenditure may be considerably less because of shortfalls in in- vestment. Increased charges for services should reduce the government's revenue burden for railways and ports. Major Issues of Transport Policy The fourth plan's investment program for transport is indicative. It includes numerous projects proposed by individual modal agencies and screened on a preliminary basis by wGT; these projects will be separately reviewed by EPB along with the annual budgets. Under existing pro- cedures, neither project formulation nor review gives adequate consid- eration to new transport technologies or to intermodal comparisons in meeting a particular transport demand. The charter of the interminis- terial transport coordination committee (TCC) includes intermodal co- TRANSPORT 319 Table 10.8. Transport Investment by the Public Sector, 1975-81 (billions of won in 1975 prices) Proposed One-fifth of Revised budget total planned budget for investment Mode for 1975 1976- for 1977-81 Railways 56.0 72.9 79.4 Highways" 54.8 54.1 71.5 Ports - 43.7 47.9 Ordinary 20.5 32.8 28.0 Industrial - 10.9 19.9 Airports 2.8 5.7 10.9 Total - 176.4 210.1 - Not available. Source: Figures supplied by EPB. a. The 1976 budget figures have been deflated by 1.12 to convert them to 1975 prices. b. Includes only the budget of the Ministry of Construction. ordination of transport sector investment, but TCC has not exercised this function. Because EPB exercises budgetary control, it inevitably has the dominant voice in decisions on investment projects. EPB' plans to make much more intensive reviews of investment projects presented to it under the budget and to increase its staff for that purpose. Large-scale fixed investment projects, such as railway lines, will be referred for review bv EPB's budget officers to the Infrastructure Division of EPB'S Planning Bureau. But other investments, such as railway car purchases, will continue to be analyzed only in EPB's Budget Bureau. It would be highly desirable for these other investments to be subjected to an eco- nomic review based on intermodal comparisons. This end could be achieved by making the same group of budget officers responsible for all transport investments and training them in the appropriate analytical techniques. Intermodal coordination of transport investment projects would thus be provided by EPB. The final size and structure of the transport investment program must await detailed review, primarily through feasibility studies, of all its components. Feasibility studies are available or planned for the bulk of 320 PART THREE: MAJOR SECTORS the program of ordinary ports and national road works. Based on the results of studies already completed, most of these projects are likely to be approved. But few feasibility studies are available for railways and industrial ports. Review of these modes nevertheless suggests some pre- liminary conclusions. Many problems of line capacity, which the railway proposes to increase by doubling or electrifving tracks, could be met by cheaper methods, particularly the use of longer and heavier trains. In addition, the design of several industrial ports could be made more economical. Design alternatives, scaling and timing of investments, and locational considerations have not yet been properly analyzed. Gen- erally scope still exists for improving maintenance performance and substituting modest expenditure of this nature for larger investments in capital assets. The need for further studies is particularlv urgent for proposed invest- ments that are part of a much larger transport complex. For example, rail investments in metropolitan Seoul need to be reviewed as part of an overall transport plan for that region. Improvements in the Seoul- Busan rail line should be implemented only after a comprehensive investigation of all transport needs in the Seoul-Busan corridor. In some cases meeting demand by lower cost modes, such as moving large volumes of petroleum products by pipeline rather than by truck or rail- way car, should be considered. In addition, some physical investment goals, particularly the program to pave nearly all national highways and the plan to move 50 percent of Korea's foreign trade in Korean flag vessels, need to be moderated by economic criteria. The overall scope and direction of modal investment plans thus are generally suited to meet projected demands, but there remains consider- able scope for economizing on transport investment. Perhaps the largest single item that could be reduced without undue risk is the expansion of Korea's merchant marine. Continued use of some foreign shipping provides a readily available alternative. Economies are also likely to emerge from the detailed reviews of some major infrastructure projects: railways, roads, and ports, especially industrial ports. Finally, improved use, particularly of railways and ports, may enable canceling or delaying some investments. Assessments of transport demand For major bulk commodities the working group on transportation (WGT) based its transport demand projections on govemment forecasts TRANSPORT 321 of production and consumption and on analyscs of consequent traffic flows. These projections show that major added loads on the domestic transport system will arise from increased consumption of petroleum products, timber, anthracite, and cement. According to revised govern- ment data, production of anthracite is slated to rise from 17.6 million tons in 1975 to 24 million tons in 1981; production of cement from 10.9 to 19.7 million tons. There appears to be no possibility that anthracite, which is primarily consumed as household fuel, could be economically supplanted by imported oil in that use. Government now provides heavy subsidies for coal production and transport, but even if these were elimi- nated, local coal would probably be cheaper than oil as a household fuel. Imported coal from Australia or North America might make inroads on the local anthracite market. If that occurred, total transport demand would remain unchanged, though traffic flows would be substantially altered. For nonbulk commodities and passenger traffic, wGT based its projections on the extrapolation of regression analyses of past trends. After determining the projected flows of major commodities in Korea's transport system, WGT assumed that individual modes would divide traffic as follows: short- and medium-range movements, generally 200 kilo- meters or less, would go by road; hauls longer than 200 kilometers by rail; bulk commodity movements between coastal industrial areas by coastal shipping. This assumption reflected the prevailing pattern and the belief of authorities that such a split would be a least-cost solution. But traffic patterns are complex, and there are important instances of short hauls by rail and of long distance movements by truck. This con- firms the observation made in other countries that distance is not a reliable guide for choosing the low-cost mode, which may instead be determined by such factors as speed, reliability, and terminal handling costs. Finally, WGT'S assumption made little allowance for shifts in modal use arising from technological innovations, such as pipelines, or from improved operations. Another example of a possible change in modal use emerged from an investigation of cement distribution costs. It appears that the introduction of large bulk cement trucks could raise the eco- nomical distance for road hauls, which compete with rail movement plus local delivery, from 75 kilometers to more than 150 kilometers. WGT made traffic projections by routes only for railway and for major coastal movements and road expressways. By assuming continuation of the past modal split, WGT may have made inadequate allowance for diversion from railways to other modes and thereby overstated the de- mand on some railway routes. On the other hand, WGT did not analyze the likely flows on most of the individual road links and therefore did 322 PART THREE: MAJOR SECTORS not provide a basis for identifying potential bottlenecks in the highway system. For its traffic projections, WGT assumed the demand as given. It did not analyze whether the provision of transport facilities would itself influence transport demand or how transport considerations could affect industrial location and the demand for transport services. Nor did it consider the relationship between transport prices and demand. Korea has uniform countrywide prices for several basic commodities-including anthracite, petroleum products, cement, and fertilizer-and uses uniform railway tariffs without taper for distance. Some shifts in traffic patterns and modal splits might occur if government were to alter these pricing and regulatory policies. Although changes would develop slowly and be minor over the medium term of the fourth plan, they would probably be significant for long-term planning. The five-year horizon for the demand projections corresponds to the fourth plan period. Because most investments proposed in the fourth plan will continue in use beyond 1981 and because some investments may be justified solely by post-1981 needs, demand forecasts should ideally have gone well beyond 1981. Nevertheless the fact that projec- tions are medium in term is not likely to cause serious investment errors for such equipment as rail rolling stock, motor vehicles, and ships. If demand rises unexpectedly, more equipment can be produced and in- stalled quickly. If post-1981 demand falls below expectation, adjustments can be made by reducing replacements. A part of the fourth plan's in- vestment program for transport does include long-range projects, such as new ports, roads, and rail lines (mainly industrial sidings), and these should reflect corresponding long-term needs. Investments of this nature represent less than 20 percent of the total. Moreover feasibility studies will be made for many projects and include forecasts corresponding to their economic lives. In summary, WGT made a major effort to measure overall future trans- port demand and provided a useful basis for planning Korea's transport system. Nevertheless further refinements of this work may be desirable. Forecasts should be made for longer time spans of ten years or more and be periodically repeated. Forecasts should also investigate alterna- tive traffic assignments and analyze road traffic potential in greater de- tail. Developing these refinements will take time, however, and existing forecasts will have to be used to review the fourth plan. Although these forecasts provide a general guide to the level of investments in the entire sector, they may have limited use for allocating funds to individual modes. TRANSPORT 323 Pricing and finance The government's principal aims in setting transport prices are to allocate traffic io the appropriate mode and to eliminate deficits of publicly owned transport enterprises, particularly KNR. Traffic allocation is supposed to ensure optimum use of Korea's transport facilities; estab- lishing financial self-sufficiency is a reflection of the policy applied by government throughout the economy. Government has nevertheless had some difficulty putting its policies into practice. In general, price regu- lation appears not to have had a major effect on modal choice and therefore not to have created distortions in the allocation of resources. Cost data are lacking to analyze the effects more precisely. Cost data are also lacking to identify optimum, low-cost modes. But even with much better data, government might still have difficulty specifying the optimum modal allocation without applying enormously complex and finally ineffective regulations. Because government's general policy is to use regulation sparingly, price regulation in transport may need to be redefined in order to accommodate more limited and attainable objectives. Similarly, inadequate accounts and limited cost and marketing data have prevented the execution of an effective policy of financial self- sufficiency. Despite numerous attempts to improve the financial perfor- mance of KNR, it is not financially self-supporting. Financial targets must be based on appropriate accounting systems, which only now are being developed for KNR and KMPA and which should be extended to KHC and the agencies responsible for aviation infrastructure. These procedures will be an essential prerequisite to formulating and coordinating the financial policies of these organizations. In general, government trans- port enterprises need far greater autonomy in their management and accounting if the fourth plan's ambitious targets for the improved finan- cial performance of these enterprises are to be met (see chapter 11 on public finance). Pricing and finance policies probably would be more effective if gov- ernment concentrated on areas in which it has inescapable responsibility and direct control: on the pricing and finance of government-owned enterprises, and on the development and promotion of improved trans- port systems. The fourth plan emphasizes the importance of making government-owned enterprises self-supporting. Pricing is a primary tool for meeting this objective. But effective pricing policy should not rely on across-the-board rate increases. It should be based on careful analysis of individual tariffs, on consideration of trade practices and 324 PART THREE: MAJOR SECTORS demand elasticities, and on comparisons of cost structures of competing modes. Government also exercises considerable influence on the prices of transport inputs, such as equipment and fuels, and thereby on trans- port services themselves. Government could use this influence to foster the development of better transport technologies. For example, govern- ment could facilitate much broader use of larger, more efficient trucks, either by removing the heavy import duties now levied on them or by encouraging Korean manufacturers to enter this product line. If government is to revise its policies for pricing, it will need to expand the collection and analysis of data: for prices actually charged for transport services; for seasonal variations in these prices; for total distribution costs for typical major bulk movements; and for the finance of transport sector enterprises, which would imply conducting another transport census. With this data, policies could then be developed for: * financial objectives of public sector transport enterprises and the establishment of accounting systems to support those objectives. * pricing major bulk commodities and the transport charges for mov- ing these commodities. D pricing railway services, with emphasis on the traditional role of railway charges as the price leader in the transport sector. *pricing port services, with particular attention to the effect of com- petition among ports and the tradeoff between the use of ports and inland transport. * improving efficiency in public sector transport enterprises. *identifying and promoting improved transport technologies. Tco has so far played only a limited role in formulating government's pricing and finance policy in the transport sector. With its limited staff, TCO may not be able to perform all the data gathering and analysis which is needed. It could nevertheless supervise a large program of research by other government bodies and private organizations. It could also formulate government policv and coordinate its execution. Steps should therefore be taken to establish a work program for TCO and to provide TCO with funds to carry it out. Tco should collaborate wvith PB in three major activities: monitoring the technical progress of the transport sec- tor; identifying promising new transport technologies; and designing programs to put them into effect. Efficiency of private transport Although government restricts entry into the road and coastal trans- port industries, the restrictions have not been so onerous as to discourage TRANSPORT 325 growth of these industries or to prevent the infusion of additional cap- ital. Nevertheless the mechanisms to control entry may hinder increased efficiency. In addition, the continued growth of private haulage-that is, own-account haulage-has not been explained. It may be a result of the existing licensing system, and it may imply an inefficient use of re- sources. Government may therefore wish to consider to what extent it should continue to restrict entry into road and coastal transport, whether it should seek structural changes in these industries, and how it can best effect such changes. A thorough study of these issues is recommended. Comprehensive information is not available on the performance of the road and coastal transport industries. Small sample surveys of the financial operations of land transport enterprises by the Bank of Korea (BOK) indicate rapidly rising revenues and profits in 1973 and 1974 after declines in 1972. Similar patterns apply to a sample of ocean and coastal transport companies. Productivitv measured in turnover of assets also rose for both groups of companies throughout this period. No measures of physical performance and service characteristics, such as transit times, are available. Inquiries suggest that shippers were generally satisfied with goods transport services by road. One major policy goal of government is to improve the management of all transport enterprises. This policy is particularly directed to road transport, where more than 2,000 separate enterprises operated about 90,000 vehicles in 1975. Most of these enterprises were quite small and were subjected to indirect management-that is, the enterprise consisted of an association of owner-operators whose activities were loosely co- ordinated by the enterprise management. Only 390 companies, or 18 percent of the total, had assets wholly owned and operated by the com- pany itself. The system of indirect management is, in the opinion of government, responsible for the low efficiencv of road transport enter- prises and for their failure to grow through internally generated funds. Government therefore proposes to eliminate indirect management by introducing tax penalties for enterprises using this form of organization and by requiring vehicles to be transferred to direct company ownership as they are worn out and replaced. Concurrently government wishes to encourage the trend toward heavier trucks and to stimulate the construc- tion of freight terminals. These proposals imply a fundamental change in the structure of the road transport industry. By eliminating the small owner-operator, this change would cut off an important source of capital for the industry and force road transport to rely much more on organized capital markets. Because the sums to be raised are very large, this change may have direct and measurable effects on the fourth plan's targets for capital 326 PART THREE: MAJOR SECTORS mobilization. Moreover the evidence does not demonstrate that obvious inefficiencies in road transport are the result of the prevailing pattern of indirect management or that companies not using this system are notice- ably better than the rest. Before attempting such a major change, government should consider making a thorough survey of the industry. Such a survey should include a transport census and measure the service characteristics and physical performance indicators of various enterprises. A similar survey should be conducted for coastal shipping to establish realistic goals for this industry's role in the transport system and to enable preparation of a feasible plan for its technical improvement. Chapter 11 Marinus van der Mel Public Finance PUBLIC SAVINGS are an important part of the financial resources that determine the rates of investment and economic growth feasible in Korea. From the early 1960s onward, public savings rose sharply and made an increasing contribution to the financing of development. A setback occurred in the first half of the 1970s, particularly during 1972- 74. It was caused by a temporary slowdown in economic growth, weak- nesses in the tax effort, and rising subsidies. Since then there has been a turnaround, reflecting the renewed buoyancy of the economy and demonstrating the responsiveness of govemment and its capacity to gather in incremental income. This chapter examines the principal factors determining public savings and their prospects. It also reveals the continuous and changing role of government in stimulating and guiding economic growth by the use of budgetary policies. Government does this not only by channeling resources for investment through the public sector budget, but also by providing incentives and mitigating undesirable effects, particularly on prices and income distribution, with various forms of tax relief and subsidies. Composition and Size of the Public Sector The public sector comprises central government, local govemments, and public sector enterprises. The predominant share of public sector financial transactions is included in the budget of central government. Not included in it are the own finances of local govemments, a large share of the transactions of government-invested corporations, public sector enterprises at provincial and local level, and extrabudgetary accounts. 327 328 PART THREE: MAJOR SECTORS Central government In 1976 the budget of central government comprised the general account, the economic development special account, and twenty-four other special accounts; the first two accounts constitute what is termed the general budget sector. The general account includes tax revenue, a part of other current revenue, and the current expenditure of govern- ment departments and offices not covered by special accounts. The gen- eral account is also fed by funds received in trust from the foreign loan special account and other special accounts. In turn, it feeds large amounts into the economic development special account and directly transfers lesser sums to other special accounts for current and capital expenditure. The economic development special account provides funds to central government for investment and loans to a number of separate government entities. In order to analyze revenue and expenditure trends and measure the net size of transactions in the public sector, it is necessary to net out gross data for transfers between all these accounts. Unfortunatelv this cannot be consistently done from budget summaries produced by the Economic Planning Board (EPB). Special accounts complicate the struc- ture of the budget; extrabudgetary accounts complicate it further.1 The national accounts prepared by the Bank of Korea (BoK) provide the best consolidated information about public finance (table 11.1). During 1963-73 net current expenditure of general government, which includes local governments and extrabudgetary accounts but excludes govern- ment enterprises and public corporations, was in the range of 12 to 13 percent of GNP. Its share of GNP rose to 15 percent in 1974 and 16 per- cent in 1975, largely because in rising subsidies and defense expenditure. Gross investment in the noncorporate public sector since the mid-1960s has typically ranged between 5 and 6.5 percent of GNP; capital transfers to the private sector usually were about 0.5 percent of GNP. Conse- quently total general government expenditure has typically been less than 20 percent of GNP. It went above this level for the first time in 1974, when it rose to 20.5 percent, and then rose further to 23 percent in 1975. 1. This structure was simplified in the 1977 budget. The economic develop- ment special account and three other special accounts were included in the general account; two special accounts were consolidated into a single account; one special account was closed. PUBLIC PINANCE 329 Local government Local government operates at three levels: provinces; cities and countics; administrative districts under the cities and counties. In addi- tion to the nine provinces, the two special cities of Seoul and Busan have provincial status and are under the direct control of central govern- ment. Seoul is under the direct administrative supervision of the Prime Minister; Busan and the other nine provinces are supervised primarily by the Ministry of Home Affairs. Each province embraces cities and counties. A county obtains city classification when its population reaches 50,000. Local governments have substantial administrative responsibility, particularly for education, but only limited authority for raising revenue and borrowing. They consequently dcpend heavily on transfers, mostly current, from central government. Provinces receive transfers from cen- tral governmenit; counties and cities from provincial governments. Total current revenue of local governments, including transfers from central government, increased from W202 billion in 1971 to W721 billion in 1976.2 Transfers from central government increased commen- surately, providing on average about 60 percent of total current revenue at the disposal of local governments. Morc than one-half of these trans- fers were channeled through the Ministrv of Education in support of the educational activities of local government. Ccntral government also ex- tends current and capital subsidies to provinces for special projects. Other current transfers were allocated in lump sums to the Ministry of Home Affairs, wvhich decides on the distribution of these funds among provinces. Most taxes levied by local governments are property-related, and in recent years these have accounted for about two-thirds of total local tax revenue. Seoul and Busan, even though containing about 30 percent of the population, collect about 60 percent of local taxes. To offset regional imbalances in revenue collection, tax revenues are re- apportioned. Between 60 and 70 percent of local tax revenue accrues to provinces other than Seoul and Busan; the rest to cities and counties. Public enterprises Public sector enterprises are highly concentrated at the central level. Those operating at national level comprise a group of six government enterprises, which are administered as departmental agencies, and a group of thirty-three government-invested corporations, of which gov- ernment holds a majority interest in twenty-four (see table D.22 in 2. EPB, Overall Resources Budget, 1976 and earlier issues. 330 PART THREE: MAJOR SECTORS Table 11.1. Revenue and Expenditure of General Government, 1963-75 Percentage of gross national product Item 1963 1966 1968 1971 Income from property and entrepreneur- ship (net of interest) 1.6 1.9 2.7 1.7 Indirect taxes 6.3 7.0 9.2 9.5 Direct taxes 2.5 3.7 5.1 5.8 Other current transfersb 1.4 1.0 1.2 1.7 Current transfers from rest of the world (net) 5.3 3.2 2.1 0.7 Total current revenue 17.0 16.8 20.3 19.3 Current civil expenditure 7.1 6.3 7.0 7.1 Current defense expenditure 4.1 3.9 4.0 4.2 Current subsidies 0.2 0.0 0.0 0.0 Other current transfers' 1.3 1.1 1.5 1.7 Total current expenditure 12.7 11.3 12.5 13.1 Net savings 4.3 5.6 7.9 6.3 Gross investment in noncorporate public sectord 3.6 4.8 6.7 6.2 Capital transfers to private sector 0.3 0.5 0.5 0.6 Total 3.9 5.3 7.2 6.7 Financed from: Net savings 4.3 5.6 7.9 6.3 Depreciation' 0.7 0.5 0.5 0.4 Capital transfers received' 0.59 0.1 0.1 0.2 Net borrowing -1.6 -0.8 -1.3 -0.2 Gross national product (billions of won) 488.5 1,032.5 1,598.0 3.151.6 Note: General government includes local government and extrabudgetary ac- counts, but excludes government enterprises and public corporations. All figures are in current prices. Figures may not reconcile because of rounding. Sources: Bank of Korea (BoK), National Income in Korea, 1975, table 1.9; BOK, Economic Statistics Yearbook, 1976, table 146. a. Provisional. appendix D). In addition to these enterprises, a large number of other enterprises are indirectly held or controlled without equity. It is difficult to disentanglc the share of government ownership of these enterprises, but the group as a whole is not important in relation to total public finance. There also are forty-six local government enterprises which en- gage in such activities as water supply, hospital administration, and PUBLIC FINANCE 331 Percentage of gross national product 1972 1973 1974 1975' Item Income from property and entrepreneur- 1.8 1.0 1.0 1.1 ship (net of interest) 8.9 8.9 10.2 12.1 Indirect taxes 4.6 4.3 4.8 4.7 Direct taxes 1.5 1.6 1.5 1.5 Other current transferse Current transfers from rest of the world 0.5 0.3 0.4 0.3 (net) 17.3 16.1 18.0 19.6 Total current revenue 7.1 6.2 6.5 6.7 Current civil expenditure 4.3 3.5 4.4 5.3 Current defense expenditure 0.0 0.4 2.7 1.5 Current subsidies 1.9 1.6 1.3 2.4 Other current transfers' 13.3 11.7 15.0 16.0 Total current expenditure 4.0 4.4 3.0 3.6 Net savings Gross investment in noncorporate public 5.7 5.4 4.2 6.5 sectord 0.6 0.6 1.3 0.7 Capital transfers to private sector 6.3 6.0 5.5 7.2 Total Financed from: 4.0 4.4 3.0 3.6 Net savings 0.4 0.4 0.4 0.4 Depreciatione 0.1 0.1 0.1 0.2 Capital transfers received' 1.9 1.1 2.0 2.9 Net borrowing 3,860.0 4,928.7 6,779.1 9,080.3 Gross national product (billions of won) b. From households and private nonprofit institutions. c. To households and private nonprofit institutions. d. Includes increase in stock. e. Includes government enterprises. f. From households, private nonprofit institutions, and corporations. g. Includes net transfers from rest of the world. transport. This group is not important in quantitative terms, either. Their combined budgets in 1976 amounted to less than W50 billion. Financial transactions of government enterprises are fully incorpor- ated in the central government budget and managed through special accounts: monopoly enterprise for tobacco and ginseng sales; communi- cations service; national railroad; grain management; postal life insur- 332 PART THREE: MAJOR SECTORS ance and annuitv; and government supply. Only the first three are im- portant in quantitative terms; the monopoly account generates large surpluses which arc transferred to the general account and form an important part of tax revenue. The laroest government-invested corpora- tions are BOK, Korea Development Bank (KDB), and Korea Electric Company (KECO). Financial transactions of these corporations are re- flccted in the central government budget only to the extent that central government contributes to their current and capital resources and re- ceives from thcm profits and service payments on debt. Some of them cover a substantial part of their financial needs for investment and work- ing capital from sources other than central government. Government- invested corporations have much greater operational independence than govcrnment enterprises. In addition, their profitability tends to increase as the degree of government ownership declines: the nine corporations in which government has only a minority interest are much more profitable than thosc in which government has a majority interest. Gov- ernment's policy is to dilute its majority interest in many corporations, but it is clear that the private sector will invest only as the profitability of these corporations rises. The relativc importance of the various classes of public sector enter- prises emerges from a study analvzing their contribution to value added.5 Gross value added generated in 1972 by public enterprises, including those held indirectly or controlled without equitv, amounted to more than 9 pcrcent of GDP and more than 13 percent of nonagricultural GDP. The public entcrprise sector accounted for seven-eighths of value addcd in finance, twvo-thirds of that in clectricitv, water, and sanitary services, and slightly less than one-third of that in mining, transport, and com- munication. It also accounted for 15 percent of manufacturing and 5 percent of construction. Local government enterprises contributed only 4 pcrcent of total value added of public sector enterprises. Extrabudgetary accounts Some government operations remain entirely outside the budoet and are financed through extrabudgetary accounts. The largest of these are for the Grain Management Fund (GMF) and the Fertilizer Fund (FF). The extrabudgetary account for the Government Supply Fund (GSF) is much smaller. It assumed special significance in 1975, however, when 3. Leroy P. Jones, Public Enterprise and Economic Development: The Korean Case (Seoul: Korea Development Institute, 1975), pp. 44-48, 67-68, 74-76, app. A. PUBLIC FINANCE 333 government took over a part of private inventories as a temporary meas- ure to relieve financial pressure in the private sector. Other extrabud- getary accounts finance the Agricultural Price Stabilization Fund (APSE) which provides loans to the farming community; the National Invest- ment Fund (NIF) which provides loans to the private sector for invest- ment and is rapidly growing in importance; two pension funds; and six other funds having small receipts and disbursements. Some of the funds perform noncommercial functions, but others engage in financial (APSF and NIF) and nonfinancial (GMF, FF, and GsF) enterprise activities. GMF was established in 1970 to provide finance for government food- grain operations. Until 1973 subsidies on foodgrains were relatively small, arising principally from the sale of barley below cost to stimulate its consumption in preference to that of rice. Losses were largely offset by profits from the sale of rice. The financial position of GMF worsened somewhat in 1973 and then suffered a major deterioration in 1974 and 1975. In both years substantial losses resulted from the operations in rice and barlcy and even more from the steep rise in the flour subsidy. By 1976 the virtual elimination of the flour subsidy, made possible by de- clining import prices, reduced the GcIF operating deficit. The extra- budgetary deficits of FF also grew rapidly in 1974 and 1975, principally because prices charged to farmers were not increased sufficientlv to cover rising costs. After the average selling price of fertilizer was raised by 79 percent in December 1975, further deficits in this account were not expected. Past Trends of Public Savings During the earlv 1960s public savings were negative (table 11.2).4 Then there was an almost uninterrupted improvement through 1970, 4. A detailed analysis of past trends of public savings is complicated bv the limitations of government statistics. Data for total public savings are a part of the national accounts, whereas analysis of underlying trends has to be based primarily on budget summaries of EPB. Data from these sources differ in coverage and are not really comparable. It nevertheless is possible to identify major factors, expressed as a ratio to GNP, which wvere responsible for changes in public savings after 1962. Tables 11.2 and 11.3 summarize major relevant items. It should be noted that estimates of public savings exclude counterpart funds, which are government revenues derived from the sale of imported aid goods. Counterpart funds initially were substantial: they reached their highest level of 8.2 percent of GNP in 1962, fell to 4.5 percent in 1965, and tapered off to 0.6 percent in 1970. Counterpart funds financed a part of defense expenditure, but also enabled higher levels of other public current and capital expenditure than would otherwise have been feasible. 334 PART THREE: MAJOR SECTORS Table 11.2. Public Savings and Related Factors, 1962-76 Percentage of gross national product Item 1962 1964 1966 1968 1970 Central government taxes 9.3 6.0 9.2 13.2 14.1 Local government taxes 1.5 1.3 1.6 1.2 1.3 Total taxes 10.8 7.3 10.8 14.4 15.4 Central government nontax revenue 1.9 0.8 1.0 1.2 0.8 Total current revenue' 12.7 8.1 11.8 15.6 16.2 Central government general expenditure5 12.6 4.8 5.8 7.3 7.3 Defense 5.9 3.6 3.9 4.1 4.0 Total 18.5 8.4 9.7 11.4 11.3 Investment and loans' 6.8 2.3 4.0 5.0 5.7 Central govemment total expenditure 25.3 10.7 13.7 16.4 17.0 Public savings -1.4 0.5 2.8 6.3 7.0 Note: Because of statistical problems, it was not possible to provide a complete reconciliation between public savings and the underlying factors. Sources: EPB, Sumrnmary of the Budget, various fiscal years; EPB, Overall Re- sources Budget, various years; EPB, "Major Economic Indicators of the Fourth Five-Year Economic Development Plan, 1977-81" (Seoul, 1976; processed) BOK, Economic Statistics Yearbook, various years; Box, National Income in Korea, vari- ous years. when the ratio of public savings to GNP reached 7 percent. Most of the increase was concentrated in 1965-68, when the ratio moved up from 1.7 percent to 6.3 percent. After 1970 public savings markedly declined, falling to a low point of 3 percent in 1974. The ratio has since moved up again, and the estimate for 1976 is 5.6 percent. The deterioration of public savings during 1971-74 was principally rooted in the weakness of the tax effort, the rising subsidies for foodgrains and fertilizer, and the slowdown of economic growth. The recovery of public savings from 1975 onward reflects the reversal of these three trends. The 1960s Tight control over government current expenditure and the rapid in- crease of domestic revenue after 1964 made possible the spectacular rise of public savings in the 1960s. Because prices rose more than 60 percent between December 1962 and May 1964, absolute priority was given to eliminating the high rate of inflation, even at the cost of a severe tem- PUBLIC FINANCr. 335 Percentage of gross national product 1972 1973 1974 1975 1976 Item 12.3 11.7 13.5 15.3 16.7 Central government taxes 1.2 1.5 1.6 1.7 1.7 Local government taxes 13.5 13.2 15.1 17.1 18.3 Total taxes 0.7 0.6 0.5 0.6 0.7 Central government nontax revenue 14.2 13.8 15.6 17.7 19.0 Total current revenuea Central government general 7.4 5.9 6.2 6.8 7.5 expenditure5 4.5 3.7 4.4 4.9 6.1 Defense 11.9 9.6 10.6 11.7 13.6 Total 6.3 3.6 4.4 5.6 5.3 Investment and loanse 18.2 13.2 15.0 17.3 18.9 Central government total expenditure 3.9 4.6 3.0 5.3 5.6 Public savings a. Excludes counterpart fund receipts, grants, and local government nontax revenue. b. Does not strictly conform to the definition of current expenditure; covers general account and economic development special account only; excludes extra- budgetary expenditure. c. Does not strictly conform to the definition of capital expenditure; covers general account and economic development special account only. porary decline of current expenditure and public investment. Between 1962 and 1964 total government spending was cut from 25 percent of GNP to only 11 percent (table 11.2). The rise in domestic revenue came largely from a higher tax burden imposed by central government; the tax effort of local governments remained minor. Internal taxes, which are exclusive of customs duties and monopoly profits, represent the largest share of central government revenue. Although the new goveTn- ment introduced major tax reforms in 1962 and largely rewrote existing laws, a major breakthrough in the tax effort occurred only in 1965-68, when the internal tax ratio rose from 5.2 percent of GNP to 9.8 percent. The dramatic improvement in tax administration was probably the most important single factor in this breakthrough.5 Additional explana- 5. For a detailed description see Gilbert T. Brown, Korean Pricing Policies and Economic Development in the 1960s (Baltimore: Johns Hopkins University Press, 1973), pp. 63-68. 336 PART THREE: MAJOR SECTORS tory factors were the consequences of GNP growth and tax law changes, but quantifying their effects is difficult. GNP growth was 12.6 percent in 1968 and 15 percent in 1969; it fell to 7.9 percent in 1970. Tax laws were substantially adjusted in 1966. Another reform in 1968 reduced the income tax for lower income groups and small businesses and in- creased it for the higher income groups and large businesses. A number of indirect taxes were also raised. 1970-76 The severe drop in public savings after 1970 was largely concentrated in the years 1972-74 (table 11.2). Because of the statistical complica- tions noted earlier, consistent analysis of all factors underlying the trend of public savings is not possible. What is clear, however, is that the main cause of the deterioration through 1973 was a decline in the tax ratio. Thereafter the tax ratio staged a marked upturn, from a low of 13.2 percent of GNP in 1973 to 17.1 percent in 1975 and an estimated 18.3 percent in 1976. This led to a recoverv of the public savings ratio, except in 1974 when it was overcompensated by a sharp rise in extra- budgetary expenditure. The decline in the tax ratio in 1972 and 1973 was the result of a relative decline in revenue from income and corporation taxes, selective excises, and customs duties. This decline in revenue was, in turn, the result of a slowdown of business activity and increased tax incentives. Indirect taxes and customs duties recovered strongly after 1973, as did monopoly profits after 1974. This recovery reflected the combined impact of inflation, increases in various indirect taxes in 1974 and 1975, and the introduction of special defense surtaxes in August 1975. Direct taxes in aggregate have shown much less buoyancy. In August 1972 generous, although temporary, allowances for depreciation and investment tax credits were extended to enterprises in specified industries. In January 1974 the income tax for wage earners was substantially reduced. In 1975 a global income tax was introduced to replace the taxes previously levied under five different rate schedules. This global svstem further reduced the tax burden for lower income groups. Current civil government expenditure, as recorded in the budget, seems to have been kept well under control. It is characterized in several years by relative contraction and, on balance, little change. To have full information about current nondefense expenditure, however, the deficits of the extrabudgetarv accounts, notably those of GMF and FF, must be added. In 1974 and 1975 the subsidies on foodgrains and fertilizers, PUBLIC PINANCE 337 financed from these funds, represented a substantial negative element in public savings. The GMF deficits in these years respectively were 1.8 percent and 0.2 percent of GNP; those of the FF, 0.5 percent and 0.8 percent. In 1976 the fertilizer deficit was eliminated, and the deficit for foodgrains were expected to be 0.5 percent of GNP.6 Defense expendi- ture, in contrast, has been rising in recent years; the financial link be- tween this increase and the introduction of the defense surcharges is direct. On a trend basis government has been successful in allocating 30 percent or more of the general budget to the investment and loan account which largely represents investment expenditure (table 11.3).7 The combined share of general expenditure, which is roughly current, and defense expenditure was kept within a ceiling of about 65 to 70 percent of the total. In 1976, however, the share of defense expenditure is estimated to have been sharply higher and that of general expenditure correspondingly lower. This shift is similarly rcflccted in their shares of GNP. In the context of total (current plus capital) government spend- ing, social and economic services have accounted for roughly one-half of general budget expenditure, considering that a large share of transfcrs to local authorities is for education and that most of the unallocable item represents spending on transport. Agriculture and transport have been by far the largest claimants on resources on capital account. In 1971 agriculture accounted for 24.3 percent of government capital ex- penditure; transport for 23.4 percent. In 1976 agriculture's share was 21.8 percent, and transport's 19.2 perccnt.8 On capital account, there- fore, a much larger share of total expenditure has been allocated to the economic sectors than the social sectors; the opposite is true on current account. Public Savings under the Fourth Plan A key objective of the fourth plan for domestic resource mobilization is to have the public savings ratio reach 6.1 percent of GNP in 1981 and 6. Calculation of the GMF operational deficit leads to different results depending on the method adopted for the valuation of sales and stocks. An alternative cal- culation for 1975 places the deficit at I percent. On the basis of this calculation, public savings in 1975 were lower by an amount equal to 0.8 percent of GNP. 7. Known exceptions are expenditure for export promotion and a number of . interest subsidies. 8. Figures were derived from data provided by Korean authorities. 338 PART THREE: MAJOR SECTORS Table 11.3. Budget Expenditure of General Government, 1968-76 Item 1968 1969 1971 1972 Amount (billions of won) General expenditure 116.4 153.8 234.6 286.8 National defense 64.7 84.4 134.7 173.9 Investment and loan 80.3 131.7 176.4 239.9 Other 0.7 0.7 0.6 0.5 Total 262.1 370.5 546.3 701.1 Composition (percent) General expenditure 44.4 41.2 42.9 40.9 National defense 24.7 22.5 24.6 24.8 Investment and loan 30.7 36.1 32.4 34.2 Other 0.2 0.2 0.1 0.1 Total 100.0 100.0 100.0 100.0 Percentage of GNP General expenditure 7.3 7.4 7.4 7.4 National defense 4.1 4.1 4.3 4.5 Investment and loan 5.0 6.3 5.6 6.3 Other . ... . .. Total 16.4 17.8 17.3 18.2 Gross national product (billions of won) 1,598.0 2,081.5 3,151.6 3,860.0 Zero or negligible. Note: Investment and loan includes the economic development special account and capital transfers to other special accounts. Other special accounts are not included. Source: Data supplied by Korean authorities. a. Revised budget figures. average 5.9 percent for the period as a whole (table 11.4).9 During the initial stages of preparing the plan these targets seemed ambitious. The prospects for realizing or even exceeding them have since brightened markedly. The buoyancy of the economy in 1976 turned out to be much greater than expected. The estimate of the public savings ratio for 1976, the base year for the plan's projections, initially stood at 3.8 percent of GNP.10 In early 1976 it was raised to 4.4 percent, and later to 5.6 percent, 9. EPB, "Preliminary Outline of the Fourth Five-Year Economic Development Plan" (Seoul, 1976; processed). 10. EPB, Overall Resources Budget for 1976 (Seoul, 1976), pp. 238-39. PUBLIC FINANCf 339 1973 1974 1975 1976' Amount (billions of won) 291.4 421.0 703.9 826.1 General expenditure 183.5 296.8 442.4 738.0 National defense 176.8 301.1 485.7 705.9 Investment and loan . ... .. ... Other 651.6 1,018.9 1,632.0 2,270.0 Total Composition (percent) 44.7 41.4 43.1 36.4 General expenditure 28.2 29.1 27.1 32.5 National defense 27.1 29.5 29.8 31.1 Investment and loan ... ... .. .. Other 100.0 100.0 100.0 100.0 Total Percentage of GNP 5.9 6.2 7.8 6.9 General expenditure 3.7 4.4 4.9 6.1 National defense 3.6 4.4 5.3 5.9 Investment and loan ... ... ... ... Other 13.2 15.0 18.0 18.9 Total Gross national product 4,928.7 6,779.1 9,080.3 12,008.7 (bilions of won) largely because of more abundant tax revenue. The average projected ratio for the fourth plan of 5.9 percent is substantially higher than the 4.5 percent average estimated for the third plan period 1972-76.11 It would equal the level realized during 1967-71, the second plan period in which CNP growth was extremely rapid (table 11.5). During the fourth plan period government intends to raise the tax ratio by almost 3 percent of GNP, from an estimated 18.3 percent in 1976 11. At this level the average ratio for the third plan would be 60 percent of the initial target. 340 PART THREE: MAJOR SECTORS Table 11.4. Government Savings, 1976-81 Item 1976 1977 1978 1979 1980 1981 Government savings (billions of won in current prices) 674.6 818.2 1,046.0 1,226.8 1,472.0 1,741.9 Current surplus 736.6 871.8 1,096.4 1,266.9 1,507.8 1,768.5 Grain account deficit 62.0 53.6 50.4 40.1 35.8 26.6 Percentage of GNP 5.6 5.5 5.8 5.9 6.0 6.1 Marginal saving ratio' (percent) 25.0 25.3 27.9 21.8 24.8 22.9 Percentage of domestic savings 26.7 24.9 25.5 24.5 24.1 23.5 Percentage of total investmene 20.3 20.3 22.2 22.6 23.3 22.5 Sources: EPB, Overall Resources Budget for 1976; EPB, The Fourth Five-Year Economic Developmnent Plan, 1977-81 (Seoul, 1976); data provided bv Korean authorities. a. The increase in government savings as a proportion of the increase in gov- ernment current revenue. b. Private and public. Table 11.5. Government Savings, 1967-81 Second Third Fourth plan plan plan Item 1967-71 1972-76 1977-81 1976 1981 Government savings (billions of won in 1975 prices) -a 1,756.6 4,057.2 586.6 994.8 Percentage of rNP 5.9 4.5b 5.9 5.6 6.1 Percentage of total investment' 22.4 17.3d 22.5 20.3 23.6 Sources: EPB, Overall Resources Budget for 1976; EPB, Fourth Plan; and data supplied bv Korean authorities. a. Not calculated. b. The original target was 7.5 percent. c. Private and public. d. The original target was 30.3 percent. to 21 percent in 1981. Nontax revenue is to rise on balance, and the deficit of the GMF account is taroeted to be brought back gradually to 0.1 percent of GNP. The reason the public savings ratio will not rise faster than indicated, despite an aggressive revenue effort and the reduc- PUBLIC FriNANCE 34I tion of grain subsidies, is that substantial provision has been made for additional current government expenditure. Its share in GNP is to rise over the five years by 2.5 percent and reach 18.5 percent in 1981 (table 11.6). In Korea the realization of public savings depends largely on the efforts of central government. Excluding transfers from central govern- ment, the savings of local governments are heavilv negative. Public sec- tor enterprises are not an important source of public savings, either. Current revenue Introduction of the value-added tax (VAT) is a key element in the anticipated expansion of the total tax effort. As part of the VAT reform, selective excises, the liquor tax, and defense surcharges will be con- tinued. Other elements of the additional tax effort include direct taxes, monopoly profits, and local taxes. The total burden of customs duties is to be reduced from 1978 onward, resulting in its lower ratio to GNP (table 11.7). Altogether, the share of indirect taxes, including customs duties and monopoly profits but excluding local taxes and defense sur- charges which are also levied on direct taxes, would rise from 10.3 percent of GNP in 1976 to 11.5 percent in 1981; the share of direct taxes, from 4.4 percent to 5.1 percent. The result of the planned effort would be slightly greater emphasis on direct taxes, taking into account that local taxes are largely direct. VALUE-ADDED TAX, SPECIAL EXCISES, AND DEFENSE SURCHARGES. The major objectives of VAT are to meet more effectively the rising need for financial resources, to strengthen the incentive schemes for exports and investments, and to maintain the neutrality of indirect taxes. VAT was introduced in July 1977. The proposed rate of 13 percent was reduced to 10 percent to minimize inflationary effects. The law nevertheless stipulates a basic rate of 13 percent and the 10 percent rate is temporary. The system of indirect taxes is complex and, because of variations in coverage and rates, involves substantial price distortions. The most important taxes are the business activity tax, which is a cascade-type tax levied on gross receipts of most businesses, the commodity tax, which is an excise tax on a wvide range of items, and the petroleum products tax. Together these three taxes account for about three-quarters of revenue from indirect taxes other than customs duties and monopoly profits. VAT has essentially replaced eight taxes: the three taxes just mentioned, the textile products tax, the electricity and gas tax, the transportation tax, the admissions tax, and the entertainment and restaurant tax. Exceptions 342 PART THREE: MAJOR SECTORS Table 11.6. Consolidated Current Account of the Government Sector, 1976-81 (billions of won in current prices) Item 1976 1977 1978 Current revenue 2,654.3 3,223.0 4,039.5 Tax revenue 2,198.2 2,731.6 3,431.7 Net surplus of government enterprises 106.3 123.5 159.3 Other nontax revenue 349.8 367.9 448.5 Current expenditure 1,979.7 2,404.8 2,993.5 Defense 738.1 925.4 1,128.4 Other budgetary expenditure 1,179.6 1,425.8 1,814.7 Total budget expenditure 1,917.7 2,351.2 2,943.1 Grain account deficit 62.0 53.6 50.4 Government savings 674.6 818.2 1,046.0 Gross national product 12,008.7 14,932.6 17,891.7 Percentage of GNP Current revenue 22.1 21.6 22.6 Tax revenue 18.3 18.3 19.2 Net surplus of government enterprises 0.9 0.8 0.9 Other nontax revenue 2.9 2.5 2.5 Current expenditure 16.5 16.1 16.7 Defense 6.2 6.2 6.3 Other budgetary expenditure 9.8 9.5 10.1 Total budget expenditure 16.0 15.7 16.4 Grain account deficit 0.5 0.4 0.3 Government savings 5.6 5.5 5.8 Note: The government sector comprises the central and local governments, in- cluding special accounts. The net surplus of government enterprises includes that of the departmental enterprises, but most monopoly profits are transferred to central government and included in tax revenue. Central government enterprises do not are that the commodity tax will be continued for a limited range of items, that part of the textile products tax will be retained, and that the admissions tax will be continued for casinos and golf courses. All these are central taxes. VAT is levied on the value added by each enterprise, including the PUBLIC FINANCE 343 1979 1980 1981 Item 4,869.1 5,856.4 7,036.7 Current revenue 4,134.7 4,974.9 5,970.8 Tax revenue 184.2 208.8 239.1 Net surplus of government enterprises 550.2 672.7 826.8 Other nontax revenue 3,642.3 4,384.4 5,294.8 Current expenditure 1,359.7 1,623.5 1,952.3 Defense 2,242.5 2,725.1 3,315.9 Other budgetary expenditure 3,602.2 4,348.6 5,268.2 Total budget expenditure 40.1 35.8 26.6 Grain account deficit 1,226.8 1,472.0 1,741.9 Government savings 20,880.4 24,351.2 28,391.2 Gross national product Percentage of GNP 23.3 24.1 24.8 Current revenue 19.8 20.4 21.0 Tax revenue 0.9 0.9 0.9 Net surplus of govemrnment enterprises 2.6 2.8 2.9 Other nontax revenue 17.4 18.0 18.6 Current expenditure 6.5 6.7 6.9 Defense 10.7 11.2 11.6 Other budgetary expenditure 17.2 17.9 18.5 Total budget expenditure 0.2 0.1 0.1 Grain account deficit 5.9 6.0 6.1 Government savings include the thirty-three government-invested corporations, whose modest profits are included in other nontax revenue. Sources: EPB, Overall Resources Budget for 1976; EPB, Fourth Plan; data sup- plied by Korean authorities. retail and import stages. Some necessities and a number of other items, which are considered undesirable objects of tax for administrative or welfare reasons, are exempt. These items include unprocessed food, water, briquettes for home heating, medical services, education, books, and finance. A zero rate is applied to exports and other specific foreign 344 PART THREE: MAJOR SECTORS Table 11.7. Total Tax Revenue, 1976-81 Percentage of gross national product Item 1976- 1977 1978 1979 1980 1981 Internal taxes 11.2 11.0 12.1 12.5 12.8 13.2 Direct taxes 4.4 4.1 4.3 4.6 4.8 5.1 Indirect taxes 6.4 6.7 7.6 7.7 7.8 7.9 Value-added tax . . .. 4.9 4.9 4.9 4.9 Excise taxes ... 1.8 1.9 2.0 2.1 Other 0.4 0.2 0.2 0.2 0.2 0.2 Customs duties 2.0 2.2 1.6 1.6 1.6 1.6 Monopoly profits 1.5 1.5 1.5 1.6 1.7 1.8 Local taxes 1.7 1.7 1.9 2.0 2.1 2.2 Defense surtaxes 1.9 1.9 2.1 2.1 2.2 2.2 Total tax revenue 18.3 18.3 19.2 19.8 20.4 21.0 Gross national product (bil- lions of won) 12,008.7 14,932.6 17,891.7 20,880.4 24,351.2 28,391.2 Zero or negligible. Note: All figures are in current prices. Sources: EaP, Fourth Plan; data supplied bv Korean authorities. a. Revised estimates. exchange items.12 The tax by its nature implies that each taxpayer can deduct VAT paid on inputs, including imports and investment items, from VAT payable on output or sales. It is expected that VAT Will increase the revenue potential by broaden- ing the tax base and making evasion more difficult. As a safeguard against initial revenue shortfall, government continues to levy special excises on about thirty items. These items are mostly of a luxury nature, but one important item is sugar. As follows from table 11.7, excises in aggregate continue to be an important source of revenue. In 1978 their yield would equal more than one-third of that from VAT; in 1981 more than 40 percent. Three other indirect taxes to be continued are the 12. A zero rate differs from a direct exemption because VAT is levied on inputs which enter into the production of a commodity. This tax on inputs is subse- quently refunded under a drawback system intended to prevent diversion of inputs exempted for export to inputs used in production for the domestic market. PUBLIC FINANCE 345 liquor tax, the tclephone tax, and stamp duties. The first of these is most important in quantitative terms. In addition, defense surcharges are to be continued for items on which thcy are now levied. Many commodi- ties are exempted from the defense surcharge because it does not apply to transactions under the business activity tax. OTHER CENTRAL GOVERNMENT INDIRECT TAXES. Reform of customs duties is not planned to contribute substantially to a stronger tax effort. It is estimated that the ratio to GNP of revenue from this source will decline from 2 percent in 1976 to 1.6 percent in 1981. This decline will be the result of the projected reduction of the average effective tax rate, excluding raw materials for export production, from 9.7 percent in 1977 to 7 percent from 1978 onward. This should be seen as a broad objec- tive. Detailed fiscal planning in this area beyond the immediate future is particularly difficult because of the active use of tariffs as an instru- ment of general economic policy, cspecially for protecting domestic in- dustries, promoting investment and exports, and pricing. In 1976 about 90 percent of the valuc of imports, other than raw materials, of fourteen key industries was exempted from duty. Firm decisions have not yet been taken about changes in exemption policy, but the scheduled lower- ing of the average effective tax rate will probably be accomplished mainly by overall, not selective, rate reductions. Monopoly profits are derived almost entirely from sales of tobacco. Government intends to raise the ratio of monopoly profits to CNP from 1.5 percent in 1976 to 1.8 percent in 1981, principally by keeping the rate of increase in tobacco profits in line with the nominal rate of growth in GNP and by moving the purchase price of leaf tobacco up in accord with the rise of the general price level. Monopoly profits are projected to increase from W384 billion in 1976 to W872 billion in 1981; profits from sales of tobacco from W371 billion to W844 billion. The surplus available for transfer to general account is expected to be W512 billion in 1981. These objectives appear realistic. DIRECT TAXES. In January 1974 government introduced an emergency economic program. Many provisions of income tax legislation were sus- pended and replaced by ad hoc provisions for 1974 only. The objective was to reduce the increase that would otherwise have occurred in the income tax burden because of the rapid rise in prices and nominal in- comes. Until this time the income tax system had been largely schedular. Tax was levied separately on five categories of income: wages and salaries, business income, rents, interest and dividends, and miscel- 346 PART THREE: MAJOR SECTORS laneous income. The rate structure was progressive for the wage, busi- ness, and rental schedules and proportional for the other two schedules. In addition, a global tax was levied at progressive rates on high-income earners. A comprehensive reform of the personal income tax system took effect in 1975. The new tax system is almost entirely global. In contrast with the six bases and four rate structures of the schedular system, the global system has, with few exceptions, a single base and rate structure. It thus eliminated disparities in the tax burden for different types of income and reduced distortions in the use of productive factors resulting from in- equalities in taxes. The new system extends personal deductions to dependents of taxpayers and provides for a large increase in maximum deductions, a reduction in tax rates over most ranges of taxable income, and an increase in tax credits for savings. Other important provisions are the introduction of a generous exemption for bonuses and the taxa- tion of capital gains. It also incorporates manv features of the schedular taxes, including expenses allowable under the business and rental income taxes and exemptions and tax credits available under the business and wage income taxes. The rate structure of the new svstem is less progres- sive for most incomes than the old system. Marginal tax rates for sixteen income brackets range from 8 to 70 percent; average tax rates from 8 to 55 percent. The highest tax rate applies to taxable incomes of more than W48 million ($99,000); the lowest to taxable incomes of less than W240,000 ($494). In addition, defense surcharges have been payable since August 1975; these increase the amount of tax due by 10 percent and, for incomes of more than W84 million, by 20 percent. Because of the numerous deductions and exemptions allowed, these rates give a misleading impression of the actual income tax burden. Government introduced numerous detailed changcs in January 1977, including further income tax relief for low- and middle-income wage earners, certain concessions for taxes on inheritance and land sales, and more extensive relief to corporations. With the exception of changes introduced for nonfiscal reasons, such as reducing the number of child exemptions from three to two and discouraging the setting up of new industrics in urban areas, the effect of these measures is to reduce revenue. The main intentions are to provide tax relief as a correction for inflation and, more generally, to stimulate incentives. Detailed changes have also been made in the corporation tax; some of these will stimulate corporate savings by increasing the incentives to retain earn- ings and charge more depreciation. Even with the large number of exemptions and deductions which characterize the direct tax system, PUBLIC FINANCE 347 direct taxes are to contribute to the additional revenue effort during the plan period. The ratio of direct taxes to GNP is to rise from 4.4 percent in 1976 to 5.1 percent in 1981. The basic structure of direct taxation is to remain unmodified during the plan period. One distinctive feature of corporate taxation is that it clearly favors open corporations over closed corporations. The opening up of corpora- tions by enlarging the number of shareholders and listing on the stock exchange is considered desirable for fostering economic development. For closed corporations income up to W3 million is charged at a rate of 20 percent, that between W3 and W5 million at 30 percent, and that in excess of W5 million at 40 percent. For open corporations income up to W5 million is charged at a rate of 20 percent and that in excess of W5 million at 27 percent. Defense surcharges increase the tax pay- able by 20 percent and, for taxable corporate income in excess of W500 million, by 25 percent. Other special provisions for open corpora- tions are more favorable tax treatment of retained earnings and some types of interest and dividend income, as well as automatic listing as a green-return corporation, which opens the door to a number of addi- tional tax favors. Direct taxes at central government level bevond personal income and corporation taxes are less important. The major revenue earner is the registration tax, which covers a wide range of assets. Jurisdiction over this tax was transferred to local governments in 1977 as compensation for the loss associated with the abolition of the entertainment and res- taurant tax. Some direct taxes, principally the property tax, are levied at local level. The incidence of the property tax is to increase substan- tially as part of the effort to upgrade local taxation. To improve revenue collection and equity, government also has initiated a program to improve the administration for direct taxes. It has a special enforcement plan, which is being gradually improved and ex- tended. Under this plan, special and very thorough audits are being intensified on a selective basis. It is being assisted by the computer center, with the expectation that VAT will provide much information for crosschecking the income of firms against their turnover. It also is expanding its special training program for young staff joining the tax administration. LOCAL TAXES. Local taxes in 1976 were about one-tenth of total tax revenue including monopoly profits. This share is to rise slightly during the plan period; its ratio to GNP is to increase from 1.7 percent in 1976 to 2.2 percent in 1981. This additional effort involves a number of taxes. 348 PART THREE: MAJOR SECTORS The registration tax, until recently levied by central government on a wide range of assets and turned over to local governments in 1977, is tentatively planned to go up threefold. A similar increase has been tar- geted for the property tax. Other relatively important local revenue earners are the acquisition tax (a property transfer tax), the inhabitant tax (a combination of a tax per capita and a levv based on taxable in- come), the farmland tax, and automobile and license taxes. As a result, the already preponderant role of direct taxes at local level will become evcn more pronounced. In 1976 direct local taxes constituted 83 percent of total local taxes of W199 billion; in 1981 they are to constitute 96 percent of total local taxes of W623 billion. Property taxation is being increased primarily by extending the tax base to farm and forcst lands and to certain industrial assets. The inclusion of farm and forest lands is feasible because of the existence of good cadastral surveys, which are also kept up to date. The inclusion of these lands seems justified because, with incomes rising rapidlv, the ability to pav of many rural land-owners has greatly increased. The tax on farmland, called the agricultural land tax, is based on income from this land, but only in principle. Elasticitv has been limited because of generous exemptions and the use for rice land of standard, not actual, incomes. In the past, industrial assets other than land were exempted from property tax for income reasons; this is now, being changed. The property tax also leaves scope for incrcased rates and improved adminis- tration. The possibilities of additional local taxation are bcing actively studied by a committee under the Ministry of Home Affairs. Another committee under the same ministrv is working out proposals for implementing a further basic objective announced in the plan: reform of the finance system of local governments. This committee is to make proposals for correctino present regional disparities in local financial resources in rela- tion to needs; these disparities have persisted despite the large subsidies granted by central government to provinces. NONTAX REVENUE. The ratio to GNP of nontax revenue, excluding the surpluses of govemrnment enterprises, is expected to be about 2.9 percent in 1981. This level is about the same as that recorded in 1976, when nontax revenue was abnormally high. The 2.5 percent ratio for 1975 was more typical and, on a trend basis, the ratio will increase over the plan period. Government has planned for substantial increases in salaries and wages in the public sector. These increases will raise contributions to pension funds which, in turn, will contribute to the targeted total PUBLIC FINANCE 349 growth of nontax revenue. School fees are expected to provide additional resources. At local level gross nontax revenue is to rise by more than 150 percent. Because the efforts to achieve this target must be concerted, the Ministry of Home Affairs established a special division to formulate proposals and scrutinize various user charges, such as those for water and transport. The modest net surpluses of government enterprises are expected to rise from W106 billion in 1976 to W239 billion in 1981. This increase is just sufficient to keep the ratio of these surpluses constant at 0.9 percent of GNP. Communications, bolstered by the profitability of tele- phone and telegraph services, is to provide W93 billion of the planned increase of W133 billion. The contribution of the railways, which in recent years bave incurred operating deficits, will remain small and rise from W9 billion in 1976 to W17 billion in 1981. The thirty-three government-invested corporations, whose budgetary contributions are included in other nontax revenue, are not a major source of revenue. In 1976 they provided only an estimated W3.8 billion, of which a mere W500 million came from the twenty-four corporations in which govern- ment holds a majority interest. During the fourth plan period the con- tribution of these corporations to government revenue is expected to remain small. The railways, communications, and iEco all have to finance large investments programs, to which only insubstantial contributions can be made from operating surpluses. Moreover a larger share of their revenue will be needed to service their growing debt obligations. To meet their financial requirements, they will have to raise prices and rates over the next several years. Government, in its planning, faces the difficulty that large increases conflict with basic objectives of limiting the rate of domestic price increase during the plan period. Current expenditure The plan includes a substantial allowance for raising current govern- ment expenditure from 16.5 percent of GNP in 1976 to 18.7 percent in 1981 (table 11.8). Because the deficit of GMF is to fall from 0.5 percent of GNP to 0.1 percent, the margin for raising all other items will be correspondingly larger.13 13. Detailed analysis of current expenditure trends, both past and future, is greatly hampered by the lack of suitable statistics. 350 PART THRERE: MAJOR SECTORS Table 11.8. Current Expenditure of Government, 1976-81 Percentage of gross national product Item 1976 1977 1978 1979 1980 1981 Total expenditure 16.5 16.1 16.7 17.4 18.0 18.7 General expenditure 6.7 6.5 7.0 7.3 7.5 7.7 Defense 6.1 6.2 6.3 6.5 6.7 6.9 Education expenditure 3.1 3.1 3.2 3.5 3.7 4.0 Grain account deficit 0.5 0.4 0.3 0.2 0.1 0.1 Gross national product (billions of won) 12,009 14,933 17,892 20,880 24,351 28,391 Note: All figures are in current prices. Sources: EPB, Fourth Plan; data supplied by Korean authorities. In building up the expenditure projections, planners looked separately at defense, education, and GMF. The defense projections have two com- ponents, one reflecting the normal effort, the other the special effort. From the financial point of view, the special effort is related to receipts from defense surcharges put into effect in August 1975. Education ex- penditure reflects general objectives to expand enrollment at all educa- tional levels above primary school and to improve the quality of voca- tional and scientific training. Part of the expansion of educational facilities will be privately financed, as in the past, but the public sector will finance most of the expansion. The rise in spending will also be the result of efforts to relieve the burden of educational outlays on house- holds by gradually providing free compulsory education and reducing the cost of middle school education. Government regards the high burden of educational outlays as an impediment to household savings. Thus the rise in government spending on education would to some extent support the objective to increase household savings. Beyond specific projections for defense, education, and GMr, planneTs have built two general assumptions into their projections of current government expenditure: one about wages and salaries, the other about the growth rate of other general expenditure. For government wages and salaries, planners have projected a sharp rise, which accounts for most of the targeted increase of general expenditure. The objective is to raise government wages and salaries gradually to the levels paid in government- invested corporations. These corporations, it must be remembered, are PUBLIC FINANCE 35I distinct from government enterprises, which have salary scales generally in line with those of civil servants. Parity between salaries of civil servants and employees of government corporations is to be reached by 1981. To do this, civil service salaries will increase by 25 percent a year (30 percent in 1977) for a compound increase of 217 percent during the plan period. Salaries in government-invested corporations are expected to increase by 10 percent or less a year and therefore generate a com- pound increase of 61 percent or less. The growth rate of other general expenditure has been targeted at about 3 percent a year in real terms. The planned increases in government salaries follow steep raises in 1974-76. It is difficult to express these increases in percentage terms because salaries of public servants consist of a number of components and the total effect of increases has not been uniform. This diFfculty notwithstanding, the raises for a middle level civil servant, sav a grade 3B section chief, came to about 44 percent in 1974, 28 percent in 1975, and 42 percent in 1976. Despite these raises, the average salaries of civil servants in 1976 were no more than about one-half of those in corre- sponding ranks in govemment-invested corporations. Total salaries and pensions in central government's general budget sector rose even faster, from W141 billion in 1974 to W202 billion in 1975 and W325 billion in 1976. These raises must be viewed against the background of infla- tion. The increase in the consumer price index for urban households was 24 percent in 1974 and 25 percent in 1975. The rate of inflation in the Overall Resources Budget was estimated at 12 percent in 1976. The operating deficit for GMF in 1976 was modest. The major reason for the deficit's decline was the elimination of the subsidy on flour, which the fall in the world price of wheat made possible. In 1974 and 1975 the flour subsidy alone was about W55 billion. The total operating deficit for rice and barley in 1976, when increases in domestic buying prices were larger than had been initially budgetcd, was about W50 billion. During the fourth plan period, government intends regularly to raise the domestic procurement and selling prices of rice and gradually to eliminate deficits. A two-tier selling price for rice is to be maintained, one for rice mixed with barley, the other for unmixed rice. Barley is to be sold at about half the price of rice, and this will require an operat- ing subsidy of W25 to W30 billion annually. Cash deficits of GMF will be larger because of the need to finance an increasing value of stocks. Calculations of the operating deficit in the plan are based on the first-in-first-out principle of historical cost. With rising prices assumed, this gives more favorable operating results than the alternative use of the lastin-first-out principle, which more closely approximates the valuation 352 PART THREE: MAJOR SECTORS of sales at replacement cost. The plan also assumes that the Fertilizer Fund will show no further deficits. Extrabudgetary deficits of GMF and FF during the last several years have been the most important subsidies burdening government finance. Current government subsidies are also provided to the railways, the coal sector, financial institutions, and a number of health and other social activities in the private sector. These subsidies, combined, are not of major importance to total public finance. The size of government subsidies, after the sharp rise in 1974 and 1975, has recently been reduced to more acceptable proportions. Never- theless both the operating and cash deficits of GMF need further reduc- tion if domestic finance targets of the fourth plan are to be met. A second area of concern is the prospective financial position of some public sector enterprises. The most immediate problem is that financing large addi- tional investments will also affect their operating results.14 Capital expenditure Government plans a relatively modest rise in expenditure on capital account. The ratio of investment and loan expenditure to CNP is to fall from 9.1 percent in 1976 to 8.4 percent in 1981. This performance would be in accord with the general strategy of limiting annual invest- ment growth. The ratio of total investment to GNP is to decline from 27.7 percent in 1976 to 26 percent in 1981. One result is that borrowing by government from BOK would be minor, corresponding to 1 percent or less of total capital expenditure (table 11.9). Together with proceeds from the issue of grain bonds, government borrowing is to correspond roughly to the amounts required to finance the increase in value of the foodgrain stockpile, which must be financed while awaiting recovery of sales proceeds."5 For central government three sectors stand out in importance: agri- culture, transport, and communications. The relatively large allocations 14. Interest subsidies, which in 1975 were equal to about 0.6 percent of GNP, are also significant, but seem on the whole justified as part of government's incen- tive policies. Other direct government subsidies to the private sector are minor and for the most part equally justified. 15. Amounts shown for bond issues in table 11.9 are gross-that is, they are not net of amortization, which is included in expenditure. Amortization of bonds in 1976 was W20 billion. Amortization is intended roughly to match new issues of bonds during 1977-81. PUBLIC FINANCE 353 Table 11.9. Government Expenditure on Capital Account and Its Financing, 1976-81 (billions of won in current prices) Item 1976 1977 1978 1979 1980 1981 Expenditure 1,279.2 1,529.1 1,806.4 2,032.3 2,310.8 2,622.3 Investment and loans 1,088.5 1,257.2 1,539.7 1,758.0 2,054.3 2,363.5 Other' 190.7 271.9 266.7 274.3 256.5 258.8 Revenue 818.7 982.2 1,240.8 1,458.9 1,735.9 2,032.6 Savings on current account 674.6 818.2 1,046.0 1,226.8 1,472.0 1,741.9 Other capital receipts 144.1 164.0 194.8 232.1 263.9 290.7 Deficit 460.5 546.9 565.6 573.4 574.9 589.7 Financed from: Net foreign borrowing 266.5 291.9 325.6 348.4 384.9 419.7 Grain bonds 100.0 105.0 90.0 75.0 50.0 50.0 Net borrowing from Bank of Korea 94.0 150.0 150.0 150.0 140.0 120.0 Source: FPB, Fourth Plan. a. Includes changes in the value of grain stocks. b. See table 11.3. to agriculture and transport conform to the broad sectoral pattern of the third plan; the share allocated to communications is higher. For local governments the major allocations are for the nondirectly productive sectors. Table 11.9 includes only investment funds that will be chan- neled directly through central and local governments. In addition, a number of public sector enterprises, such as KDB and KiCO, are expected, as in the past, to attract domestic and foreign savings directly. Some funds included under government investment and loans are lent to the private sector. For these reasons it is not possible to derive total public sector investment readily from budget documents or from forecasts of public finance in the fourth plan. Financial transactions between govemment and other sectors of the economy can be measured only by systematic flow-of-funds analysis. As is pointed out in the analysis of historical flow-of-funds data in 354 PART THREE: MAJOR SECTORS chapter 12, total financial transactions of government, excluding public sector corporations, were in surplus through 1970. The surplus in that year declined to about 1 percent of GNP.16 From 1971 to 1975 govern- ment was a deficit sector. The deficit ranged from I to 3 percent of GNP and was highest in 1974 and 1975 because of the losses of GMF and FF. The flow-of-funds projections for 1977-81 imply financial deficits in the government sector of less than 1 percent of GNP. Despite the deficits in the 1970s it may be concluded that, contrary to the experience of many developing countries, Korea's government sector has not been a large-scale net borrower of funds, if the activities of corporate enterprises are excluded. Until 1973 government was even a net supplier of funds to the financial system. In 1974-76 the deficits of GMF and FF gave rise to net government borrowing from the banking system, but' even then the amounts were neither excessive nor unmanageable. Major Issues of Fiscal Policy The main conclusion from the foregoing analysis of current revenue and expenditure trends is that the postulated public savings targets are realistic-given determined policy action on the part of government and barring unforeseen major setbacks to the economy from external causes. The results of independent macroeconomic modeling work, described in appendix A, even seem to indicate, on the basis of observed structural relations and plausible assumptions included in a number of variants of the model, that it might be possible to exceed the public savings targets. It is appropriate to bear in mind that private savings, particularly house- hold savings, are likely to fall short of projected targets during the fourth plan period. Higher-than-targeted public savings, if they materialize, could offset some of this shortfall. For revenue much depends on the efficient introduction and imple- mentation of VAT and related taxes. The extensive work preparatory to its introduction augurs well for the success of this far-reaching reform. The prospects for customs duties as a source of revenue, given the con- siderable weight of nonfiscal considerations, are difficult to assess. At 16. The flow-of-funds accounts define government as including general govern- ment and government enterprises, which subsume monopoly profits, grain management, railways, and communications, but not the postal and supply special accounts. Box, Economic Statistics Yearbook, 1976 (Seoul, 1976), p. 319. PUBLIC FINANCE 355 local government level, little specific action is in evidence for stepping up the revenue-raising effort, but the subject is under active study. Direct taxation has been projected to absorb a gradually rising share of income, but the total impact of exemptions and deductions remains large. This pattern probably is unavoidable, given the importance of fiscal incentives in overall development strategy. Nevertheless the price and equity effects of all parts of the proposed revenue package should be monitored. One way to test the realism of the tax targets is to examine their implied buoyancy. In aggregate the target for taxation implies a tax buoy- ancy ratio for the five-year period of 1.26 (table 11.10). For VAT, the most important single tax, the buoyancy ratio for 1979-81 would be 1.02. For direct taxes, excises, monopoly profits, and local taxes, the implied buoyancy would be higher. These buoyancy ratios seem feas- ible when compared with past performance. In fact, tax buoyancy during the 1965-68 period was considerably higher. But in these years the rise in revenue was exceptionally strong. This rise started from a low base in 1963-64 and was made possible bv the dramatic improvement in tax administration. Neither condition is applicable to the future. The Table 11.10. Tax Buoyancy Ratios, 1976-81 1977- Item 1976 1977 1978 1979 1980 1981 81 Internal taxes 1.02 0.88 1.61 1.24 1.22 1.20 1.31 Direct taxes 1.21 0.61 1.33 1.52 1.39 1.39 1.29 Indirect taxes 1.09 1.20 1.81 1.10 1.14 1.10 1.41 Value-added tax - - - 1.01 1.05 .0.99 1.021 Excise taxes - - - 1.39 1.39 1.42 1.49' Other -0.66 -1.07 0.61 0.61 0.63 0.51 0.07 Customs duties 1.00 1.69 0.42 0.96 0.97 0.94 0.62 Monopoly profits 0.97 0.97 1.24 1.33 1.43 1.40 1.37 Local taxes 0.78 1.89 1.66 1.34 1.34 1.35 1.56 Defense surtaxes n.a. 0.80 1.72 1.17 1.17 1.15 1.26 Total tax revenue 1.29 1.00 1.29 1.23 1.22 1.20 1.26 - Not applicable. n.a. Not available. Note: Tax buoyancy is the percentage change in revenue expressed as a ratio to the percentage change in GNP. See table 11.7. Sources: EPB, Fourth Plan; data supplied by Korean authorities. a. Total for 1978-81 only. 356 PART THREE: MAJOR SECTORS buoyancy ratios reached in 1974 and 1975 were also substantially higher than 1.26. Key factors in those years were the introduction of defense surcharges and the strong performance of the business tax and monopoly profits. These factors cannot be counted on in the years to come, either. Insofar as the income tax has been restructured and many indirect taxes have been replaced by VAT, the past has become an even more imperfect indicator of future performance of both direct and indirect taxes. Reali- zation of the taxation targets will therefore depend primarily on achiev- ing the targets for economic growth and, within this context, on the timely implementation and efficient administration of the various meas- ures proposed. With respect to government current expenditure, some projected in- creases appear liberal. Moderation of the targeted increases in public sector wages and salaries may be advisable for reasons of general wage policy. Moreover across-the-board wage increases may not be the best way to improve the remuneration of employees in the public sector. The salary structure in the public sector is complex and needs to be ration- alized. To this end the establishment of an independent pav commission deserves consideration. The projected increase in educational expendi- ture, particularly for vocational and technical training, may leave scope for reduction. In the sensitive area of foodgrain pricing, the needed con- sensus appears to be lacking. The temptation to use subsidies as a cushion, if prices rise more than projected, offers no real solution and should be resisted. Effects of tax reforms on revenue, prices, and equity Indirect taxes, not direct taxes, are to make the largest contribution to additional current revenue. This policy will have some regressive effects, even though special efforts are being made to minimize the burden on low-income earners.17 In the context of a system in which heavy reliance is placed on financial incentives to stimulate economic growth in the private sector, some regressiveness of the tax system is difficult to avoid. The past incentive system also relied heavily on tax incentives, and it succeeded in stimulating investment, exports, and economic growth. This process benefited many low income earners, not so much by reducing their tax burden as by creating additional employ- 17. The formulation of future tax policy would benefit from fiscal incidence studies for selected income groups; such studies do not now exist. PUBLIC FINANCE 357 ment and income on a large scale. Altogether the reforms of the direct and indirect tax system make it necessary to minimize undesirable fallout effects for income distribution, prices, and household savings. Some tradeoffs between conflicting policy objectives are inevitable. GLOBAL PERSONAL INCOME TAX. Under the schedular system wage earners were entitled only to a basic deduction of W180,000; tax credits were offered for contributions to housing funds; rates for the five sep- arate schedules and for the global tax were separate. Under the new system wage earners became entitled to a basic deduction of W180,000, a deduction for a spouse of W120,000, deductions for family dependents of W60,000 to a maximum of three (two after 1977), and an earned income deduction of W180,000. Tax credits are now offered for con- tributions to housing funds, but to a higher limit, for savings deposited with national savings associations, and for the purchase of corporation stocks by employees. As mentioned earlier, the rate structure is progres- sive, with rates rising from 8 percent on the first W240,000 of taxable income to 70 percent on taxable income of more than W48 million; corresponding average rates range from 8 percent to 55 percent. Under the new system almost all income is subject only to global tax. But contrary to the basic principle of a global income tax, special tax rates exist for certain types of income not aggregated in total income. Interest income from banks is taxed at only 5 percent, not as part of total income.18 Certain types of dividend income are subject to with- holding tax rates ranging up from only 5 percent, whereas the general withholding rate for dividends is 25 percent. Capital gains are taxed at a flat rate of 50 percent for gains from land and 30 percent for gains from buildings and other assets. These special rates have been devised to foster savings and the development of the capital market and thus are justified. To be avoided, however, is broadening the application of special rates. This would undermine the basic objective of the reform: taxing total income at a unified rate. The introduction of the new income tax was initially expected to result in a reduction of revenue from this source, mainly because of an increase in personal deductions. In fact, revenue in 1975 was about W200 billion, compared with W128 billion initially budgeted and 18. Interest from a substantial number of bonds-government, government- guaranteed, and certain corporate issues-is not taxable under the new system, just as it was not taxable under the old system. 358 PART THREE: MAJOR SECTORS W165 billion collected in 1974. The large surplus in 1975 occurred mainly because increases in nominal incomes, including of those in the public sector, were much higher than expected. Similarly the initial estimate of revenue of W217 billion in the 1976 budget turned out to be low by about W90 billion. Wages again rose more than anticipated. In addition, collections from taxpayers having sources of income not subject to withholding substantially exceeded expectations. The respon- siveness of the tax to price and income changes appears to have im- proved, and this may in part be the result of better tax administration. Many low- and middle-income earners have benefited from the larger deductions and exemptions. But the tax is much less progressive than it appears at first glance because numerous additional exemptions and de- ductions are available to higher income eamers, particularly businessmen and property owners. In addition, the new rate structure for wage, busi- ness, and rental incomes is less progressive than that under the schedular system. Consequently the new tax structure for most incomes is less progressive than was the total impact of schedular and global taxes under the old system. VALUE-ADDED TAX. The introduction of VAT, following the introduc-- tion of the global income tax which itself contributed to simplifying and modernizing the fiscal system, was a major step forward in creating a more efficient tax system. VAr will be instrumental in broadening the indirect tax base and improving the responsiveness of revenue efforts to the future growth of the economy. Because of the complexity and scope of the reform of indirect tax, its effects on revenue, prices, and equity required close attention, as did the problems of transition and administration. Government has studied all of these extensively. The transition to VAT resulted in changes in relative prices because of the many indirect taxes and rates which VAT replaced. Calculations of likely price effects indicated that many price changes would be relatively small. Nevertheless some upward pressure on prices was inevitable despite the reduction of the VAT rate to 10 percent. The total tax burden of the business tax, taking into account its cascade effects, was probably close to 5 percent. But for many commodities the indirect tax burden was estimated to be about tvvice as large, if other, varying taxes on these commodities were taken into account. During preparations for the introduction of this reform, attention was paid to the likely impact on expenditure patterns of different household groups. Little change was expected in the distribution of disposable incomes as a result of VAT. One merit of VAT is that it eliminates distortions in the price structure PUBLIC FINANCE 359 caused by the indirect taxes it replaces. Because of the comprehensive nature of the reform, this objective is being substantially achieved. But the continuation of special excises and defense surcharges implies that some distortions in the price structure will persist. Effective administration is especially crucial to the success of VAT. It requires an adequate audit system, which will be supported by a computer to crosscheck invoices and by field audits. Crosschecking in- voices by computer is already being successfully done for the business tax, but the number of invoices will increase greatly under VAT, espe- cially because of the intended reduction of the exemption limit. Further- more price control will be needed to prevent unjustified increases and to force businesses to reduce prices when justified. Data collected on the indirect-tax content of a substantial number of commodities will facili- tate this task. The relatively high VAT rate of 10 percent involves some danger of evasion, which improved administration may keep within minor proportions. Containing current expenditure A major issue arising from the plan's targets for current government expenditure concerns public sector salaries. The general intention of upgrading these salaries appears sound, but the rate of salary increases in the public sector can easily become a pacesetter for similar raises in the private sector which may exceed productivity gains and be harmful to export competitiveness. A word of caution also seems in order about undifferentiated increases. The pay structure is complex and includes differences in pay among various parts of the government. For example, police and military personnel are relatively well paid, but considerable differences exist in remuneration among govemment enterprises. Because the entire pay structure apparently has not been systematically studied for many years, it seems timely to appoint an independent pay commis- sion to undertake such a study. Meanwhile it would be advisable to reserve some part of available budgetary funds for salary increases that correct anomalies, but not for undifferentiated, across-the-board, per- centage increases. Proposed increments in education expenditure also appear to be high. There probably is scope for economies. For example, more efficiency seems possible in the field of technical and vocational education. The proposal to relieve the burden of educational spending on households by having government shoulder a larger share of this expense also de- serves more scrutiny. First, it is not certain that increasing government's 360 PART THREE: MAJOR SECTORS share would increase household savings at an equal pace. Second, even if this were certain, the question may be asked: What is to be gained from an increase in household savings at the expense of a corresponding fall in public savings? A third question arising from the proposal relates to the burden of educational expenditure for different income groups and, hence, to equity. Are the poorest groups to benefit most, or will all groups benefit equally? The introduction of a health insurance scheme, especially for low- income earners, has been advanced as a matter of urgency. The bud- getary implications of such a scheme, especially in the long run, should be carefully assessed. The initial objective was first to improve the or- ganizational set-up of health services and only then to allocate more funds. If decisions are made to implement the scheme prematurely, the likelihood of waste is high. As explained in chapter 5, unless the supply of medical manpower and facilities is substantially increased, the effect of health insurance would be to drive up the prices for health care. If the domestic financing targets of the fourth plan are to be met, the policy for pricing foodgrains needs to be reassessed. Little consensus appears to exist in government about the preferred future course of action. Major questions relate to the desirability of continuing to adjust the terms in trade in favor of agriculture through pricing policy-as against the use of other instruments to upgrade incomes in rural areas- and to the merits of continued subsidization of barley-as against the elimination of this subsidy, as proposed earlier, by cross-subsidization. The foodgrain subsidy is only one aspect of agricultural strategy, which also embraces policies for production and procurement. The objective should be not only to reduce and eliminate consumer subsidies, but to prevent an unwarranted growth of stockpiles and related financing needs. To reduce the inflationary impact of the GMF cash deficit, part of the 1975 crop was paid for by issuing W54 billion in six-month grain purchase certificates to farmers. These were soon discontinued because of their unpopularity.19 To replace them government has been issuing grain bonds to financial institutions; sales were W20 billion in 1975 and W100 billion in 1976. Tapping the domestic credit market to 19. The end of the crop year for rice is October; that for barley is June. Certificates were issued at an annual interest rate of 18 percent. Farmers were nevertheless entitled to cash them after three months, in which case they received interest at an annual rate of 12.6 percent. They unanimously chose to do so because they could earn higher returns elsewhere. PUBLIC rGINANCE 361 finance part of GMF cash deficits has the obvious disadvantage of further reducing the already tight availability of credit to the private sector.20 Subsidies to public sector enterprises are another aspect of subsidy policy requiring special attention. Direct budgetary subsidies to public enterprises are now scattered and do not as yet constitute large amounts. In the 1976 central government budget, the largest subsidy element identified was a WI 1.7 billion contribution to the national railway. Although this item was listed under government investment and loan, almost all of it supported the railway's operating account. Also in 1976 KEco received bond interest subsidies of W3.5 million, the Korea Trade Promotion Corporation received a budgeted subsidy of W3.5 billion, and the Mining Development Corporation received W4 billion, which was listed as investment but in fact was a pure subsidy. During the fourth plan period, one key problem will be to prevent additional current subsidies to these enterprises. This can be accomplished by having them generate adequate revenue to service the rapidly rising burden of addi- tional borrowing. Capital requiremiients of public enterprises Three public enterprises facing large capital needs are the Korea National Railroad (KNR), the communications service, and KECO-the only government enterprises which have recently had their prices con- trolled. The national railway operated profitably until 1970. Since that year its revenue has not been sufficient to meet operating expenses and amortization payments. Tariff increases permitted by government over the last several years have been too infrequent and too small to compen- sate for inflation. In addition, growing borrowing to finance capital expenditure has rapidly built up the cost of debt service. During the fourth plan period, the key problem will be to generate enough cash flow from current revenue to service the continuing increase in borrow- ing. Whether this will be possible is doubtful. Early in 1976 rail rates were increased only by 35 percent. Government's price objectives con- flicted with a larger increase and are likelv to constrain future increases. The communications service is a substantial contributor to govern- ment revenue. But the phenomenon is deceptive: the heavy backlog of investment in this sector could not be reduced because of insufficient 20. The interest rate for grain bonds is 17 percent for six-month issues and 18 percent for twelve-month issues, well below the effective lending rates of banks. 362 PART THREE: MAJOR SECTORS resources. Demonstrating this backlog are the facts that the cost of a telephone alone is about $400, but the waiting period for installation is six to twelve months. Defense requirements also exert pressure on de- mand for communications services. Proposals have been made for rate increases, but the constraints on such increases are similar to those for the railway. KEco's capital shortage is severe because of its heavy investment pro- gram. The budgetary capital contribution in 1976 was held down to W20 billion, even though it was known that requirements were at least W100 billion. No other sources could be found, and the supplementary budget requirement for 1976 was about W100 billion. KECO was ex- pected to make a profit of W13 billion in 1976, but one-half of its shares are privately held and it pays a 15 percent dividend to private stock- holders. KEco has made numerous requests for rate increases, but most of these have been rejected for reasons of general price policy.2' On the whole, power rates are not low by international standards, but they do not adequately reflect the true economic costs of power generation, which are high because of the scarcity of domestic energy sources. And in contrast with govemment enterprises, KECO has been paying substan- tial taxes. To relieve the pressure to increase rates, policvmakers decided in 1977 to exempt KECO from taxes-a move which conceals, but does not solve, the problem of insufficient cash generation. If rates were increased by 10 percent annually, KEco would probably generate enough current revenue to service loans completely. But that would still leave KECO with too little cash to contribute to the financing of its investment program. Another question is whether KECO should be subsidized for rural electrification, nuclear power development, and the defense burdens imposed on it. Improvement of budgetary accounts and public finance statistics The key problem in analyzing revenue and expenditure trends is that it is difficult to derive a fully consistent and consolidated picture of government financial transactions from budgetary accounts and public finance statistics. A related problem is that the classification and descrip- tion of items are inadequate. EPB annually undertakes one form of 21. Rate schedules were revised in December 1974 and December 1975. The rate structure is complex; power rates are subsidized for certain export industries and the national railway. PUBLIC FINIANCE 363 consolidation in its overall resources budget. The main limitations of these consolidated data are that they are highly aggregated, lack clarity about financing items, especially foreign borrowing, and do not include extrabudgetary accounts. A second source of consolidated public finance data is the series of national accounts prepared by BOK. These accounts set out consolidated data for general government, which includes local governments and extrabudgetary accounts but excludes government en- terprises and public corporations. The BOR data have the considerable merits of having been systematically netted out and of being more com- prehensive than those of EPB. They become available, however, only substantially later than the budget data. In addition, numerous correc- tions are made to fit national accounting conventions. It thus is difficult to use these data for operational purposes, including projections. Although BOK extensively analyzes and processes public finance data, the useful- ness of this work could be substantially increased by a more informative and systematic explanation of methodology, coverage, and reconciliation with budget data. There also seems to be scope for better coordination of this work within BOK. The budget data prepared by EPB are primarily seen and used as an internal management and planning tool. This may explain why rela- tively little attention has been paid to their form and presentation. Nevertheless considerable scope exists for improvement. The need is particularly great for more clear and comprehensive reporting of foreign debt transactions in the budget document and for a more systematic tabulation and analysis of the numerous transfers. The second improve- ment would enable greater understanding of the differences between gross and net flows of government revenue and expenditure. Another related and more basic aspect of budgetary practices is the use of nu- merous special accounts, some of which are small. Although there is merit in separately studying the expenditure and related financing needs of certain types of government activity, this merit by itself does not justify building into the budget as many special accounts as currently exist. It would seem that the justification for these accounts, with the exception of some for government enterprises, is inadequate. It would, in addition, seem appropriate to reexamine the role and justification of some extrabudgetary funds. As noted earlier, government introduced a number of changes in the accounting system starting with the 1977 budget. These changes appear to be a first step toward reducing the complexity and difficulty of interpreting data. In sum: a concerted effort is needed to upgrade public finance statis- tics, to streamline the structure of budgetary accounts, and to facilitate 364 PART THREE: MAJOR SECTORS understanding of the budget by improving the classification and descrip- tion of items, particularly those for expenditure. This effort will con- tribute to the efficiency of budgetary and planning activities and increase the usefulness of public finance statistics as an important tool of eco- nomic analysis. Chapter 12 Agustin V. Que Finaincial System THE GROWTH AND DIVERSIFICATION of Korea's financial system contrib- uted significantly to the high rates of economic growth registered be- twveen 1963 and 1976.1 Rapid economic growth, in turn, was an impor- tant factor in the expansion of the financial system. Government used existing institutions to channel credit to preferred sectors and projects. It also created new institutions to engage in more specialized activities. The broadened scope and range of financial institutions facilitated the flow of funds between the saving and investing sectors. For the most part these allocational mechanisms were adequate. It is difficult to dem- onstrate, for example, that a different pattern of resource allocation would have led to even higher rates of economic growth, greater devel- opment of the export sector, and lower incremental capital-output ratios. But levels of domestic savings, although improving, were decidedly in- adequate. This resulted in heavy dependence on foreign savings to finance investment. By the mid-1970s government authorities recognized that the country's economic growth could not be sustained by continuing heavy reliance on foreign borrowings and that domestic savings would have to be further encouraged and enlarged. As a result, self-sufficiency in financing investment became one basic objective of Korea's fourth plan for the 1977-81 period. Financial Structure and Development At the end of 1975 the central bank, Bank of Korea (BOX), accounted for about 16 percent of total financial assets; the commercial banking system for about 61 percent (table 12.1). With more than three-quarters 1. The results of a study by Adelman and Morris show that the degree of improvement in financial institutions is the strongest index of dynamic economic performance. Irma Adelman and Cynthia Taft Morris, "Performance Criteria for Evaluating Economic Development Potential: An Operational Approach," Quarterly Journal of Economics, vol. 82, no. 2 (May 1968), pp. 260-80. 365 366 PART THREE: MAJOR SECTORS Table 12.1. Structure of the Financial System, 1975 Annual growth Percent- rate' Assets age of (percent) (millions total 1965- 1971- Institution of won) assets OfFices 71 75 Monetary system 10,793.9 76.8 1,478 27.1 13.9 Bank of Korea 2,173.7 15.5 1 -0.3 12.8 Commercial banking systemb 8,620.2 61.3 1,477 41.5 11.4 Commercial banks 5,302.3 37.7 640 36.8 18.9 Nationwide commercial banks 4,112.1 29.2 383 36.6 14.3 Local banks' 661.0 4.7 203 128.0d 49.9 Foreign banks' 259.8 1.8 9 75.2d 48.1 Banking accounts of the Korea Trust Bank 269.4 1.9 45 9.3 46.4 Specialized banks 3,317.6 23.6 837 47.1 3.0 Korea Exchange Bank' 1,927.3 13.7 28 29.1d 0.2 Medium Industry Bank 287.7 2.0 79 26.7 8.3 Citizens National Bank 265.7 1.9 126 35.9 4.7 Korea Housing Bank' 167.7 1.2 50 44.4d 14.4 Credit Sector of Agri- cultural Cooperatives 608.3 4.3 489 15.1 7.6 Credit Sector of Fish- eries Cooperatives 60.9 0.4 65 31.1 3.2 Nonbank financial institutions 3,264.6 23.2 1,348 26.9 10.8 Development banks 2,381.5 16.9 11 25.6 8.6 Korea Development Bank 2,328.5 16.6 10 25.2 8.4 Korea Development Finance Corporation' 53.0 0.4 1 47.0' 17.9 Trust accounts' 244.4 1.7 - 41.4 -2.8 Insurance companies 206.7 1.5 1,235 24.8 12.3 Short-term finance companiesg 221.7 1.6 10 - 142.7h Mutual finance companiesg 63.7 0.5 71 - n.a. Postal savings accounts 47.0 0.3 n.a. 24.9 10.3 Securities firms 99.6 0.7 102 25.8 52.8 Total 14,058.5 100.0 2,826 27.1 13.1 - Not applicable. (Notes to table 12.1 appear n.a. Not available. on the following page.) FINANCIAL SYSTEM 367 of total assets accounted for by the monetary system, the Korean finan- cial system was in many respects still at an early stage of development, even in comparison with other LDCS.2 The most important reason for this preponderance of the monetary system has been the use by govern- ment of commercial banks to implement a comprehensive system of se- lective credit to ensure the flow of funds to preferred sectors. Commer- cial banks were used because of their extensive branch networks; enterprises needed only to maintain relationships with their primary banks to apply for and receive earmarked funds. Another reason for using commercial banks was the substantial direct and indirect control that the Ministry of Finance (MoF) and BOK exercise over these institutions. The uniformity of profits of the five nationwide banks, which control more than 70 percent of total commercial bank assets, reflects the oligo- polistic structure of commercial banking in Korea. Recent information 2. A survey of nineteen LDCS conducted by Goldsmith in 1967 showed that monetary systems accounted for 67 percent of total assets: assets of central banks constituted 24 percent, and those of commercial banks 43 percent. Securities firms and finance companies accounted for only 1 percent. For seventeen developed countries Goldsmith's figures showed that central banks accounted for only 9 percent of total assets, commercial banks 33 percent, and nonbank financial institutions 58 percent. Raymond W. Goldsmith, "The Development of Financial Institutions during the Post-War Period," Banca Nazionale del Lavoro Quarterly Review, no. 97 (June 1971), table 15, p. 166. (Notes to table 12.1.) Note: Assets and liabilities among classes of institutions are not netted out. Com- pared with statements of outstanding assets and liabilities presented in BOK'S flow- of-funds accounts, the aggregate figures here are much higher. In fact, they are about double those shown in the BOK tables. Nevertheless the main outlines of the financial system are not materially changed by using either gross or net assets. Sources: BOX, Monthly Economic Statistics, February 1976; BoK, Economic Sta- tistics Yearbook, various years; figures supplied by Korea Investment and Finance Corporation (KIFc) and Korea Securities Finance Corporation (KSFC). a. Deflated by implicit cNP deflator (1970 = 100). b. Corresponds to deposit money banks as defined by BOK. c. Established in 1967. d. 1967-71. e. 1968-71. f. Beginning in 1970, trust accounts of commercial banks were transferred to Korea Trust Bank. g. Established in 1972. h. 1972-75. 368 PART THREE: MAJOR SECTORS on bank profitability is not available, but a study covering the 1970-72 period indicates little variance in the profit levels of these five banks.3 This oligopolistic structure is reinforced by two basic facts: profit has not been a major objective of bank managements; and commercial banks continue to be used by government as tools of national economic policy. In addition, to enable nongovernment shareholders of banks to receive a reasonable dividend, the nominal income of these banks is usually supplemented by interest received on reserve deposits held by BOK. The rate of interest paid by BOK is retrospectively set after profits are com- pared with target rates of return for a specific period. A large specialized banking sector functions side by side with com- mercial banks, giving the Korean financial system a dual structure (table 12.2). Because of MOF's concern with directing funds to preferred 3. Leroy P. Jones, Public Enterprise and Economic Development: The Korean Case (Seoul: Korea Development Institute, 1975), pp. 187-89. Table 12.2. Specialized Banks: Functions and Sources of Funds Bank/year Principal sources established Principal functions of funds Korea Development Largest term-financing institu- Borrowings from gov- Bank' (snB) tion; used by government as ernment (general fiscal (1954) principal financing channel for funds, National Invest- implementing five-year plan proj- ment Fund, and USAID ects; services the electricity, coal, counterpart bor- and shipbuilding sectors, and rowings), the World most industries in the manufac- Bank, Asian Develop- turing sector; provides equity ment Bank (ADn), investments, guarantees for cor- Kreditanstalt fur porate borrowings in domestic Wiederaufbau (KFw), and foreign currencies, long and various foreign banks; medium-term loans for equip- equity. ment, and loans to term clients for working capital. Korea Exchange Handles foreign exchange trans- Foreign currency de- Bank (xEra) actions; buys and sells foreign posits of nBOK and bor- (1967) exchange; holds foreign currency rowings from foreign funds; issues letters of credit, banks. loans, discounts, and acceptances of bills; guarantees foreign obli- gations of Korean corporations; depository for the foreign assets of BOsO. FINANCIAL SYSTEM 369 Table 12.2 (continued) Bank/year Principal sources established Principal functions of funds Medium Industry Provides term and working- Time and savings de- Bank (MIm) capital loans to, and discounts posits; borrowings from (1961) the bills of, small- and medium- government, BOK, ADB, sized firms. the World Bank, and KOB. Citizens National Provides banking services to Time and savings de- Bank (cNB) households and small-sized firms, posits; mutual install- (1963) mainly petty loans secured by ment deposits; demand mutual installment accounts; deposits; borrowings established to take over mutual from government and savings and loan functions of BO5K; equity. traditional cooperative financial societies. Korea Housing Provides housing loans to low- Demand, time, savings, Bank (sHB) income families; extends con- and housing-install- (1967) struction financing to Korea ment deposits; Korea Housing Corporation (nRc) and housing bonds and de- private developers for apart- bentures; borrowings ments; extends long-term and from government. low-interest construction financ- ing to local governments and individuals. Credit Department, Provides loans to member coop- Borrowings from gov- National Agricul- eratives, local governments, and ernment and BoX; tural Cooperatives nonprofit organizations to demand, time, and Federation and finance agricultural and fishing savings deposits. member coopera- activities. tives (1961) Credit Department, Provides banking services to Borrowings from gov- Central Federation fishermen, to manufacturers of emient, BOX, and of Fisheries fishing equipment and materials, nationwide commercial Cooperatives and and to fishing villages for devel- banks; demand, time, member coopera- opment; extends loans and and savings deposits. tives (1962) credits to member cooperatives. Note. All operate as commercial banks, except siB which is a nonmonetary fi- nancial institution. Sources: Bol0, Financial System in Korea, 1974; BO1K, Monthly Economic Sta- tistics, January 1976. a. Successor to the Industrial Bank of Korea, established in 1918. 370 PART THREE: MAJOR SECTORS sectors, each specialized bank was separately chartered. Commercial banks, which operate as general purpose institutions, were initially pro- hibited from engaging in the same line of business as specialized banks. This functional definition worked reasonably well in meeting the objec- tives of selective credit policies. But it also tended to compartmentalize and insulate classes of institutions, which in turn resulted in a lack of competition in certain areas and a lack of funds for many borrowers.4 Government control over the most important financial institutions is the predominant factor to consider when evaluating the structural characteristics and performance of the Korean financial system. BoK, under MOF supervision, exercises normal control and supervision over the commercial banking system. MOF directly supervises specialized banks and all other nonbank financial institutions. The five nationwide commercial banks are fully controlled by government, as are such gov- ernment banks as Korea Exchange Bank (KEB), Medium Industry Bank (Mm), and Korea Development Bank (KDB). The extent of this control covers a wide range of operating aspects. BoK and MOF set all nominal interest rates, specify the composition of asset portfolios, and effectively influence the decisions to allocate loanable funds. All banks are subject to periodic examination by the Superintendent of Banks to ensure that records are accurate and banking regulations complied with. The most adverse consequence of financial dualism and tight govern- ment control was the resurgence of the curb or unorganized money market (uMm) in the late 1960s. Not subject to MOF and BoK control and supervision, the UMM served to relieve pressure in the organized financial sector. In August 1972, however, a presidential decree effec- tively minimized the size of the U1MM. The decree froze all UMM loans and required licensed business enterprises to report to government all borrowings raised from sources other than organized financial institu- tions. It also lengthened the maturities of some loans and converted loans of shareholders to their own firms into equity. The total of UMM borrow- ings declared by firms under the decree was W345.6 billion, an amount equivalent to 80 percent of the money supply (M1) at that time and to 24 percent of the outstanding domestic credit extended by the bank- 4. Commercial banks and Korea Development Bank (KDB) were subsequently permitted to engage in business previously designated only for specialized banks. Nationwide commercial banks were permitted to deal in foreign exchange, along with Korea Exchange Bank (Knia), and encouraged to lend to small- and medium- sized enterprises, along with Medium Industry Bank (MIs). FINANCIAL SYSTEM 37I Table 12.3. Distribution of Assets of Financial Institutions, 1963-75 (percent) Institution 1963 1967 1971 1975 Monetary system 76.2 77.2 64.3 79.1 Bank of Korea 38.3 23.6 17.9 19.8 Commercial banking system 37.9 53.6 46.4 59.3 Commercial banks 19.4 25.9 32.1 37.4 Specialized banks 18.5 27.7 19.3 21.9 Nonbank financial institutions 23.8 22.8 35.7 20.9 Development banks, insurance companies, and thrift organizations, 23.2 22.6 34.4 17.1 Investment institutions' 0.6 0.2 1.3 3.8 Total 100.0 100.0 100.0 100.0 Source: aoK, Flow of Funds in Korea, 1963-70; noK, Economic Statistics Year- book, 1973 and 1976. a. Includes mutual savings companies, trust accounts, and postal savings. b. Includes short-term finance companies and securities firms. ing system.5 Concurrently MOF encouraged the establishment of a new class of institutions-short-term finance companies (SCs)-to handle transactions of unsecured commercial paper having maturities of one to ninety days. The Korea Investment and Finance Corporation (KIFc), an institution founded in 1972 by the Korea Development Finance Corporation (KDFC), was the first sTFc to be established. Nine other sTFcs have since been established, five of them in Seoul. The institutional reforms of 1972 signaled a shift in the direction of the MOF policies guiding the development of the financial system. Based on the success of STFCS, MOF adopted a policy of encouraging the devel- opment of the secondary finance bloc-consisting of such nonbank financial institutions as development banks, STFCS, finance companies, and securities firms-in order to diversify the means for collecting and allocating financial savings. Despite the growth in the number of these institutions in recent years, the commercial banking system continues to be dominant (table 12.3). Fluctuations in the share of specialized 5. Ml is currency in circulation plus demand deposits (narrow money). Data on curb market borrowings are from aox, Report on the Results of the Presidential Emergency Decree on August 3, 1972 (Seoul, 1973), p. 28. 372 PART THREE: MAJOR SECTORS banks and development institutions resulted from their limited capital bases, their dependence on government's fiscal position, and MOF'S per- mitting commercial banks to engage in certain lines of business pre- viously reserved for specialized banks. The last factor was the strongest determinant of the relative decline of specialized banks, particularly KFs. Nonbank financial institutions, specifically KDB, lost the large share of institutional assets they had in 1971. But the true extent of the shift is not evident from the figures shown here because securities firms act as agents in transactions, not as principals, and their balance sheet positions do not adequately reflect the volume of financing they facili- tated. This has been especially true in recent years, when corporations raised substantial amounts through the securities markets. Efforts to promote the securities markets began in 1968 with the pro- mulgation of the Capital Market Promotion Law. The law established conditions for the sale of govemment-owned securities to the public, gave preferential subscription rights to employees of issuing corporations, and created the Korea Investment Corporation (KIC) to underwrite the sale of securities and perform other functions contributing to the develop- ment of Korean capital markets. This was followed by the August 1972 presidential decree which put pressure on the corporate sector to balance their financial structures through increased equity capitalization. The decree provided a significant stimulus to the latent corporate bond and equity markets which had been moribund since the stock market col- lapsed under a wave of speculation in 1962. Then the Public Ownership Inducement Law was passed in December 1972. It provided a legal basis for rationalizing the financial structures of large business enterprises by compelling them to open up to public ownership by the sale of exist- ing stock or the public offering of new issues. The intent was to raise the equity base of these corporations and achieve a better balance between their capital resources and borrowed funds. Insurance schemes have not been quantitatively important in the Korean financial system. This situation contrasts with that in a number of countries where insurance organizations are significant mobilizers and sources of long-term finance. There is no organized social security system. Except for pension schemes instituted only in 1975 for the military, government employees, and school teachers, no substantial pension funds are administered by government. The National Welfare Pension Fund was approved by the National Assembly for introduction during the fourth plan period, but this has been deferred. Private life insurance companies have not done well either, mostly because the FINANCIAL SYSTEM 373 Table 12.4. Measures of Financial Development, 1963-75 (percent) Measure 1963 1967 1971 1973 1975 Total assets of financial institutions to GNP 32.4 40.4 65.5 74.1 56.31 Financial intermediation ratio (Fix) 23.9 43.4 38.0 50.9 58.7 Source: Calculations by Dr. Park Yung Chul, Korea University, Seoul. a. Excludes rest-of-the-world holdings, in accord with Box's 1975 definitions for flow-of-funds accounts. policies marketed to date have not provided adequate protection and because institutional problems have beset the life insurance industry. Effects of Institutional Reform Assets of financial institutions grew from just over 30 percent of GNP in the mid-1960s to more than 70 percent by the mid-1970s. Measured by another indicator-the financial intermediation ratio (FIR) which is the proportion of total flows of nonfinancial sector issues held by finan- cial institutions-the importance of financial institutions in facilitating the transfer of resources among sectors nearly trebled (table 12.4).o The decline of the FIR in the 1967-71 interval resulted from the resur- gence of the UMM that followed the gradual dismantling of the regime of high interest rates that lasted from September 1965 to April 1968. Banking system As tools of govemment's selective credit policies, banks grew rapidly in the 1965-71 period. Growth was spurred by the regime of high in- terest rates adopted in September 1965.7 At the start of the regime, 6. Goldsmith's more commonly used ratio is that of total assets of financial institutions to GNP. But by using flow-of-funds accounts and making possibly egregious assumptions about the complete netting out of security issues between the financial and nonfinancial sectors, the financial intermediation ratio can be calculated, albeit imperfectly. Raymond W. Goldsmith, Financial Structure and Development (New Haven and London: Yale University Press, 1969). 7. For a thorough discussion see Gilbert T. Brown, Korean Pricing Policies and Economic Development in the 1960s (Baltimore and London: Johns Hopkins University Press, 1973), chap. 7. 374 PART THREE: MAJOR SECTORS nominal interest rates for time and savings deposits dramatically in- creased. Rates for time deposits of three and six months doubled; the rate for those of more than one year increased from 15 percent to 26.4 percent. Nonpreferential lending rates accordingly increased; discounts on bills rose from 14 percent to 24 percent; call loans from 12 percent to 22 percent. The purposes of the reform were to mobilize additional domestic resources, minimize inflation by dampening investment de- mand, and improve resource allocation. The effects of the 1965 reform on institutionalizing financial trans- actions in Korea were remarkable. Although the assets of financial insti- tutions rose only from 27.4 percent of GNP in 1964 to 30.8 percent in 1965, the FIR more than doubled during that period, rising from 21.4 percent to 52.5 percent. Bank deposits subsequently increased substan- tially, further enhancing the primacy of banks in the financial system. Specialized banks, as a group, also grew rapidly in the 1965-71 period. Two of the most important ones, Korea Exchange Bank (KEB) and Korea Housing Bank (KHB), were established in 1967; the operations of Medium Industry Bank (MIB) and Citizens National Bank (CNB) benefited from increased govemment impetus on lending to smaller firms. As events subsequently demonstrated, however, the period of reform was too short to sustain the momentum initially generated. In April 1968 BOK began to lower the rates for bank deposits and loans, and financial disintermediation through the UMM accelerated until 1972. But with the implementation of financial innovations specified in the presi- dential decree of August 1972, the Fm jumped to 54.9 percent in that year. By the end of 1975 the FIR, buoyed by the increase in the number and diversity of financial institutions after 1972, reached 58.7 percent.8 The growth of financial institutions during the 1963-75 period was sustained by the expansion of banking liabilities. This expansion was a consequence of govemmenes emphasis on time and savings deposits as the primary instruments of resource mobilization during the regime of high interest rates between 1965 and 1968. Table 12.5 shows the struc- ture of changes in the outstanding liabilities of financial institutions during three periods, 1963-67, 1967-71, and 1971-75. In all three periods M1 and time and savings deposits accounted for about one-half 8. Financial layering ratios, calculated as the ratio of total issues of financial institutions absorbed by the financial sector to those absorbed by the nonfinancial sector, provide another measure of transactions among institutions. The financial layering ratio in Korea was 9.2 percent in 1963-66 and 38.3 percent in 1972-75. FINANCIAL SYSTEM 375 Table 12.5. Sources of Growth of Financial Institutions, 1963-75 Change in outstanding stock 1963-67 1967-71 1971-75 Per- Per- Per- Bil- cent- Bil- cent- Bil- cent- lions age lions age lions age of of of of of of Instrument won total won total 'won total Money (M1) 78.6 22.1 240.8 15.5 820.9 18.0 Time and savings deposits 121.8 34.2 585.2 37.7 1,401.9 30.8 Insurance and trust 32.3 9.1 126.0 8.1 130.0 2.9 Securities 10.1 2.8 65.3 4.2 243.6 5.3 Equities other than stocks 6.4 1.8 42.7 2.8 260.3 5.7 Government loans 31.8 8.9 109.8 7.1 333.4 7.3 Foreign claims and debts 3.0 0.8 173.3 11.2 776.7a 17.1 Other' 71.8 20.2 207.5 13.4 586.5 12.9 Total 355.8 100.0 1,550.6 100.0 4,553.3 100.0 Note: All figures are in current prices. Sources: BoK, Flow of Fuands in Korea, 1963-70; BOK, Flow of Funds in Korea, 1963-75; BOK, Economic Statistics Yearbook, 1973; and World Bank estimates. a. Estimated on the basis of outstanding indebtedness of BOK, KDB, and com- mercial and specialized banks at the end of 1975. b. Calculated as a residual. of the increase in liabilities of financial institutions. These deposits were largely provided by the household and corporate sectors. Foreign and government sources have been the major suppliers of funds to the financial system. The substantial increases in foreign claims and debts in the 1967-71 and 1971-75 periods coincided with rapid economic growth and balance-of-payments difficulties. Total investment requirements outstripped the supply of domestic funds, and aggregate savings targets were not met. Foreign commercial banks provided most of these funds through transactions with Korean commercial banks. An associated factor contributing to the growth of financial institutions has been the large volume of transactions between institutions, of which the relationship between BOK and commercial banks is one significant aspect.9 In recent years particularly, the expansion of bank assets, or 9. Commercial banks include specialized banks, such as nr.a and Korea Housing Bank CxHB), which perform commercial banking functions. This definition con- forms to BoK's designation for deposit-money banks. 376 PART THREE: MAJOR SECTORS bank credits, has been stimulated by BOK rediscount facilities and direct govemment loans (table 12.6). Securities market Momentum generated by the Public Ownership Inducement Law was sustained by measures announced in May and June 1974. The May 1974 measures focused on three points: dispersion of ownership of large, privately held business enterprises; improvement of banking credit con- trols, especially over privately held enterprises heavily dependent on bank credit; and rationalization of the financial structures of large firms. Enterprises subject to these measures were separated by MOF into two Table 12.6. Flow of Funds between Government and Financial Institutions, 1963-75 (billions of won in current prices) Flow of funds 1963-66 1967-71 1972-75 From government to financial institutions 38.6 232.6 731.7 Government loans' 22.5 123.6 386.3 Equity subscriptions 4.5 41.9 185.9 From financial institutions to government 5.2 66.9 845.8 BOK loans 1.5 7.7 572.2 Commercial bank loans 1.0 28.0 28.5 Purchase of securities 2.6 8.3 155.0 Net flows from government 35.4 165.7 -114.1 From banks to BOK 32.4 149.0 389.4 Reserve deposits 27.0 69.3 273.1 From BOK to banks 4.8 115.4 559.4 BoK loans 4.4 105.4 555.0 Net flows from banks 27.6 33.6 -170.0 Sources: BOK, Flow of Funds in Korea, 1963-70; Box, Economic Statistics Year- book, 1976. a. Loans to BoK, commercial banks, and xDB for on-lending to government supported projects. b. The definition of commercial banks corresponds to Box's definition of deposit money banks which include nationwide and regional commercial banks and spe- cialized banks, but exclude KDB. FINANCIAL SYSTEM 377 categories on the basis of their financial standing: sound and unsound. Sound companies were the first to be enjoined to go public. Failure to comply would have led to suspension of government financial support, foreign exchange privileges, and tax exemptions. Unsound companies were given three years to improve their financial structures. In the in- terim they were neither granted payment guarantees of foreign loans for new projects nor allowed to establish, purchase, or own new enter- prises or to invest in shares of stock. The measures encouraging companies to open up to the public were supplemented by MOF's announcement in June 1974 of a comprehen- sive capital-market development plan intended to facilitate the absorp- tion of new issues of stocks and debentures. The plan had two main features: recapitalizing and restructuring the Korea Securities Finance Corporation (KSFC) to administer credit programs for underwriting new issues; and introducing a number of structural reforms in the securities markets. First, installment savings plans administered by brokerage firms were introduced to facilitate the investment by small investors in stocks and bonds. Employee stock-ownership systems were also improved by making stock-distribution rights mandatory. The installment savings plans for acquiring securities were subsequently included in the 1976 Wage-earners' Savings Scheme.10 Second, investment trusts were devel- oped. The Korea Investment Trust Company (KITC), owned by twenty- seven brokerage firms, and Korea Investment Corporation (KIc) started marketing shares in bond and stock trusts. These trusts provided attrac- tive yields in the wake of the stock market boom of mid-1975. Other market-oriented steps were introduced in 1974 and 1975. First, a firm underwriting system for the sale of new issues was initiated to improve the placement of securities. Second, requirements for finan- cial disclosure were improved by requiring comprehensive prospectuses for all new security issues and by upgrading accounting practices and standards through the promulgation of strict principles. Third, a new auditing division was established in the securities and insurance bureau of MOF to broaden MOF'S supervisory and regulatory activities. More important, AOF began to consider transforming Kic into an autonomous 10. Under this scheme about 60 percent of the labor force is eligible to establish installment savings accounts in annuity contracts of two, three, or five years. No interest payments are made during the contract period; at the end of the period government will pay interest plus a lump sum bonus equivalent to 9.6 to 15 per- cent a year, depending on the maturity. Interest income will be tax free, and an income tax credit equal to 15 percent of the principal saved will be given. 378 PART THREE: MAJOR SECTORS securities and exchange commission to oversee open and closed corpora- tions. Finally, pricing in the bond markets was somewhat improved by an announcement in March 1976 that new issues of bonds could be sold with coupons ranging from 15 to 21 percent. Complementing the new yield structure, the March 1976 measures permitted the issue of new types of corporate debentures which carry detachable rights, have sinking funds, and are partially convertible into stock. To encourage longer maturities, MOF recently permitted the issue of five-year, partially indexed debentures, for which the coupon rate after three years could be adjusted upward or downward to maintain a fixed percentage spread with the rate for time deposits held at commercial banks. These policy directives from MOF were all steps in the right direction. But the bond market would function more efficiently if new issues of bonds, instead of being sold only at par, could be sold at discounts or premiums on par to reflect current market conditions. The response of the issuing and investing sides of the securities markets to these measures was remarkable (table 12.7). The increased supply of new stocks attested to the success of government's going-public program, which induced not only the companies subject to the program to go public, but other firms not so designated. Corporate debt issues also grew in size, but in 1975 and 1976 underwriters had difficulty placing them because authorized coupon rates were less than the yields of 21 to 23 percent in the secondary market and discounts to ultimate purchasers were not allowed by MOF. At these less-than-market rates, Table 12.7. Public Offerings of Stocks and Bonds, 1968-76 Stocks Bonds Total Billions Billions Billions Year Companies of tvon Companies of won Companies of -won 1968 2 0.16 2 0.16 1969 12 2.21 ... 12 2.21 1970 9 2.07 ... .. 9 2.07 1971 4 0.85 . ... 4 0.85 1972 7 1.08 35 9.93 42 11.01 1973 47 21.47 12 3.45 59 24.92 1974 19 14.34 59 27.87 78 42.21 1975 62 39.88 67 33.45 129 73.33 1976 87 74.00 112 86.28 199 160.28 . I .Zero or negligible. Source: Data supplied by the Securities and Insurance Bureau of the Ministry of Finance (MoF). IJINANCIAL SYSTEM 379 underwriters held unsold portions of the issues, usually about half, for periods of up to two months, after which the issues were sold to banks and insurance companies at 100 percent. Taking accrued interest into account, this practice increased the yield to purchasers. Another diffi- culty has been that individuals are not attracted to debt issues. They do not benefit from the commission sacrificed by underwriters, and they can obtain higher returns from equities and investment-trust certificates. The March 1976 measures improved this situation somewhat, but it seems clear that the market could be made more fluid and orderly only by allowing discounts and premiums on par. The secondary or trading market, buoyed by the increased supply of securities arising from the going-public measures, has also been very active. Trading in stocks listed on the Korea Stock Exchange (Is;E) rose from W20 billion in 1968 to only W33.8 billion in 1972, reflecting the market's moribund condition before reforms were introduced in 1972. In 1973 the value of stock trades more than doubled. The trading volume was higher, and the price index increased 37.4 percent. In 1974, despite a drop in stock prices, trading volumes again more than doubled. Since then the market has boomed, benefiting from the increased pace of economic activity after mid-1975. Despite price level fluctuations, the trend of the KSE composite stock-price index has been upward, rising 39.8 percent from year-end 1974 to year-end 1976. Dividend yields stabilized at about 13 percent during the same period, and price-earnings ratios at about 9:1. This indicated a fairly substantial increase in the earnings and dividend payments of listed companies. As a result of all this activity, the share of stocks and debentures in gross financial assets acquired by households rose from 5.1 percent in 1968 to 29.2 percent in 1972 and stabilized at that level until 1975. Total securities-com- prising national and public bonds, financial debentures issued by finan- cial institutions, and corporate issues-correspondingly accounted for 5.7 percent of gross financial assets acquired by individuals in 1968 and rose to about 29.8 percent in 1972-75. Mobilizing Financial Savings Before 1971 Korea was in the unique position among LDCS of realizing a surplus from its govemment sector. This reflected the large transfers received by government from abroad and the fairly tight budgetary con- trols in effect. During 1963-65 government surpluses of 2 to 2.5 percent 380 PART THREE: MAJOR SECTORS of GNP covered one-third to one-half of total financial deficits. Financial surpluses subsequently declined as government investment increased and tax exemptions reduced the rate of growth of total revenue. Since 1971 government has been a deficit sector. These deficits were exacerbated in 1974 and 1975 by deficits in the Grain Management Fund and the Fertilizer Fund. Buoyant tax revenue and changes in purchase and release prices reduced these deficits in 1976. Household and unincorporated enterprises were negligible suppliers of funds before 1965. After the interest-rate reform of September 1965 household surpluses quadrupled and continued to increase until 1967, when they reached 3.5 percent of GNP. With rising incomes and positive real interest rates, the household financial surplus rose to 4 percent of GNP in 1972 and to an estimated 4.5 percent of GNP in 1976. These domestic surpluses were not sufficient to cover the deficits of the cor- porate sector; beginning in 1971 they were insufficient to cover the deficits of the government sector (table 12.8). Consequently Korea has been dependent on borrowings from the rest-of-the-world (ROW) sector, and the pattern of foreign borrowing has been influenced by the balance of payments performance. In 1972 and 1973, for example, ROW sur- pluses amounted to about 3 percent of GNP. These were well below the 7 percent average in 1967-71 and the 11 percent average for 1974-75. In 1976, with the strong performance of Korean exports, ROW surpluses were only 1.1 percent of GNP, or substantially less than EPB'S onginal estimate of 3 percent. Table 12.8. Financial Surpluses and Deficits, by Sector, 1963-76 Percentage of gross national product Sector 1963 1967 1971 1975 1976k Financial -0.6 0.1 -0.1 0.5 0.1 Government 2.5 1.5 -1.2 -3.2 -1.1 Corporate -7.0 -10.8 -9.7 -10.2 -6.5 Household 0.1 3.3 1.9 5.0 4.5 Rest of the world 3.8 4.1 9.4 10.2 3.0 Statistical discrepancy 0.7 1.8 -0.2 -2.3 Financial surplus = deficit 7.1 10.8 11.3 15.7 7.6 . . . Zero or negligible. Note: Includes net transfers from the ROW sector to the government and house- hold sectors. Sources: BOK, Flow of Funds in Korea, 1963-70; Bo}:, Economic Statistics Year- book, 1976. a. Estimated by EPB. FINANCIAL SYSTEM 38I Factors influencing household financial savings Time deposits of commercial banks were the means used by monetary authorities to implement the reform instituted in 1965, when nominal interest rates on all time and savings deposits were nearly doubled (figure 12.1). The response of households to these measures was imme- diate. Their acquisition of time and savings deposits increased in current terms from W1.5 billion in 1964 to W14.6 billion in 1965. Spurred by Figure 12.1. Nomnzinal Rates on Time and Savings Deposits, 1962-76 Annual nominal interest rate (percent) 30 --r______ 28- 26 - 24- L 2 2 20 1 8 16 ' 4L 12 1 0 8 6 4 1962 I 1964 1 1966 I1968 1 1970 1 1972 1 1974 I 1976 1963 1965 1967 1969 1971 1973 1975 three-month time deposits six-month time deposits -. -. - one-year and longer time deposits -- - -- installment savings deposits Source: BOK, Monthly Economic Statistics (February 1976), table 11, p. 23. 382 PART THBEE: MAJOR SECTORS Table 12.9. Composition of Gross Financial Investments of Households, 1963-76 (percent) Source 1963-66 1967-71 1972-75 1976k Money (Mi) 30.9 22.1 19.9 14.7 Time and saving deposits 40.9 49.5 35.2 33.4 Other deposits and loans 13.3 11.8 6.1 13.1 Insurance and trust 10.3 13.1 5.6 8.0 Securities' 8.9 15.9 33.5 29.7 Miscellaneous -4.3 -12.4 -0.3 1.0 Total (billions of won in current prices) 130.3 920.2 2,011.5 1,180.5 Sources: aoK, Flow of Funds in Korea, 1963-70; Bo1, Economic Statistics Year- book, 1973 and 1976. a. Estimated by EPB. b. Includes equities other than stocks. rapid economic growth and substantial increases in real disposable in- comes, this growth continued until 1969 despite the lowering of nominal. deposit rates in successive steps starting in April 1968 (table 12.9). During this period of relative price stability, the demand for all financial assets, except government bonds which were being redeemed, grew alongside time and savings deposits. Curb market loans also grew, in- dicating that the retreat from the 1965 reform, although not adversely affecting the growth of financial assets, was too early for the savings effort to be institutionalized adequately. When the economy entered into recession in 1970, the demand for financial assets declined. The decline was arrested only when reforms were introduced in 1972. Because of the abatement of inflation, the freezing of curb market loans, and the introduction of new instruments, households acquired financial assets worth W420.8 billion in 1972, double the amount in real terms of acquisitions in 1971. The gross acquisition of financial assets in real terms subsequently slowed down despite government measures intended to increase the supply of and demand for securities."1 The effect of these measures was to alter the 11. Under the Public Ownership Inducement Law passed in 1972, privately held corporations designated as eligible by MOP had to sell at least 30 percent of their shares to the public through a legitimate underwriting. This requirement was subsequently raised to 40 percent. FITNANCIAL SYSTEM 383 composition of gross financial savings: the 33 percent share of securities in 1971-75 was more than triple the average for 1965-70. The slower rate of financial asset acquisition by the household sector in 1974 and 1975 was primarily the result of the slower real growth of incomes. Inflation rates in these two years reached record levels in the wake of worldwide inflationary and recessionary conditions. Under these circumstances real rates of return on financial assets fell. Even though nominal rates of interest were raised in 1974 and 1975, the realized real rates of interest for three-month and one-year time deposits were more negative than they had been since 1964. With the economy recovering in 1976, however, the rate of financial asset acquisition by households again increased. And in connection with the institution of a limited prime-rate system, the monetary board increased the inter- est rates of the commercial banking system in August 1976. Time de- posits were increased by I to 2.4 percentage points, with the bigger increases applying to deposits of shorter maturity. Other rates were increased by 1 percentage point. Factors influencing corporate savings During the 1963-75 period corporate savings in 1970 prices grew at a compound rate of 11.2 percent a year; corporate investment grew at 12.5 percent a year. The growth of corporate savings was constrained by economic and institutional factors: the incentive to retain earnings has been weak; depreciation has not been significant as a source of corporate savings.12 As a result, corporate finance was characterized by the steady decline of corporate savings as a proportion of total cor- porate funds (table 12.10). Because corporate savings were inadequate, investments were financed largely from sources external to the corpo- rate sector. This pattern might be expected, however, given the rapid growth of GNP. Corporate financial deficits fluctuated around a level of 7 to 10 percent of GNP during 1963-73 and reached a high of 14.3 percent in 1974. By 1976, however, they were reduced to 6.5 percent. 12. The discussion of corporate savings in this section builds upon the compre- hensive analysis of corporate savings in II SaKong, "An Overview of Corporate Finance and the Long-Term Securities Market in Korea," Korea Development Institute Working Paper, no. 7505 (Seoul, 1975). It has also benefited from the treatment of these issues in Jeffrey E. Williamson, "Domestic Savings in Korea: Can a Pessimistic Past be Reconciled with an Optimistic Plan?" (Washington, D.C.: World Bank, 1976; processed). 384 PART THREE: MAJOR SECTORS Table 12.10. Composition of Sources and Uses of Corporate Funds, 1963-76 (percent) Item 1963-66 1967-71 1972-75 1976' Sources Corporate savings 27.3 16.9 19.7 29.5 Retained earnings 14.1 5.8 n.a. n.a. Depreciation 13.2 11.1 n.a. n.a. External finance 67.0 85.9 83.5 70.5 Securities 12.2 13.5 18.9 16.7 Domestic loans 24.0 32.5 39.8 43.3 Foreign loans 19.5 23.6 21.8 14.4 Miscellaneous 11.3 16.3 3.0 -3.9 Adjustment 5.7 -2.8 -3.2 ... Uses Real investment 82.1 66.5 65.1 60.4 Fixed capital formation 67.9 60.2 n.a. n.a. Inventories 14.2 6.3 n.a. n.a. Financial investment 17.9 33.5 34.9 39.6 Money and deposits 3.1 10.2 16.8 16.7 Securities 2.6 3.4 5.9 6.7 Net trade credit 2.2 2.9 2.1 4.6 Otherb 10.0 17.0 10.1 11.6 Total sources and uses (billions of won in current prices) 347.9 2,387.9 5,617.9 2,516.1 n.a. Not available. . Zero or negligible. Note: Breakdowns of corporate savings and real investment were not available for 1975. For 1972-74 retained earnings and depreciation respectively accounted for 5.3 percent and 14.5 percent of total sources; fixed capital formation and inventories respectively accounted for 57.4 percent and 8 percent of total uses. Sources: Box, Flow of Funds in Korea, 1963-70; BOK, Economic Statistics Year- book, 1976. a. Estimated by EPB. b. Comprises deposits with nonbank financial institutions, private loans, checks being cleared, and miscellaneous investment. IFINANCIAL SYSTEM 385 RETAINED EARNINGS. The ratio of net profits to net worth of Korean manufacturing firms-a reasonable indicator of economywide retums- fell from 17 percent in 1960-62 to 12 percent in 1967-71.13 This trend was reversed in 1972-74, when profitability reached 23 percent. Al- though profit levels were squeezed because of recessionary conditions in 1975, a turnaround was under way in 1976. Throughout the period the increases in the ratios of value added to tangible fixed assets in manufacturing and of net profits to total assets were less than might have been expected. The main reason for this was the changing pattern of industrialization, but there were four additional factors: low prices in international markets for Korean exports; govemment policies that possibly lowered the relative price of capital until the early 1970s; the increasing leverage of firms; and the near tripling of financial expenses as a proportion of net revenue between 1965 and 1971. The rate of earnings retention has been low for institutional and fiscal reasons as well. Investors in companies listed on KSE demand high cash-dividend payments, and govemment exerts pressure on listed firms to maintain high dividends, even when business conditions are not favorable. Cash dividends of 20 percent on par are quite common. In the 1970-75 period these dividend levels meant fairly high retums to shareholders. Despite the upward drift of stock prices, the average divi- dend yield was 13.8 percent. On an after-tax basis the effective dividend yield averaged between 12.4 and 13.1 percent. Adding capital gains, the effective nominal return for the period rose between 23 and 27 percent a year. Because there are no taxes on capital gains from securities, this nominal return yielded average real returns of 9.3 to 10 percent an- nually.14 Although shareholders gained, the retained earnings of corporations stagnated. Dividend payout ratios, as a share of net profit after taxes, ranged from 40 to 60 percent. Consequently some firms had to use discretionary accounting practices, usually reducing depreciation charges, to justify the payment of high dividends. For owners of privately held firms there has been a tax disincentive to retaining earnings in the 13. Published figures tend to be unreliable, and caution is needed in interpret- ing results derived from Bori surveys of corporate finance. 14. The impact of this exemption on tax revenue has been small because the securities market only recently emerged as a major source of corporate finance. The tax exemption has nevertheless been an effective incentive to the development of the securities market. 386 PART THREE: MAJOR SECTORS business. Fifty percent of retained earnings are construed as having been paid as dividends, and hence as taxable income of shareholders. This practice is a clear encouragement for companies to open up to the public, but it acts as a powerful deterrent to higher eamings retention for the majority of Korean firms which will remain privately held. DEPRECIATION ALLOWANCES. Depreciation or capital consumption al- lowances have not been important as a source of corporate finance and, consequently, of gross corporate savings. Part of the reason for this is the very rapid growth of investment. But even taking this growth into account, depreciation appears to be underused as a potential source of corporate finance. Depreciation allowances accounted for only 11 per- cent of gross sources of corporate finance in 1967-71. They rose to 15 percent in 1972-74. In comparison, depreciation accounted for 20 per- cent of total sources in Japan in 1954467 and 42 percent in the United States in 1946-63.15 One reason for the small contribution of depreciation in Korea is institutional: the legal depreciable lives of equipment have been long. In manufacturing, for example, the average depreciation rate in 1967- 74 was 9.6 percent, the equivalent of a depreciable life of 10.4 years. Even though the August 1972 presidential decree increased depreciation rates on fixed assets of major industries from 30 percent to between 40 and 80 percent, corporations apparently have not availed themselves of this provision. Another reason for the small contribution of depreciation is that firms have not taken full advantage of the 1965 Assets Revalua- tion Law, even in the face of high inflation.16 The explanation usually cited for the relative underuse of the law by publicly held firms is that, as stock dividends are distributed to reflect the higher book value, total cash dividends correspondingly increase. As a result, depreciation allow- ances are understated through the use of discretionary accounting poli- cies and become an inadequate source of funds for future expansion. Privately held firms, on the other hand, have no incentive to revalue assets because they are arbitrarily taxed on the basis of asset size. Both classes of firms are deterred by a 3 percent revaluation tax. 15. n1 SaKong, "Corporate Finance," table 4, p. 13. 16. This law permits the revaluation of fixed assets to reflect replacement costs each year, provided the wholesale price index rises by 25 percent from the date of the last revaluation. The revaluation surplus is taxed at the rate of 3 percent. FINANCIAL SYSTEM 387 Allocating Financial Resources The growth of domestic credit facilitated the allocational decisions of government. Domestic credit rose by 25.8 percent a year during 1963-66 and by 33.9 percent a year during 1972-76. Throughout the period the flows of credit from institutional sources-the commercial banking system and KDB-were concentrated on manufacturing. These flows preceded the shift in the role of manufacturing as a major con- tributor to value added. Manufacturing accounted for 43 percent of institutional credit in 1963-64 and 52 percent in 1975; its share in GNP rose from 11 percent in 1963-64 to 35 percent in 1976. The fact that flows of credit preceded these structural shifts clearly shows that govern- ment policies succeeded in stimulating the development of export-led, labor-intensive manufacturing. But in pursuit of this objective, credit to such sectors as construction, particularly housing, and to community, social, and personal services was perhaps unavoidably reduced. Social overhead activities, such as electricity and transport, also received sub- stantial credits, and this was in line with maximizing industrialization in Korea. Allocational mechanisms The most important mechanisms for allocating funds in Korea are commercial banks and KDB. Although control of the system of institu- tional finance rests with MOp and BOR, some latitude in decisionmaking is allowed. Commercial banks, subject to broad sectoral guidelines, may and do allocate banking funds-that is, funds raised through deposits and the sale of securities-to specific firms and projects. Since the early 1960s banks have used various practices to discriminate among borrowers in response to conditions of excess demand for funds under regimes of credit rationing and fixed interest rates. To supplement the discretionary allocational mechanism, which is based on the system of sectoral guide- lines and has not always been reliable or satisfactory, a strong set of preferential credit schemes ensures that funds reach a target sector at a lower cost than market. In recent years the most visible selective credit scheme has been that of the National Investment Fund (NIF); its programs are directed toward heavy and chemical industries, energy, export finance, and the New Community movement. 388 PART THREE: MAJOR SECTORS DISCRETIONARY LENDING AND INTEREST RATES. Allocational decisions of banks are heavily influenced by a traditional asset-based approach to lending. Banks give more weight to collateral than to the ability of borrowers to service their indebtedness on the basis of demonstrated and projected cash flows. The major institutional factor reinforcing the traditional predisposition for collateral is the prohibition in the Banking Act of unsecured loans having maturities of more than one year. Among the five nationwide and ten regional commercial banks, 98 percent of domestic loans and discounts outstanding in September 1975 were se- cured by various forms of collateral. Real estate has long been the main form of collateral, but recent patterns indicate the increasing use of installment savings to constitute collateral and serve as sinking funds. In general, banks require real estate, comprising land, plant, and equip- ment, valued at 130 to 200 percent of the principal value of loans so secured. The collateral-based lending approach has been complemented by a regime of fixed rates. Until August 1976 borrowers were ostensibly charged one nominal rate, irrespective of their credit standing. Then MOF announced the institution of a prime-rate system for general working-capital loans made by banks. This merely ratified an existing situation: commercial banks and insurance companies had long used various practices to charge different effective rates. Commercial banks collect interest quarterly in advance from nonprime customers; monthly in advance from prime customers. Amortization schedules are accelerated for nonprime companies. Compensating balances are established by the tacit understanding that borrowers will maintain with their banks a pre- determined level of demand, time, and savings deposits.17 Most stringent is the requirement to open installment savings accounts, which are used both as collateral and as sinking funds. When this requirement is applied, banks usually ask borrowers to open a 24-month installment account; payment of the eighth monthly installment usually results in the borrower's obtaining a net amount of less than two-thirds of the approved loan. This practice raised the effective cost of borrowing from a nominal 15.5 percent to 35 percent at year-end 1975 (see table D.25 in appendix D). For insurance companies the approach is straightfor- 17. Borrowers are required to maintain a balance equal to about 20 percent of their outstanding borrowings in demand, time, or savings deposits. Because the full amount of a loan is not available for the borrower's use, the effective cost of a loan increases. This practice is not explicitly permitted by law. FINANCIAL SYSTEM 389 ward. Loan terms usually are one to three years, and interest is collected one year in advance. SELECTIVE CREDIT POLICIES. Selective credit schemes consist of direc- tives to banks regarding the composition of their portfolios and loans to banks earmarked by government for specific projects and sectors. These schemes constitute an important part of institutional credit: they accounted for 31 percent of the increase in institutional credit to indus- try in 1965-71 and 43 percent in 1971-75 (see table D.26 in appendix D). In both periods the share of export loans, mostly short-term produc- tion loans against letters of credit, was large. In 1974 and 1975, how- ever, manufacturers relied heavily on these loans to stockpile raw ma- terials as a hedge against price increases resulting from the 1974 devaluation and to finance inventories of finished goods which piled up as export demand slowed down in the later stages of the recession. These uses of funds were sanctioned by government and in mid-1975 were intended to stimulate the economy. The large share of preferential credit extended to manufacturing reflects government policy to support manufactured exports by providing sufficient incentives to ensure international competitiveness. In 1971-75 export and equipment loans to manufacturing companies were extended at nominal rates 3.5 to 8.5 percentage points below the general banking rate. The effective interest-rate subsidies were even larger because finan- cial institutions could not add on as many charges to these loans. Sub- sidized credits in this period amounted to 44 percent of the W1,375 billion in institutional credit to manufacturing. Most of these credits were for terms longer than one year. Assuming a 5 percent effective interest subsidy, the savings to manufacturing firms in aggregate would have been at least W30 billion. These credits were extended to export- ing firms, particularly those in textiles, clothing, and electronics, and to firms engaged in the machinery industry for their purchases of new equipment. The agricultural and construction sectors also received large amounts of subsidized funds relative to the total increases of institutional credit. The share of agriculture in the increase in subsidized credit declined from 54 percent to 41 percent over the 1971-75 period, but this may only have been symptomatic of the substitution of direct subsidies for credit by Grain Management Fund purchases and Fertilizer Fund oper- ations. It therefore seems that to evaluate incentives to agricultural production, the complementarity of budgetary subsidies and preferential credit schemes must also be considered. Subsidized equipment finance 390 PART THREE: MAJOR SECTORS for housing rose substantially in the 1971-75 period. This financing from commercial bank funds, raised through deposits and sales of se- curities, rose quickly in the 1972-75 period; the amounts extended under this program before 1972 were negligible. Housing loans from govern- ment funds also doubled in 1971-75. Because about one-half of subsidized credits were from general bank- ing funds, not from government funds, deposit-money banks were unable to earn sufficient spreads to cover the high cost of administering these loans. Consequently BOK has been paying interest on reserve de- posits and special accounts to compensate the banks. This practice en- abled banks to double their effective spreads and pay 15 percent divi- dends on par to shareholders other than government. Because this dividend was equal to the deposit rate, investors treated bank shares as quasi fixed-income securities. On the whole, selective credit schemes have been successfully imple- mented in Korea. This success can be seen in the alteration of the composition of aggregate output: that is, in the shift in industrial struc- ture, the export-led development of manufacturing, and the rapid growth of investment and output in manufacturing. This ultimate objective was accomplished by directing the flow of sufficient credit to manufacturing and by reducing the cost of funds to these sectors. Because the selectivity of fund flows extended down to the level of individual firms, the ad- vantages of the selective credit schemes accrued only to a small number of business groups and the companies affiliated with them. Korea's selective credit policies therefore succeeded in altering the composition of output, but at the cost of the equitable distribution of credit. The benefits of growth appear to have outweighed this cost. THE NATIONAL INVESTMENT FUND. Government instituted NIF in Jan- uary 1974 as a temporary instrument to correct the structure of the financial system by improving the flow of domestic savings into invest- ments in heavy and chemical industries. Under MOF jurisdiction, NIF is in effect managed by BOK. Its funds are mainly derived from direct government contributions, the compulsory purchase of NIF bonds by financial institutions, and loan repayments. Its loans may be used for fixed capital, working capital, and export finance. As the financial system and the securities market develop, the need for direct intervention of government should diminish, as should the role of NIF. Government set certain boundary conditions in establishing NIF. The first such condition specifies that NIF lending terms be in line with international terms so as not to impair the competitiveness of capital- FINANCIAL SYSTEM 391 intensive industries. Another boundary condition ensures NIF'S operating flexibility: as a special government fund, it is not administered as part of the government budget, as were earlier programs of government to support loans to industry. Instead, NIF supplies funds through existing banking institutions, which in turn appraise projects and administer the loans. NIF is thus an intermediary and a guarantor; it acts to convert savings, which would otherwise be directed to short-term loans, into guaranteed medium- and long-term loans. For relending, banking insti- tutions receive a spread of only 1.5 percent. As a result, commercial banks have been reluctant to handle NIP loans, and the bulk of NIF operations have to be coursed through KDB. In the future it appears that NIF will be KDB'S most important source of domestic funds. Financial Savings under the Fourth Plan Korea's fourth plan calls for higher levels of internal savings by the corporate sector and larger financial surpluses of the household and unincorporated enterprise sector net of its real investment needs. These surpluses are to be transferred to the corporate and government sectors through the intermediation of the financial system. Two major policy measures will be used to accomplish these objectives: high real rates of return on financial assets to stimulate household savings; and more efficient allocation mechanisms to ensure that funds are directed to the most economically beneficial and financially profitable projects. The plan projects that the retained earnings of the corporate sector will increase from 2.9 percent of GNP in 1976 to 3.9 percent in 1981 and that depreciation will increase from 7.9 percent of GNP to 9.4 percent.'8 Treating all depreciation as being incurred by the corporate sector, gross corporate savings are expected to rise from an average of 6.8 percent of GNP during 1968-75 to 13.3 percent in 1981. Consequently, corporate deficits are projected to decline to 5 percent of GNP by 1981, a fairly substantial reduction from 10.2 percent of GNP in 1975 and 6.5 per- cent in 1976. For the government sector, the plan projects that financial deficits will persist, but that these will decline continuously to the man- ageable level of 0.1 percent of GNP by 1981 (table 12.11). Among the surplus sectors, households are projected to perform a 18. Economic Planning Board (EPB), The Fourth Five-Year Economic Devel- opment Plan, 1977-81 (Seoul, 1976). 392 PART THREE: MAJOR SECTORS Table 12.11. Financial Surpluses and Defcits, by Sector, 1977-81 Percentage of gross national product Sector 1977 1981 Financial 0.2 0.3 Government -0.7 -0.1 Corporate -6.6 -5.0 Household 5.0 6.8 Rest of the world 2.1 -2.1 Financial surplus = deficit 7.3 7.2 Note: Includes net transfers from the ROW sector to the government and household sectors. See table 12.8. Source: EPB, Fourth Plan. major role in financing the deficits of the government and corporate sectors, particularly because the ROW sector is projected to become in- significant as a source of funds. It is envisaged that the financial surplus of households and unincorporated enterprises will steadily increase, rising from 4.5 percent of GNP in 1976 to 6.8 percent in 1981 and averaging 6 percent of GNP during the plan period. This substantial increase will depend on the success in achieving real savings targets, which imply a marginal saving rate of 35 percent for the plan period. At these levels financial surpluses would on average represent 55.2 per- cent of sector savings during the period. These surpluses are intended to supplant foreign capital and cover 68.3 percent of deficits of all other domestic sectors in 1977 and 95.4 percent in 1981. To achieve the plan's targets for corporate savings, institutional and tax hindrances will be reduced. First, the plan proposes that depreciation rates be increased. To implement this, the tax deliberation committee convened by MOF has recommended that depreciation allowances in any one year could be matched by an equal amount of special deprecia- tion reserves. This provision effectively doubles depreciation rates on the straight-line method. FOT small- and medium-sized firms, special replacement reserves are proposed as charges to income in addition to normal depreciation expenses. Second, to increase the retained earnings of privately held corporations, the tax deliberation committee has recom- mended that the proportion of retained earnings subject to personal income tax as presumptive income of the private owners be lowered. This provision should reduce the extent of double taxation on income of closed corporations and their owners. F1NANCIAL SYSTEM 393 To achieve the plan's targets for mobilizing household financial sav- ings, government authorities contemplate a series of measures that essen- tially would make financial assets more attractive relative to other forms of asset accumulation, particularly real estate. First, real interest rates on time, savings, and other deposits are to be maintained at a minimum level of 5 percent. Government authorities intend to accomplish this by keeping inflation in check and by making selective increases in nominal interest rates similar to those implemented in August 1976. Inflation, measured by the consumer price index, is projected to abate from the 15 percent level that prevailed in 1976 to between 7 and 10 percent during the plan period. If this occurs, nominal time-deposit rates of 15 to 18 percent-the levels prevailing in mid-1976-would result in real rates of about 5 to 8 percent. The stress on high real rates of return is fully justified by previous experience. Second, securities are to become more important as channels of savings. Government intends to increase the supply of equity issues by encouraging closely held firms to open up to the public and by encouraging all corporations to issue more bonds. MOF is planning an attractive package of incentives to investors, featuring high nominal-dividend yields, an upward-sloping bond-yield curve, and a more varied array of maturities and kinds of securities. On the supply side, MOr is planning to allow listed firms to sell rights issues of new shares at market prices, not at par value as under the present system, to increase the net proceeds from rights issues. For the allocation of financial resources during the plan period, greater emphasis is to be placed on furthering the growth of heavy and chemical industries, exports, small- and medium-sized firms, and housing: * Growth of the machinery industry is to be encouraged through loans from NIF and commercial banks. These two sources are pro- jected to account for 43.9 percent of the domestic finance require- ments of this industry during the plan period. Internal savings and sales of new stock are to provide the rest. * In shipbuilding, the bottleneck to rapid growth appears to be the lack of export credits to purchasers of ships, not the financing of investment in shipbuilding capacity. The Korea Export-Import Bank, established as an independent entity in July 1976, should facilitate the sale of ships on credit. * Export financing, on the whole, is well developed in Korea. Most schemes, conventional and innovative, have already been imple- mented, and many of these are to be expanded during the plan period. 394 PART THREE: MAJOR SECTORS - The oldest mechanism providing finance to small- and medium- sized firms, the Medium Industry Bank, has been in operation since 1961. One of its divisions, the Credit Guarantee Fund, was estab- lished as a separate entity in July 1976. Capital is to be increased to this institution to enable the expansion of its operations. In addi- tion, the emphasis of commercial lending is to shift to smaller enterprises. Past inattention to housing requirements will begin to be rectified under the fourth plan with the proposed construction of 1.3 million new dwellings, mostly apartments. The cost of construction is pro- jected to be W2,519 billion in 1975 prices, of which W600 billion will be directed to the construction of 500,000 units by the government-owned Korea National Housing Corporation. Major Issues of Financial Policy The fourth plan's proposals for the corporate sector should result in some improvement of corporate saving rates. But it is quite important that they be supplemented by other measures: moving away from dis- cretionary accounting policies to encourage firms to charge depreciation regularly; eliminating the pressure on open corporations to pay high dividends; and possibly lowering the tax rate on the revalution surplus. In addition, a thorough review of legal depreciation rates should be made to enable their being updated. Other fiscal incentives, such as more com- prehensive investment tax credits, should also be studied. For the household sector the fourth plan's stress on savings instru- ments is sound. Thirty percent of the projected gross acquisition of financial assets by the household sector during 1977-81 is expected to be in time and savings deposits and 37 percent in securities (table 12.12). Money (Ml) is projected to account for only 12 percent, about three-fifths of its share in 1972-75. Although this expansion is con- sistent with cautious monetary policies, it does not seem realistic. Pres- sure is emanating from various manufacturing sectors for a loosening of credit so that planned investments can be financed. It is likely that M1 will account for a larger share than envisioned. The share of securities may be as large as projected if government efforts are successful to develop the capital markets as significant sources of direct corporate finance and to continue implementation of the 1972 Public Ownership Inducement Law and other capital-market development programs. But if the share of M1 is increased, the share of securities would decline. FINANCIAL SYSTEM 395 Table 12.12. Gross Financial Investments of Households, 1972-81 1972-76, 1977-81 Billions Percentage Billions Percentage Source of won of total of won of total Money (M1) 574.4 18.0 1,353.3 12.3 Time and savings deposits 1,102.2 34.5 3,314.0 30.1 Other deposits and loans 277.8 8.7 1,376.3 12.5 Insurance and trust 208.1 6.5 831.5 7.6 Securities' 1,024.7 32.1 4,060.4 36.9 Miscellaneous 4.8 0.2 60.5 0.6 Total 3,192.0 100.0 10,996.0 100.0 Note: All figures are in current prices. Sources: Box, Flow of Funds in Korea, 1963-70; Box, Economic Statistics Year- book, 1973 and 1976; EPB, Fourth Plan. a. Investments in 1976 were estimated by EPB. b. Includes equities other than stocks. Finally, the relatively smaller share of time and savings deposits-the result of a slower rate of growth projected at 26 percent a year in 1977-81 compared with 29.3 percent estimated for 1972-76-puts into perspective the impact of the Wage-earners' Savings Scheme instituted in April 1976. Its impact appears to be greater on income distribution, providing higher returns to savers not having access to other forms of savings, than on resource mobilization. It would appear that the plan's targets for household financial savings can be achieved if aggregate savings targets are met. What, then, are the prospects for aggregate savings? The mildly upward trend of finan- cial surplus as a proportion of GNP indicates that the financial resources mobilized have been less than their potential. The financial surplus ratio rose from 2.9 percent of GNP in 1965 to 5.2 percent in 1975. This general trend nevertheless disguises peaks of 4 percent in 1969 and 7 percent in 1973 in response to reforms introduced in 1965 and 1972. The subsequent retreats from these peaks point to the need for sustained programs of resource mobilization to reinforce gains achieved from re- forms in interest rates and financial structure. The policy implication seems to be that government must mount a comprehensive, continuing program for the mobilization of financial savings. At the core of such a program would be sustaining the higher real yields on time and savings deposits, insurance policies, and securities. One further implication of the movement of the financial surplus ratio is that additional headway could be made if more emphasis were given to promoting forms of 396 PART THREE: MAJOR SECTORS contractual savings, such as whole-life insurance policies and pension schemes.19 Patterns of financial resource allocation Korean decisionmakers and economic planners want to be certain that resources are channeled to designated projects and used as efficiently as possible. This concem at the macroeconomic level has resulted in gov- ernment's pervasive control of the entire financial system. Financial flows have consequently been passive, following or ratifying the investment decisions of government authorities mapping the direction of indus- trialization. Finance has been plentiful for exports, for firms that are heavily export-oriented, and for the short-term finance of large firms. Although this allocational strategy conformed to the objectives of pre- vious plans, it also resulted in a lack of adequate financing for such sectors as small- and medium-sized firms and housing. Various measures are proposed in the fourth plan to balance the allocation of funds be- tween competing sectors and yet maintain the momentum of manufac- turing development generated under previous plans. Although the fourth plan presents only the aggregate investment outlays by sector, not the manner in which these outlays are to be financed, it can be assumed that the broad sectoral credit allocations will not deviate substantially from those made during 1972-76. In manufacturing the financing patterns call for sizable infusions of foreign loans and investments, which are to account for 31 percent of fixed investment outlays and 6 percent of working-capital requirements. These requirements are concentrated in the heavy and chemical indus- tries, for which 38 percent of the financing requirements are to be supplied by foreign loans. The financing pattems for the other industrial sectors are similar. According to estimates for major investment projects, foreign flows are projected to be bunched in 1977-79. If these schedules are met, the negligible share of foreign savings projected by 1981 would be consistent with pattems of real industrial investment. Otherwise the plan's targets for lower shares of foreign financial savings would not be met. One-half of the domestic funding requirements in manufacturing is to be provided by NIF, which is essentially a set of portfolio directives 19. If there were no conflicts with objectives to improve employment and income distribution, a more effective way of mobilizing household savings might be to extend social security to workers in private industry. FlrANCIAL SYSTEM 397 to the banking system, and by loans from domestic banks. These amounts from NIF and bank loans would in turn account for 50 percent of the expansion of domestic credit projected for 1977-81. This level is consistent with the share of manufacturing in incremental domestic credit of 49 percent in 1967-71 and 60 percent in 1972-75. Nip's practice of allocating funds to eligible industries on an industry- by-industry basis appears to reduce its flexibility. Banks receive their allocations from NIP on an annual basis, but disbursements are linked to disbursements for individual projects. To a certain extent banks are free to select borrowers from among the industries specified in their annual NIp programs; but they also have specific projects allocated to them by the NIF council. In such cases banks usually approve loans after confirming that project proposals comply wvith NIF's selection criteria: soundness of financial structure, appropriateness of loan size, maturity, repayment prospects, and project and funding plans. Uncommitted or undisbursed funds allocated to any one industry cannot be transferred to other eligible industries. This seems unduly restrictive in view of the uncertainties about the availability of financial resources, particularly foreign funds. If allocations were reviewed semiannually to enable transfers from one industry to another, this might inject the flexibility required to meet rapidly changing circumstances. Government is aware of the need to improve the discretionary alloca- tional system and thereby ensure that interest rates have a larger role in allocating funds. With the August 1976 measures, authorities have taken a step toward a prime-rate system, which will enable banks to discriminate among borrowers according to risk, and an upward-sloping yield curve, which will encourage banks to lend long-term because of greater profitability. It is clear, however, that the one percentage point difference between prime and nonprime borrowers is unrealistic and unlikely to be observed in practice. Banks will continue to use discre- tionary practices to impute the correct difference in rates for prime and nonprime companies. Should positive real interest rates be maintained? An across-the-board increase in rates perhaps is not necessary to improve allocational effi- ciency. Instead, changes in nominal rates, added to the reforms instituted in August 1976, are needed to reflect present rate differences.20 A com- prehensive prime-rate system would accomplish this. Borrowers, espe- cially smaller firms, would then know the true cost of finance. Long-term 20. Nominal rates were changed in 1978. 398 PART THREE: MAJOR SECTORS rates also need to be increased to stimulate long-term loans. These recommendations are couched in the recognition that excess demand for bank loans, even at the high effective levels of general banking rates, indicates that interest rates are in disequilibrium. Excess demand will continue so long as loans are subsidized for various undertakings and strong selective credit policies favor certain projects and borrowers over others. Because these subsidies form part of a complex incentive system to maintain the competitiveness of Korean exports in world markets, some disequilibrium rates will necessarily prevail. The task of allocating funds in the fourth plan period is thus complex. First, notwithstanding the broad measures being taken by government to stimulate aggregate domestic savings in general and financial savings in particular, there is substantial doubt about whether savings targets can be met. Second, the allocation problem is likely to become acute because of government's intention to restrain inflation to a manageable level of 8 to 9 percent annually. If this is to be accomplished, the growth of domestic credit must be restricted. The fact that the proposed increase in domestic credit of only 19 percent a year during the fourth plan period may be low relative to the needs of both the private and public sectors further complicates the tasks associated with credit allocation (table 12.13). Strengthening financial institutions The mobilization and allocation of financial savings in Korea, al- though reasonable, have not been ideal. Their effectiveness could be Table 12.13. Growth of Assets and Liabilities of the Banking System, 1972-81 (percent) Annual growth rate Item 1972-76" 1977-81 Money (MN) 34.0 21.3 Quasi money 29.1 26.1 Time and savings deposits 29.3 26.0 Money (M2)b 30.8 24.5 Domestic credit 34.2 19.0 Note: Growth rates were based on current prices. Source: Data supplied by EPB. a. Assets and liabilities for 1976 were estimated by EPB. b. M2 is M1 plus time and savings deposits. FINANCIAL SYSTEM 399 improved by instituting reforms intended to increase the efficiency of institutions engaging in these efforts. Formulating such reforms, given the constraints associated with mobilizing and allocating financial re- sources during the fourth plan period, is important for the long-term development of the financial system. BANKING. Because lending limits are set at three times the deposit base, the main objective of managers of commercial banks has been to increase the base of time and savings deposits. And because bank man- agements know that BOK will supplement income, the costs associated with raising resources and operating banks have not been overriding concerns. In addition, bank profitability depends more on BOK'S policies governing rediscount facilities, interest-rate structures, and various re- serve requirements than on the inherent capability of bank managements to keep operating costs low relative to the volume handled. Because banks could not pay market rates for deposits, the noninterest cost of deposit mobilization-associated with maintaining extensive branch net- works, handling demand deposits, advertising, and contributing to or- ganizations promoting savings-has been high. The cost of administering the various selective credit mechanisms has also been high. Conse- quently the spread between the average return on bank loans, dis- counts, securities, and other deposits, and the average return on deposit liabilities and debentures issued, have had to be large. According to BOK the rate structures and operating costs of banks require a 6.02 percent spread to guarantee coverage of all expenses and dividend pay- ments of 15 percent on par to shareholders other than government. Under these circumstances it is clear that the influence of market forces on the managerial performance of banks has been and will continue to be limited. Technical banking skills in Korea compare favorably with those in most developed countries, and adherence to contractual obliga- tions and commitments is high. But bank managements are weak in introducing innovations and reducing the cost of operations. There is no incentive to do either; profit maximization is not a goal. The establishment of the limited prime-rate system in 1976 was a significant step toward bank lending based on cash flows. As the system is accepted and its implementation is broadened to other classes of loans, credit standards should be strengthened. The almost exclusive reliance on collateral for loans should diminish accordingly. As an interim meas- ure BOK began implementing a bank rationalization program in 1976 to upgrade the operational capabilities of Korean commercial banks. This useful exercise was supplemented by the recommendations of the finan- 400 PART THREE: MAJOR SECTORS cial markets deliberation committee convened by MOr to study the rationalization of Korea's financial markets. Part of the program was the internationalization of the five nationwide banks by increasing their capital. This action, completed in August 1976, provided Korean banks with a larger operating base and a stronger position for international transactions. BOK has also established a principal-banker system for 526 large firms. This system is intended to improve credit information sys- tems, enable monitoring the level of bank exposure to these firms, and facilitate syndications of large loans. MONEY MARKET. The original purpose for setting up short-term finance companies (sTFCs) was to absorb the unorganized money market (UMNI) into the organized financial system. The volume of new loans made by sTFCs rose from W2.1 billion in 1972, or less than 1 percent of total transactions of the financial sector, to W62.2 billion in 1975, or 3.4 percent of total transactions. Netting out foreign claims and debts, the share of sTiCs in domestic transactions was 4.3 percent in 1975, a figure made more remarkable by the fact that sTiC resources are raised from individuals and corporations, not from government. A subsidiary purpose for establishing STFCS was to create a core of underwriters to handle the origination and placement of corporate bond and stock issues. This has not been achieved. Money market operations have been so profitable that sTFCs have concentrated on these activities and neglected their underwriting role. Competition among sTiCs has been over the amounts and other conditions on which notes are discounted, the time taken to process loan requests, the reciprocity arrangements with major shareholders, and the convenience of office locations.21 The absence of price competition- sTFCs have been operating a cartel through the sTFC Association since 1973-has resulted in the deterioration of credit standards, a process that has continued as STFCS have sought to increase their business vol- ume by reaching out to marginal firms. 21. Commercial banks have been able to obtain money market licenses since October 1975, but the volume of transactions handled by these banks has been small relative to that of sTFcs. Another source of competition in the money market emerged with the entry of merchant banks. The first merchant bank, Korea Merchant Banking Corporation, began operations in September 1976, and others have followed this lead. Because their activities in the money market are limited in amount by law, the extent to which they will be competitive in the money market is not yet apparent. The risk of overcompetition nevertheless persists. 5IPNANCIAL SYSTEM 401 Aware of these problems, Mop has increased reserve requirements for time deposits and other liquid assets from 10 percent of total short-term liabilities, net of standby credit facilities, to 15 percent. These require- ments are based on the recognition that STFC activities have evolved into quasi banking. STFCs act mainly as principals by borrowing on their own account and relending to industrial companies. They act only to a minor extent as agents by selling the paper of these companies to the public without recourse. The reasons for this major failing of STFCS stem pri- marily from the fact that information on industrial companies is lacking or inadequate. Most enterprises do not systematically provide informa- tion to the public; if they do, accounting practices tend to be discre- tionary, which makes it difficult to determine corporate financial posi- tions with accuracy. In addition, public accounting and auditing are not yet well-established traditions in Korea. Although the accounting situation is improving, it is difficult to foresee major changes in STFC operations in the near future. Sales of commercial paper without re- course will continue to lag significantly behind sales of sTFC paper be- cause sales without recourse are sustained mainly by the demand of major shareholders for paper issued by their companies, not by strong underlying demand by independent investors. SECURITIES MARKET. The expansion of industry and the shift from light to heavy manufactures require stronger corporate financial struc- tures capable of supporting large investment programs that take longer to generate returns. Government recognizes this and correctly argues that the need for long-term finance can be met principally by having the securities markets deal in long-term securities. Despite the advances made by the securities market in the 1970s, it still suffers from one inhibiting factor: prices are not related to operating performance.22 Most investors perceive stock prices in relation to par, not as equilibrating mechanisms, and dividend rates are fixed as a per- centage of par. With par values and dividend rates so fixed, the fluctua- tion of stock prices has been influenced not by earnings prospects, but by expectations of distributions of free shares or stock dividends. These free shares arise from the transfer of retained surplus into common stock, which must then be paid out or distributed to owners of record. 22. This subject is treated in II SaKong, "Some Observations on the Securities Market of Korea," Korea Development Institute Working Paper, no. 7410 (Seoul, 1974). 402 PART THREE: MAJOR SECTORS Retained surplus refers not only to the accumulation of earned surplus, but to the increase in equity arising from the paper revaluation of fixed assets. In ensuing years the previous dividend rate is applied to par to determine cash dividends. Total cash dividend payments increase accordingly, even though there is no increase in eamings. Thus cash generation for additional dividend payments is often made at the expense of adequate provision for depreciation. This unduly increases the cost of external equity. The rigid par-dividend system also prevents previously listed corpora- tions from obtaining correct prices for new issues. Korean law stipulates that rights issues-that is, sales of new shares by previously listed firms- can be offered only at par or at a price stated as a certain premium over par, neither of which bears any relation to current market prices. Thus a profitable company cannot take advantage of the favorable price- earnings ratio that its performance warrants. Under the fourth plan, authorities intend to move away from this outdated system and allow listed firms to sell new shares at market value. This development should speed up the sound development of the securities market and result in rational expectations of corporate issuers and investors. MORTGAGE MARKET. The Korea Housing Bank (KHB) is the only in- stitutional source of mortgage and construction finance in Korea. To finance government construction and private mortgages during the fourth plan period, a larger volume of national housing bonds will be sold. These bonds are KHB'S second most important source of funds, after de- posits. To facilitate purchases, new mortgage systems are to be intro- duced, and the installment loans provided by KUB are to be expanded. It is also envisioned that firms located in certain industrial estates will be permitted to charge the housing of employees as a legitimate business expense. In addition to these measures, government should consider more com- prehensive mortgage schemes, including mortgage insurance, to reduce KHB'S downpayment requirements of 50 to 70 percent for low- and middle-cost housing. It must be stressed, however, that the provision of housing finance should be firmly based on the principle that mortgage credit is not to be subsidized: government should avoid preferential rates on mortgage and construction loans. The emphasis should be on making funds available at market rates, reducing downpayments through mortgage insurance and similar schemes, raising loan-to-value ratios, and lengthening the maturities of mortgage loans to provide manageable levels of debt service. Other mortgage schemes-such as those which minimize the burden of servicing mortgages by pegging interest or FINANCIAL SYSTEM 403 principal payments to the consumer price index or by accelerating pay- ments to accommodate the higher income earned later in life-could also be implemented.23 LIFE INSURANCE. The small share of life insurance in gross assets ac- quired by individuals in Korea stems from the unattractiveness of re- turns from policies with savings features. In the 1970s these policies yielded nominal rates of 11 to 13 percent, or much less than alternative forms of investment. In addition to low returns, operational inefficiencies and marketing problems have stymied the life insurance industry in Korea. In the 1960s life insurance companies marketed policies that were pure savings annuities. Subscribers paid in a fixed amount each month or quarter, but were provided with none of the protection of normal insurance policies. The returns were reduced further by high front-end sales charges, which often meant that subscribers could not recover the full amount of principal contributions. Consequently there was disenchantment with annuities, and insurance companies stopped promoting them as strongly as before. In addition, sales forces have been inadequately trained. Buyers of life insurance were often convinced to take larger than warranted coverages, which meant that premium pay- ments could not be sustained. The lapse or cancellation ratio of life insurance in Korea ranges from 40 to 80 percent in the first year, and is much higher than international standards of about 15 to 30 percent. This is particularly unfortunate in view of the fact that Korea is one of the few LDCs to introduce successfully the concept of industrial in- surance, under which premiums are collected monthly to reduce the burden of carrying insurance. MOF should consider ways and means to improve returns from life insurance policies by minimizing sales charges deducted at the start of the coverage. One possible solution would be to professionalize the insurance sales force by requiring agents to be full-time, not part-time. Selling techniques could be improved. Agents could be trained to assist policyholders in determining adequate insurance coverage and selecting insurance packages that yield sufficient financial returns and still provide basic protection against normal risks. The financial markets deliberation committee convened bv MOF is examining the problems of the Korean life insurance industry. Based on the results of this study, a new in- surance code should be promulgated to modernize the industry. 23. See J. Black, "A New System for Mortgages," Lloyds Bank Review, January 1974, pp. 9-16. PART FOUR Appendixes Appendix A Syamaprasad Gupta Macroeconomic Model THE MACROECONOMIC MODEL was designed to investigate the long-term implications of alternative growth strategies for the economic aggregates, the structure of production, and the growth of incomes for different occupations and income classes. In addition, the model was used to examine the consistency of projections in Korea's fourth plan. Taking the plan's assumptions for exports, the pattern of investment allocations, fiscal parameters, and government expenditure, the model examined the feasibility of aggregate targets for savings and investment and projections of growth and employment. This model drew from two other macro models of Korea. One, by Irma Adelman and Sherman Robinson, is a detailed general-equilibrium model designed to investigate the effects of alternative policies on the size distribution of income in the short to medium run.' The income distribution block of the macro model de- scribed here drew heavily from the Adelman-Robinson model. Similarly the 53-sector interindustry model developed bv Kim Yoon Hyung was used in the derivation for each sector of incremental capital-output ratios, which play the central role in the investment block of this model.2 The Korea Development Institute (KDI) was associated with the formu- lation of this model and was extremely helpful in preparing most of the inputs required for the sector classification. The model is thus closely related to the macro model used in the preparation of Korea's fourth plan. 1. Irma Adelman and Sherman Robinson, Income Distribution Policy in Developirng Countries: A Case Study of Korea (Stanford: Stanford University Press; Oxford: Oxford University Press, 1978). 2. Yoon Hyung Kim, "53-Sector Interindustry Projection Model for Korea: 1974-81," Korea Development Institute Interim Report, no. 7505 (Seoul, 1975). 407 408 PART FOUR: APPENDIXES Structure This is a 17-sector dynamic input-output model which combines na- tional accounts with input-output accounts and flow-of-funds accounts, using information from income and expenditure surveys. The relations of the model are dynamic and simultaneous. Its dynamic relations operate intertemporally through the functional relations between capital- stock investments and output (physical capital) and between population changes and migration (human capital). Its simultaneous relations operate mainly through prices. The real income of a sector is dependent on the sectoral cost of living and on prices. The sectoral cost of living depends on the sectoral consumption basket, which in turn depends on sectoral real income-that is, the distribution of income in the economy. Similarly, sectoral prices are dependent on wage costs, nonwage costs, tax rates, exchange rates inclusive of tariffs, and FOB import prices in local currency. Wage costs, nonwage costs, and exchange rates are de- pendent on the cost of living, sectoral prices, and the expenditure structure of the economy. In addition, desired and actual imports are equilibrated through changes in exchange rates, which affect both sec- toral prices and the distribution of income. The model has the following special features: * The supply-demand balance for each specific sector is ensured at 1973 prices. * Prices are estimated in the model by cost of production considera- tions, that is, long-term price trends. * Consumption in the model is income-elastic and estimated separately for each occupation, income class, and item. The consumption effect of changes in relative prices works through income shifts between different classes. * Income distribution is a function of changes in product prices, factor prices, sectoral growth, population growth, and dependency ratios in different occupation groups and income classes. * The volume of investment is endogenous and is a function of do- mestic and foreign saving. Investment allocation between sectors is partly a policy variable and, in a few sectors, is export-demand oriented. * Saving and consumption propensities are derived from family budget surveys and are not a residual, as they frequently are in national income accounts. MACROECONOMIC MODEL 409 * The wage rate is a function of productivity growth, a long-run influence, and cost of living and unemployment, short-run in- fluences akin to the Phillips curve. To summarize, the overall structure of the model is intersectoral. Current production, imports, and investment are determined by the use of input-output analysis; the loop for production, income distribution, and consumption is closed, as are the population-employment and flow- of-funds loops. The investment loop is partly open, as for the sectoral allocation of aggregate investment, and partly closed, as for total invest- ment which is constrained by savings. The export loop is open; the import loop is open when imports are estimated as a residual and closed when the import matrix is used. The model is primarily long-term in its perspective, but is not a terminal-period model. Results are simulated on a year-to-year basis, which insures feasibility and consistency. The simplified flow chart gives a summarized version of the flow relations among different economic activities in the model (figure A.1). Capital stock (KT) and the supply of labor (LAB FOR) determine pro- duction (GRoss OUTPUT) under a given set of technological conditions: input-output matrix (INP-OUT MATRIX) and production function (PROD FUc). Because the model is based on fixed capital-output relations, the level of production activities at a given period may absorb the whole labor force or result in unemployment (LAB UNM). The value of gross output net of intermediate payments is distributed between factors through the factor market (PAC MAKT) and enters as income in the household (Iio), corporate (ci), and government (GI) sectors. These sectors spend part of this income (CONS) and save the rest (Hs, Cs, and cs). The sum of these savings is total domestic savings (DS). Foreign savings (Fs), which is exogenously specified, equals the net import surplus of the country. The sum of domestic and foreign savings deter- mines total investment (INV). The total supply of goods and services of the country (DOM SUPPLY) is given by gross domestic output and im- ports (MT). The supply is matched in each sector by total demand at prevailing market prices; total demand comprises consumption (CONS), investment (INv), and exports (ExP). The capital stock of the next period (KT1) is the sum of the current period's capital stock (KT) and net investment of this period. The total labor force is estimated from population growth (pop) and from the age structure and participation rates (AGE PART). The model comprises seventeen sectors, which are listed before the equations at the end of this appendix, and eight blocks of relations Figure A.1. Sihplified Flow Chart of the Model's Relations I~~~~ u I F I F HOI r AGE PART = Age structure and GROSS = Gross output participation rates OUTPUT GS = Government savings CAP MAKT = External capital HOI = Household income market Cl = Corporate income HS = Household savings CONS = Consumption INP-OUT = Input-output matrix MATRIX CS = Corporate savings INV = Gross investment DOM SUPPLY = Domestic supply KT = Capital stock DS = Domestic savings LABOR = Labor force EXP = Exports LAB UNM = Unemployment FAC MAKT = Factor market MT = Imports FS = Foreign savings POP = Population GI = Government income PRODFUC = Production function 410 MACROECONOMIC MODEL 411 determining production, employment, income distribution, consumption and savings, public finance, prices, exports and imports, and capital accounts. These blocks are interrelated. The main driving force of the model comes from investment. Financing investment is partly domestic and partly international. External capital flows are exogenously specified. Domestic savings are largely governed by saving behavior and public policy. Household and corporate income and savings are influenced by changes in fiscal policy, price policy, and income distribution. Exports and external capital flows determine import availability and the level of feasible output. Production block Production is exogenous in three of the model's seventeen sectors: food grains, agriculture other than food grains, and mining. These sec- tors are largely affected by climate and other noneconomic factors. Out- put growth in these sectors is exogenously specified. Incremental production in the remaining fourteen sectors is determined by past investment, using fixed incremental capital-output ratios (icoRS) and assuming different gestation lags in each sector. The adjusted ICORs are estimated from unadjusted ICORs derived by KDI and EPB. Investment by destination is partly endogenous and partly policy-determined. Total investable funds determine aggregate investment and are the aggregate of domestic and foreign savings. The first charge on available funds is made for working capital, replacement investment, and fixed investment required for growth of the agriculture and mining sectors. Residual in- vestment is allocated to different sectors by exogenous allocation param- eters, initially based on the fourth plan's investment pattern. If any sector's output is to be increased, because of increased export targets or increased domestic requirements, the allocation parameters of that sector are increased endogenously on a requirements basis. Investment by source is derived from investment by destination by means of a normalized capital-flow matrix. Employment block Employment in each sector is determined by employment elasticities in each sector in relation to changes in sectoral value added. This assumption, in conjunction with the assumption of fixed icoRs, implicitly gives a production function with increasing substitution of labor by capital. Employment estimates are separately derived for wage earners 412 PART FOUR: APPENDIXES and the self-employed. The total population and working population are estimated exogenously. The model uses separate dependency ratios for urban and rural households in the base period; the rate of change of the dependency ratio is assumed to be the same in rural and urban areas over the projection period. Dependency ratios are functions of base-period dependency ratios derived from household surveys, of the index of their change, and of the number of unemployed allocated between sectors. Outmigration from rural areas is estimated on the basis of an econometric relation derived from time series data. The rate of migration is assumed to be a function of growth in the nonagricultural sectors, with some distribution lag. The natural increase of the farm population is at the average national rate, but the total is reduced yearly by outmigration to the urban sector. Income distribution block The income of the rural sector (agriculture and mining) is the income accruing to the rural sector net of depreciation, indirect taxes including customs duty, and other nontax levies. Estimates of the incidence of all these taxes have been drawn from past observations. Per capita income in the rural sectors is derived from the model's estimates of the number of wage earners and the self-employed and of the dependency ratios in each of these sectors. The distribution of the average income of a class is estimated using the variance derived from the base period data.3 In the base period a lognormal distribution has been fitted for each sector and occupation class. Wage rates in each sector respond to changes in the cost of living, the level of unemployment, and the labor productivity of each sector. The speed of adjustment of wages to changes in the cost of living is sensitive to the rate of underemployment. The elasticity of wage changes in relation to productivity changes is a policy variable. Real wage changes are derived by deflating money wages by the cost of living, using separate indexes to reflect rural and urban patterns of consumption. By rough approximation it is assumed that the rural sector is exclu- sively identified with agriculture and mining and that all other sectors belong in the urban category. The income of the self-employed in each urban sector is derived by subtracting wage payments from the total net value added in each sector. Of the unemployed labor force, 60 percent 3. Adelman and Robinson, Income Distribution Policy. MACROECONOMIC MODEL 413 is absorbed in agriculture (sectors 1 and 2); the remainder is distributed in trade (sector 12) and services (sector 14). After full employment is reached, economization of labor input is assumed to take place only in trade and services. This is based on the rationale that the base-period employment parameters in these sectors contain disguised idle man- power. Consumption and savings block Domestic savings in the model have been divided into household savings, corporate savings, government savings, and additional savings in either the household or government sectors. Household savings are the sum of savings generated in different occupations and income classes. The savings propensity in any sector is a function of real disposable income on a per capita basis. Thus total household savings are a func- tion of the sectoral composition of incomes, the pattern of incidence of all taxes, the age structure of the population, and the dependency ratios in different occupations and income classes. Government savings are the difference between government revenue and government con- sumption. Corporate savings are a function of corporate growth in gen- eral and export growth in particular. All equations pertaining to private savings, total savings, and foreign savings are identities. Public finance block Government revenue is the sum of receipts on the basis of fixed average rates applied to the relevant income categories and commodity output on trade. For direct taxes, separate rates are estimated for differ- ent occupations and income levels. For indirect taxes and custom duties, separate rates are used for different sectors. Nontax revenues and tax receipts for local taxes, inheritance tax, and monopoly gains are exoge- nous. The growth of government consumption is a policy variable; the composition of its consumption basket is exogenous. Prices block Prices in each sector are estimated on a cost-markup principle and respond to changes in the cost of the various inputs (labor, capital, and imports) and indirect taxes adjusted by changes in total factor produc- tivity. The GDP deflator is equivalent to the average of gross output prices weighted by the final demand elements of any period. Prices are 414 PART FOUR: APPENDIXES affected by changes in technology through changes in real wages, rates of return on capital, and the weights by which the primary inputs are combined. These technology changes in the model operate through changes in ICORS and employment elasticities. The effect of import sub- stitution operates through import-output ratios and changes in exchange rates. The urban and rural cost of living indexes are derived as weighted averages of sectoral prices using the relevant consumption baskets. Exports and imports block Exports from each sector are exogenous: that is, it is implicitly assumed that exports are demand-determined and based on demand in the external markets. Imports are estimated endogenously using three alter- native approaches. First, imports are estimated on the basis of a residual flow approach, such that any shortage of demand, domestic and foreign, is met by imports. Second, import demand is based on the base-year average import propensities of the intermediate and capital-goods sectors, both of which are influenced by technological developments, and of the consumption sector, which is influenced by changes in behavior. Third, the desired import demand in each year is estimated on the basis of the ratio of the previous year's imports to total demand. The rationale for having three different approaches is this: actual imports are determined by the residual flow method. This only gives a consistent equilibrium position. Each period's import coefficients refer to realized parameters for that year. Hence, a deviation from the previous year's parameters represents the import substitution achieved for the current year. The model has a floor or minimum noncompetitive level of imports based on technological grounds. On behavioral grounds, some price mecha- nism must operate to make the required amount of import substitution feasible. Changes in the effective exchange rate operate as the equilib- rium mechanism. Base-period parameters are also used to get an overall idea of total import substitution achieved over the entire projection period. Noncompetitive imports are calculated on the basis of a non- competitive import matrix derived from input-output tables for 1970.4 Capital accounts block The deficit in the current account of the balance of payments is esti- mated from the capital account. It is the sum of net official and unofficial 4. Bank of Korea (BoK), 1970 Input-Output Tables (Seoul, 1973). MACROECONOMIC MODEL 4I5 capital inflows and changes in reserves. It is affected largely by commit- ment pattems, disbursement patterns, amortization of outstanding debt, and other capital transactions. The resource balance is determined by adjusting the current account balance for net factor-service payments and net transfers, all of which are exogenous except net interest pay- ments, which are determined by the amounts and terms of past commit- ments and by earnings of foreign exchange holdings. It is matched by the excess of imports over exports in current dollars. The resource balance in local currency and base-period prices is esti- mated by deflating the current-price resource balance by ciF import price deflators and is further adjusted by terms-of-trade changes between the base year and current year. Assumptions The 17-sector classification of the model is more detailed than the fourth plan's classification, which contains eleven sectors. These sectors are aggregated from a 53-sector interindustry table, which presents do- mestic and import coefficient matrixes for 1970.5 In addition, the model uses a capital-flow coefficient matrix computed by Km; this matrix was derived from a 1968 wealth survey and subsequently revised on the basis of a 1974 survey. Before using these capital-output ratios in the model, however, adjustment was needed for incorporating gestation lags between the investment and the changes in output. Different gesta- tion lags are assumed for each sector and estimated from project infor- mation in Korea and from intemational comparisons. If ICOR (-I) represents an ICOR with one year's gestation and ICOR (-n) represents an ICOR with a gestation period of n years, then: ICOR (n) = IcoR (-1)/(l + r)"'1 where r is the rate of growth of investment in that sector over the period. Over the projection period, investment was assumed to grow at an average of 9.5 percent a year. The employment elasticities were estimated in collaboration with EPB. In broad aggregates they are almost the same as those of the fourth plan. They also are fairly close to past time series data. Exact comparison is difficult, however, because of the absence of suitably disaggregated data 5. Kim, "Interindustry Projection Model." 416 PART FOUR: APPENDrxES and the fact that time series observations of employment elasticities are strongly affected by short-term economic fluctuations. The aggregate employment elasticities for the economy range from 0.211 to 0.867 over the sample period (1964-75). Aggregating over the entire pe- riod, the elasticity comes to 0.386, which is very close to the model's assumption of 0.371 for 1976-81 and slightly higher than the model's assumption of 0.324 for 1981-90. In the plan the average employment elasticity is 0.369. The working-capital matrix was estimated from input-output tables for 1973.0 It is of the diagonal type. This assumption is fairly strong, but is necessary in the absence of more detailed information. Four different savings propensities were estimated for four separate classes and for each occupation. The ratio of the self-employed to wage eamers was estimated from the Adelman-Robinson model. The numbers of wage earners and self-employed in the base period were estimated from these ratios and from the distribution of sectoral labor force for 1973, as given by RD1.7 Total population and working population projections are based on World Bank estimates discussed in chapters 4 and 6. The log variance of income for each occupation class was derived from the Adelman- Robinson study on Korea.8 The composition of the government sector's consumption is derived from the 1973 input-output table. Direct tax rates for households are borrowed from the 1974 family budget survey and changed exogenously over the projected period from plan informa- tion. The same is true for customs duties. Other revenue items are assumed to be the same as in the fourth plan until 1981 and to be exogenous thereafter. The investment allocation for each sector, espe- cially manufacturing, is derived from the fourth plan and from detailed work done by World Bank staff. The model also estimates the number of people below the income level required to permit daily intake of 2,210 calories a person. By using 1974 retail prices, the annual expenditure of people at this line has been calculated to be W61,000 for the urban sector and W55,800 for the rural sector.9 In the base year 1974 the number of people below this line was 3.77 million; for 1976 the model estimated the number to be 2.7 million. 6. Box, "1973 Input-Output Tables" (Seoul, n.d.; processed). 7. Kim, "Interindustry Projection Model." 8. Adelman and Robinson, Income Distribution Policy. 9. This figure becomes approximately W67,000 a year in 1975 prices. MACROECONOMIC MODEL 417 The major findings of the simulations are described in chapter 3; employment projections are described in chapter 6. Sector Classification Equivalents in the 53-sector 17-sector classification classifiation 1 Grains I Part of agriculture and forestry 2 Other agriculture I Part of agriculture and forestry 2 Fishing 3 Mining including oil 3 Coal 4 Metallic ores 5 Nonmetallic minerals 4 Heavy, labor-intensive, export- 34 Electronics oriented manufacturing 5 Heavy, capital-intensive, export- 23 Rubber products oriented manufacturing 6 Heavy, labor-intensive, domestic- 32 Nonelectrical machinery oriented manufacturing 33 Industrial electrical machinery 35 Household electrical machinery 36 Shipbuilding and repairing 37 Railroad transport 38 Motor vehicles 39 Precision and optical products 7 Heavy, capital-intensive, domestic- 14 Pulp, paper, and paper products oriented manufacturing 16 Inorganic chemicals 17 Organic chemicals 18 Chemical fertilizers 19 Synthetics 20 Other chemicals 21 Petroleum products 22 Coal products 24 Cement 25 Glass, clay, and stone products 26 Iron and steel 27 Rolled steel 28 Steel pipes and plated steel 29 Cast and forged steel 30 Nonferrous metals 31 Metallic products 8 Light, labor-intensive, export- 9 Fabrics oriented manufacturing 10 Finished textiles 11 Leather and leather products 40 Other manufactures 9 Light, capital-intensive, domestic- 6 Processed foods oriented manufacturing 7 Beverage and tobacco 15 Printing and publishing 418 PART FOUR: APPENDIXES Equivalents in the 53-sector I7-sector classification c1assification 10 Light, labor-intensive, domestic- 13 Wood products and furniture oriented manufacturing 11 Light, capital-intensive, export- 8 Fiber spinning oriented manufacturing 12 Lumber and plywood 12 Trade (retail and wholesale), bank- 45 Banking and insurance ing, and insurance 49 Commerce 13 Dwellings 46 Housing 14 Education, health and other ser- 50 Education vices, including public administra- 51 Health tion 52 Other services 15 Transport and communications 47 Communications 48 Transport and storage 16 Electricity 43 Electricity 44 Water and sanitary services 17 Residence, buildings, and other 41 Residence and buildings construction 42 Public and other constructions The Equation System In the equations, variables not having the subscript t refer to the time t. An asterisk means multiplication; a bar over a symbol means exogenous. Variables ADR Average dependency ratio ADTX Additional taxation AGREMP Employment in agriculture AIDR Index of average dependency ratio ALFA4 Ratio of value added to gross output in the ith sector ASPS Additional savings effort in household sector AVWG Average wage for wage earners Bi Matrix of investment gestation lags by sector BBij Distribution matrix of fixed capital-output coefficients BWK,1 Distribution matrix of fixed working capital-output co- efficients CGi Government consumption by sector CGT Total government consumption COPE Elasticity of corporate saving CP, Private consumption by sector MACROECONOMIC MODEL 419 Variables CS Corporate savings CSS Ratio of corporate income to value added in the ith sector CUSTD Customs duties Di IDomestic demand by sector DELTA Parameter for replacement investments with respect to net output DT Direct taxes E-i Exports by sector ELNFS Exports including nonfactor services ELNWi Labor elasticity of nonwage earners for value added per sector ELW4 Labor elasticity of wage earners for value added per sector ENT Total wage earners ENWi Employment of nonwage earners by sector ENWT Total nonwage earners EWU Employment of wage eamers by sector EW1 Wage elasticity with respect to labor productivity EW2 Wage elasticity with respect to cost of living for urban wage earners EXPR Aggregate export price in dollars FP Rural population (agriculture and mining) FS Foreign savings GAMMA Incidence ratio of nontax and customs duty in ith income class GDPDEF GDP deflator GR Government revenue GS Govemment savings HS Household savings ICORP Incremental capital-output ratio by sector ID, Investment demand by destination by sector IINDT Index of indirect tax rate INDT Indirect taxes INDTS Indirect tax incidence in the ith income sector INDX4 Indirect tax rate by sector INTMNC6J Noncompetitive intermediate import coefficient matrix 420 PART FOUR: APPENDIXES Variables INTNCM, Intermediate imports of noncompetitives by sector IPLi Index of labor productivity by sector ISi Investment demand by source by sector IWKOR Incremental capital-output ratio of working capital by sector IWR4 Index of wage rate by sector LAMBDAA Investment allocation coefficients by sector MC, Competitive imports by sector MCCQ Competitive imports of intermediate inputs by sector MCCC,j Competitive import coefficient matrix MDi Imports demanded (ex ante) MLNFS Imports including nonfactor services MScs Competitive imports (ex post) NA Nonagricultural output NONTAX Nontax revenue nontax Rate of growth of nontax revenue OUTM Outmigration ratio PADTX Incidence of other taxes in the ith income class PDi Prices by sector PEE Export price index PMDi Price index of imports FOB in dollars PMM Import price index POP Population PSS Incidence of additional saving in the ith income class PXD, Price index of exports FOB in dollars PY. Labor productivity by sector RCL Rural living cost index RCNCij Rural consumption by income class by sector RCON; Rural consumption by income class RDINt Rural disposable income by income class RDTRi Direct tax rate for the ith rural income class RESI Residual investments RIi Replacement investments by source by sector RIDS Replacement investments by destination by sector RlNCi Rural income by income class of 1, 2, and 3 RPi Sectoral growth rates RRI, Rural real income by sector of 1, 2, and 3 MACROECONOMIC MODEL 421 Variables RSAVi Rural savings by income class RSB Resource balance in current won RSR, Savings ratio of the ith rural income class RU Rural unemployment RW4 Real wage rate by sector SNWINCQ Sum of income of urban nonwage earners by income class by sector SRINCq Sum of rural income in the ith sector by income class SUWINCi, Sum of income of urban wage earners by income class by sector TI Total investments TOTEMP Total employment TS Total savings UN Unemployment UNS Unemployed labor force allocation in a sector UNS4 Surplus labor allocated to a sector UNWCON, Consumption of urban nonwage eamers by income class UNWDINi Disposable income of urban wage earners by income class UNWDTR, Direct tax rate for the ith urban nonwage income class UNW4I Per capita income of urban nonwage eamers UNWINCs Income of urban nonwage eamers by income class UNWSAVi Savings of urban nonwage eamers by income class UNWSR, Savings ratio of the ith urban nonwage income class UU Urban unemployment UWCL Cost of living index of urban wage eamers UWCONi Consumption of urban wage earners by income class UWDIN, Disposable income of urban wage earners by income class UWDTRi Direct tax rate for the ith urban wage income class UWINCi Income of urban wage earners by income class UWSAV4 Savings of urban wage earners by income class UWSR, Savings ratio of the ith urban wage income class Wi Wage by sector WK& Working capital by origin by sector WKDi Working capital by destination by sector WP Labor force (working population adjusted by participa- tion rates) 422 PART FOUR: APPENDIXfiS Production block YNit = YNi, t-l RPit i= 1, 2, 3 YNit = YNj, t-I + Bi, t-l IDi, t-11lCORt-1 i 4, . .., 17 XNj = YNi/ALFAi i=1,...,17 GDPMP z YN4 i=1,...,17 Investment by sources block ISi = IBBij * IDj iz=1,...,17 1,... , 17 RIh = BBij* RIDj i=1,.,17 WK; = IBWKij * WKDj i=1,...,17 j =1,. . ., 17 Investment by destination block TI =TS ID, =IDi i= 1,2,3 WKDit =IWKOR (XNit-XNi, t-1) i= 1,.., 17 RID, =DELTA,i*YNi . i=,17 RESI =TS- RID- VD4 i= 1, 2,3 ICOR ICOR ID, LAMBDA, * RESI i =4,..., 17 Employment block (EWt - EWi, t-)/EWi, t, i = 1,..., 17 = ELWi * [(YNit- YNi, t-)/YNI, t-l] (ENW, - ENWi, t)/ENWi, t- i=1,...,17 = ELNWi * [(YNit -YNit-01M, t_J UN = WP -(EW± + ENW;) i=1,..,17 PYi =YNM/EW4 i=1,..,17 MACROECONOMIC MODEL 423 Employment block POPt = POPt-1 (1 + yrt) UNS4 =UN * UNAC i17 ADR =POPU/WP AIDR = ADRt/ADR, DRit = DR,,_1 * AIDR WP = WP ELW4 = ELWs ELNW, = ELNW, EWT =1EWj ENWT = ZENW6 TOTEMP = ENT + ENWT AVWG = (1Wi * EW4)/EWT OUTM, = 8.82* (NA,/NA,, -1) + 12.946 * (NAt_1/NAt.2 - 1) -0.349 FPt = (1 - OUTM) * (WPt/WPt-1) * FPt-1 UU = WP-FP-EW4- ENW, i =4,...,17 AGREMP = IEWT4 + YENWT4 i= 1,2 Income distribution block RRI, = [CSS2 * YNi - PSS2 * YN, - PADTX * YNi - INDTS, - GAMMA, * (NONTAX + CUSTD)]/[(EWi + ENW) * DR,] + UNS4 i = 1, 2, 3 Assuming a lognormal distribution of the foregoing expression and a given variance of total income earned in section i by class c, then: RlNCe =- SRINCi, i 1, 2,3 Wit-W, tW,/W1, t- = EW1 - [(PWt - PY4, t-) PY,,t-1] + EW2 * [(UWCLt-2 - UWCLt-2)/ UWCLI-21 - (0.06/UN.PERCENT) i =4,..., 17 424 PART FOUR: APPENDIXES Income distribution block RWV = WI/UWCL UWhi RW/DRi Similarly, assuming a lognormal distribution and a given variance, then: UWINCc = YSUWINC, i = 4,..., 17 UNWlI = K(YNM PDs - EW, W0 (ENW; + UNSj)] *(WMR (1/UNWCL) i 4., 17 Similarly, assuming a lognormal distribution, then: UNWINCc = ISNWINCj, i 4, ..., 17 PDINc = RINCc (1- RDTRc) c 4 UWDINC UWINCc (1- UWDTR,) c 1,..., 4 UNWDINC UNWINC, (1- UNWDTRo) c 1,..., 4 RSAV, =RDING * RSR, c 4 UWSAVC UWDINo * UWSRC c = 1,..., 4 UNWSAV, = UNWDINo * UNWSRc c 1, ..., 4 HS =(F(RSAV, + UWSAVc c 4 + UNWSAVe) CSt - CSt-/CSt1 = COPE * GDPMPt - GDPMP,-,/GDPMPt_ GS =GFR-GCT FS = (RSB/PMM) + ET * (PEE/PMM - 1) TS =HS+CS+GS-+ FS+ASPS CSSS = 1- CS/GDPMP COPE = 1.37 * (ET/ET- 1) PADTX = ADTX/GDPMP DS =HS + CS + GS + ASPS ASPS =ASPS PSS = ASPS/GDPMP. MACROECONOMIC MODEL 425 Consumption block RCON, = RDlN0 (1-RSR,) c = 4 UWCON, = UWDIN, (1- UWSR,) c = 1..., 4 UNWCONe = UNWDIN, (1- UNWSRO) c 1, .., 4 RCNCą, =RK-R BCON, = 4 i 1,. 17 UNCNCC = UHi, * UWCON, c 1,..., 4 i-1 ... 17 UNWCNCO = UH* UNWCON c =1,..., 4 i 1,..., 17 CPi =IRCNCi + UWCNC4 c 4 + UNWCNC,, 17 C =3CP-+CG, 17 Public jinance block DT =(BlIN0 * RDTRc + UNINC, * UNDTR, + UNWINCa * UNWDTRc) 4 INDT =XXN1 * INDX INDT 17 NONTAX _ NONTAX,- * (1 + y wontax) CUSTD = Mi TR i17 GR _ DT + INDT + NONTAX ± CUSTD + ADTX CGT CGi =KS, CGT i ADTX -ADTX ASPS = ASPS INDTSi =XN4 * INDX, * llND 17 426 PART FOUR: APPENDIXES Prices block PDi = AINVjj ([UNITWI * (TOTEMP,t/ I TOTEMP,0) * (YNj,/YNjt) IWR1] + UNITNW + UNITW - [UNITW1 * (TOTEMP,t/ TOTEMP,0) (YNI./YNit) * IRC1] + {(UNITIM, I 1PIM, * [(MT/GDPMP)/ (MTo/GDPMPo) f IEEXCHI} + UNITIT1 f IINDT) i = 1,... 17 IPLt = PY4t/PYio i-i .,17 IWRi = wi=1... 17 IINDT = IINDT ICUSTD = ICUSTD RCL = (RCNCd c = 1R.N.C. 4 4 c ic * (PD, * W1 + IPIM( * W2)] i= .,17 UWCL = If(YUWONCd/1UWONCci) c =1,4... 4 i c ic * (PD * W +IPIM, *W.)] i 1 .,17 UNWCL = [(IUNW0NCc 1, N, 4 4 c ic (PDi f WI + IPIMA W2)] .=1,, 17 PXD =PXD PMD =PMD EXPR =PXDi * EJET, when E= ET i=1,...,17 IMPR = PMD M4, t-/MT, when Mt = MT i=, ., 17 PEE = EXPR MPR GDPDEF = I(PDi + FD) F i = 1,., 17 MACROECONOMIC MODEL 427 Imports and exports block Mt =CP +CGIS+SWKi + EW i, .. , 17 + R1h-XNi + lAi1XN1 j=1,...,17 INTNCM4 -INTMNCil * XN1 i -1 ., 17 j j=1, ...,17 MCC, =YMCCC1j * XNj 17 I ~~~~~~~j=1,...,17 MDi =MCCi + INTNCMi i=1..,17 MCI =MD,-INTNCM + MC, i=1,...,17 MSC =MCj (CP + CG + IP) i =1,...,17 D, =XN,-E±+Mi i=1,.,17 MST = i=1,Y,17 MT =C+TI+ET-GDPFC ELNFS =Ei= 1, I 11 MLNFS = IMs I =I,.. ,1 ET =SE i= 1, ,17 Appendix B Jayati Datta-Mitra Structure of Exports and Imports ALTHOUGH ExPoRrs in the early 1960s were moderately buoyant, they constituted less than 5 percent of GNP. Between 1965 and 1975, how- ever, the pace of export expansion was phenomenal. Merchandise ex- ports, which totaled $175 million in 1965, grew at an average rate of about 40 percent a year to reach more than $5 billion in 1975. Measured in 1970 prices this growth rate was about 34 percent a year. In 1976 exports were about $8 billion. Manufactured exports were the impetus to export growth throughout the entire period.' As early as 1965 manu- factured exports alone totaled $144 million to constitute about 82 percent of merchandise exports and 8 percent of manufactured output. In 1976 their share in total merchandise exports exceeded 90 percent, and their share in manufactured output exceeded 25 percent. This record appears all the more spectacular when placed in the context of recent develop- ments in world trade. During the 1973-76 period the net growth in world trade in manufactures was relatively slow, yet Korea was able to almost double its export volume. Imports of goods and services consistently exceeded exports after 1960. Although import performance was erratic in the first half of the 1960s, 1966 was a watershed. During 1966-70 imports increased fivefold, and the import bill grew at about 34 percent a year in current prices. The primary impetus to this growth was export activity. In order to contain the import bill, government initiated measures to encourage the import substitution of intermediate and capital goods. During 1970-75 the growth rate of commodity imports was held to about 30 percent a year in current prices. Measured in 1975 prices, the annual growth rate of imports for that period was 11 percent. 1. As in the fourth plan documents, the term manufactured exports here encompasses all exports other than those of food and live animals, beverages and tobacco, mineral fuels, lubricants and related materials, and animal and vegetable oils and fats. 429 430 PART FOUR: APPENDIXES Export Structure The expansion in aggregate exports was accompanied only by modest changes in the commodity and market structure of exports. Analysis of export values, shares in total exports, and growth rates of a selected list of two-digit commodity groups responsible for more than 75 percent of Korea's total exports since 1965 reveals heavy reliance on a handful of export commodities (table B.1). Six major two-digit commodity groups contributed about 46 percent of export eamings in 1965-textiles, cloth- ing, footwear, wood products, electrical machinery, and miscellaneous manufactures, which include wigs and eyelashes.2 Their combined share in export eamings was 68 percent in 1970 and 60 percent in 1975. Grow- ing reliance on these commodities was not the passive outcome of neglect of new opportunities, but the result of the buoyancy of these traditional categories. A study of LDC exports to industrial countries during the 1967-73 period confirmed that exports of clothing, electronics, wood manufac- tures, and footwear registered higher growth rates than aggregate LDC exports.3 Korea appears, then, to have focused on the appropriate export categories, and this was by design. There nevertheless was substantial diversification in Korea's commodity structure. A host of new commodi- ties appeared at the two-digit level-transport equipment, travel goods and handbags, and metal manufactures are examples-but even more diversification occurred at the detailed three-, four-, and five-digit levels. This dynamism, evident in both traditional and new export groups, suggests that Korea did not become locked into a restricted product strategy. The United States and Japan absorbed more than 70 percent of Korea's aggregate exports until the early 1970s. By 1976, however, this share was reduced to about 56 percent.4 Such concentration is not sur- 2. These are the respective headings of srrc groups 65, 84, 85, 63, 72, and 89. Korean trade data are classified according to the statistical classification of com- modities for Korean trade which in turn is based on the United Nations' revised standard international trade classification (srrc). Exports are divided into nine broad one-digit commodity groups; manufactured exports comprise groups 5 through 8, except for the two-digit group 68, which covers nonferrous metals. 3. At least when measured in current prices. Donald B. Keesing and Phi Anh Plesch, "Industrial Countries' Manufactured Imports from Developing Countries," (Washington, D.C.: World Bank, 1976; processed). 4. Bank of Korea (Box), Economic Statistics Yearbook, various issues. STRUCTUREf OF EXPORTS AND IMPORTS 431 prising. The United States was the leading importer of LDC manufac- tures in the 1960s, and its share in LDC exports continued to expand in the 1970s. In the 1967-73 period, for example, the United States con- sistently absorbed more than 40 percent of LDc manufactured exports. Korea's export strategy for the United States apparently sought to exploit this pattern. The concentration on the Japanese market was similarly motivated. Japan's share in LDC manufactured exports grew substantially after the mid-1960s-from about 5 percent in 1967 to about 13 percent in 1973. An equally important factor was Korea's traditional links with Japan. Korea's strategy of widening the geographical spread, range, and skill intensity of its exports, while apparently maintaining the broad structure and pattern of labor intensity, focused on two basic objectives: diversifying product categories and markets; nimbly maneuvering in and out of particular products and geographical areas to suit the shifting tastes of markets and overall trends in the international economy. Korea's success in achieving these objectives led to a seemingly contra- dictory result. The relative constancy, even growth, of traditional two- digit export groups was coupled with a proliferation of new markets and export commodities, which demonstrated variability in their shares over time. Diversification The diversification of exports occurred not only at the detailed level- that is, beyond the two-digit groups-but at the broad two-digit level as well. Although this performance apparently did not upset the pattern of dominance of traditional exports at the two-digit level, the density of the export structure increased. In other words, the gaps in the Korean export structure were increasingly closed both at the level of broad export groups and the detailed commodity level within each group. This feature characterized Korea's performance in a variety of geographical markets. The increases in density were remarkable even at the relatively aggre- gative two-digit level. In 1967 only six two-digit groups had exports in excess of $10 million. By 1975 the number had risen to twenty-three; in fifteen of these groups, export values exceeded $50 million. Analysis at a more detailed commodity level also reveals clear signs of diversifica- tion. For example, the number of four-digit groups in the SITC 899 group with exports approximating $1 million rose from one to six between 1967 and 1975; the number of three-digit groups in the srrc 71 group, 432 PART FOUR: APPENDIES Table B.1. Export Performance of Selected Commodities, 1965-75 Annual Export vale growth (millions of dollars) rate srrc category 1965 1970 1975 (percent) 03 Fish and fish preparations 17.8 40.8 359.5 18.0 06 Sugar and sugar preparations 0.8 1.5 119.1 13.8 12 Tobacco and tobacco manufactures 0.9 13.5 66.5 72.9 26 Textile fibers 7.7 42.6 43.0 40.7 33 Petroleum and petroleum products 4.9 95.4 - 62 Rubber manufactures 1.0 3.7 90.9 28.9 63 Wood and cork manufactures (excluding furniture) 18.2 93.5 227.6 38.7 65 Textile yarn, fabrics, made-up articles, and related products 26.3 84.9 648.9 26.4 66 Nonmetallic mineral products 2.8 6.5 106.8 18.7 67 Iron and steel 12.7 13.4 231.5 1.0 69 Manufactures of metal 2.2 12.2 124.1 41.2 71 Machinery other than electric 2.5 8.4 76.8 27.4 72 Electrical machinery 1.9 43.9 441.6 87.1 73 Transport equipment 1.1 9.2 183.7 53.0 83 Travel goods, handbags, and similar articles 0.1