10961 WORLD BANK SECTORA~L PROGRAMS AND POLICIES INDUSTRY TELE:.COMMUNICATION WATER SUPPLY TOURISM POWER T1D A N IOCDG'\DTATIlQ-N I WORLD BANK OPERATIONS Sectoral Programs & Policies WORLD BANK OPERATIONS Sectoral Programs & Policies Published for The International Bank for Reconstruction and Development by The Johns Hopkins University Press, Baltimore and London 1972 Copyright © 1972 by the International Bank for Reconstruction and Development All rights reserved Manufactured in the United States of America Library of Congress Catalog Card Number 72-4032 ISBN 0-8018-1448-0 (clothbound edition) ISBN 0-8018-1449-9 (paperbound edition) PREFACE Basically the World Bank's operations are shaped by the needs of the development programs of individual developing countries, and the part the Bank can play, with other development agencies, in satis- fying those needs. For this reason our activities are usually examined on a country by country basis. However, it is useful also to look at the aggregation of Bank opera- tions by sector of activity (agriculture, industry, transportation, etc.) to determine the main problems and policy issues faced by the Bank in the various fields. The papers contained in this volume were orig- inally prepared and printed over a period of twelve months as an internal exercise to meet those objectives. The papers describe the distinctive economic and developmental characteristics of each sec- tor and review the approach and scale of Bank operations in each field. The first of the papers deals with Agriculture which is becoming an increasingly important sphere of Bank activity, and in which the great majority of the population of less developed countries is engaged. Industry-the subject of the second paper-is the main alternative occupation; though it involves fewer people, it is a mainspring of rapid growth. One way in which many of the less developed countries can earn foreign exchange from their richer neighbours is through providing tourist facilities. The paper on Tourism describes the problems the Bank has faced, and the solutions emerging in the four years it has been engaged in this new sector. The "infrastructure" on which growth in agriculture, industry and tourism can be built is dealt with in four papers: Transportation, Tele- communication, Electric Power, Water Supply. Each gives an account of the emergence of Bank policy from long experience. But development is not dependent on material growth alone. Edu- cation, or the lack of it, is recognized as a major constraint, and the paper on this subject indicates why the Bank has been placing in- creasing emphasis on this sector. Similarly it is increasingly being accepted that there cannot be social and economic progress without some check on the exponential growth of population; the paper on Population Planning lays out the global problem, as well as describing the first steps which the Bank has taken in cooperation with many other organizations, to assist Governments requesting help in this complex field. In many parts of the developing world a critical human problem is the explosion of the cities, which attract an endless stream of immi- grants from the countryside, for whom there are often neither homes nor jobs. These and other problems of Urbanization are examined, and the Bank's somewhat tentative policies in this new field of activity are described in the final paper in this volume. But the end of the volume is not the end of the Bank's operational story. These papers themselves describe policies which are in evolu- tion, being tested and changed as we learn more about the science of development. It is our hope that, just as the preparation of these papers has con- centrated our thoughts on the lessons of our experience, and our plans for the future, their publication will be useful for all those who are engaged in development, and who are trying to better the lives of the majority of the inhabitants of this globe. President, World Bank Group CONTENTS AGRICULTURE .............................................. 1 Challenges Facing Agriculture ....... ...................... 3 Some Major Policy Issues ....... .......................... 18 Past Bank Group Activities in Agriculture ..... ............... 36 The Bank Group's Future Role ....... ....................... 57 Annexes 1. Indices of Agricultural Production by Level of Development and by Region ...... .................. 65 2. Normal Variations in Economic Structure with Level of Development ........... ........................ 66 3. Agricultural Output per Agricultural Worker and Factors Associated with Differences in Output, 19 Countries, 1960 67 4. Compound Annual Rates of Growth in Total Agricultural Production, Total Agricultural Population, Agricultural Production per Farm Person, and Crop Areas per Farm Person, 25 Countries, Selected Periods 1950-1968 ...... 68 5. Actual and Projected Capacity, Production and Consumption of Fertilizers in Developed and Developing Countries ........ ...................... 69 6. Area Devoted to High-Yielding Varieties in Relation to Area of All Varieties, 1970-1971 ...... ............... 69 7. Medium-term (1974) Forecast for Cereal Production, Utilization and Net Trade in Selected Developing Countries .............. .......................... 70 8. Share of Agricultural Primary Products in Export Earnings of Selected Developing Countries, Average 1967-1969 . 71 9. Indices of Value of Agricultural Exports and Imports, by Level of Development ....... .................... 72 10. Composition of Agricultural Export Earnings of Developing Countries, 1955-1970 ...... .............. 73 11. Bank and IDA Lending by Area and Subsector for Various Periods .......... ......................... 74 12. Bank Group Investments in Agriculture-Related Industries 76 13. Average Size of Agricultural Project by Area and Subsector over Various Periods ...... ................ 77 14. Loans and Credits for Agriculture FY1969-1971: Total Cost and Number of Participating Farms .... ...... 78 15. Distribution of Project Investment and Number of Participating Farms by Size of Project Investment Per Farm 80 16. Summary of World Bank's Activities in Agriculture, with Projections through FY1976 ...... ............... 81 INDUSTRY . ................................................. 83 Industrialization and Development ...... ................... 85 Stages of Industrialization ................... ........... 87 Obstacles to Industrialization ................ ........... 91 Social Aspects of Industrialization ............ ........... 96 Strategies and Policies ........ .......................... 99 Manufacturing and the World Bank Group ..... .............. 102 Main Features of Past Operations ...... ................... 102 Projected Activities, FY1972-1976 ...... ................... 107 The Role of the Bank Group ....... ...................... 118 Annexes 1. Some Indicators of Industrial Development ..... ....... 126 2. Characteristics of Bank Group Manufacturing Projects by Organizational Unit, FY1967-1971 .... ...... 129 3. Bank Group Manufacturing Operations by Sector, FY1967-1971 ............. ........................ 130 4. Pattern of Bank Group Manufacturing Operations in Selected Countries by Organizational Unit, FY1967-1971 ............... ....................... 131 5. Bank Group Manufacturing Operations by Organizational Unit and Major Recipient, FY1967-1971 ............. ........................ 132 6. Bank Group Manufacturing Commitments by Regions and Countries, FY1967-1971 ....... .................. 134 7. Bank Group Commitments to Industrial DFCs by Regions and Countries, FY1967-1971 ..... ............. 136 8. IFC Manufacturing Commitments by Regions and Countries, FY1967-1971 .................. ......... 137 9. Direct Bank-IDA Manufacturing Commitments by Regions and Countries, FY1967-1971 ..... .......... 139 viii TRANSPORTATION ............ .............................. 141 Characteristics of the Sector ....... ........................ 143 Past Activities in the Sector ........ ........................ 151 Projection of Future Operations ...... ..................... 161 Annexes 1. Basic Transport Data ......... ...................... 180 2. Bank and IDA Lending for Transport through FY1971 184 3. Selected Statistics on Bank Group Projects FY1967-1971 Table 1-Highway .190 Table 2-Railway .191 Table 3-Port .192 4. Summary of World Bank Transport Activities With Projection through FY1976 .193 5. Graph: Relation Between Motor Vehicle Density and GNP Per Capita, 1963-1967 .194 TELECOMMUNICATION .......... ............................ 195 Introduction . ........................................... 197 International Perspective ........ .......................... 198 Special Characteristics ......... ........................... 200 Main Features of Bank Group Operations .206 Annexes 1. Telecommunication Loans and Credits FY1962-1971 211 2. International Sector Data-Telecommunication .212 3. Summary of World Bank Telecommunication Activities With Projection Through FY1976 .214 ELECTRIC POWER .215 Introduction .217 Characteristics of the Sector .218 Recent Bank Lending in Power .222 Planned Bank Activities .224 Areas of Challenge ............. 226 Annexes 1. Chart: Installed Capacity in Developing Member Countries 233 2. Table: World Bank Group Lending in the Power Sector, FY1948-1971 ............... ....................... 234 3. Table: International Sector Data ...... ............... 236 4. Table: Summary of World Bank Electric Power Activities With Projection Through FY1976 ..... ....... 238 WATER SUPPLY AND SEWERAGE ....... ....................... 239 Water and Sewerage in Urban Areas ...... .................. 241 Sector Features Significant for the Bank Group ..... ........... 243 Main Features of Bank Group Lending ...... ................. 247 Annex Tables 1. World Bank/IDA Lending for Water/Sewerage, FY1962-1971 ............ .......................... 252 2. Comparative Data Sheet ....... ..................... 253 3. Summary of WolId Bank Water Supply and Sewerage Activities With Projection Through FY1976 ..... ....... 254 EDUCATION ................................................ 255 Trends in Educational Development ...... ................... 257 Review of World Bank Policies and Operations, 1963-1971 ...... 265 World Bank Education Policies and Operations in the 1970s ..... 271 Conclusions ............................................. 279 Annex Tables 1. Enrollment Growth Rates, 1960-1966, and Comparative Enrollment Ratios, 1968, by Regions and Levels .... .... 281 2. Inefficiency in Primary Education ..... ............... 282 3. Public Expenditure on Education as a Percentage of the Budget and National Income, 1960 and 1965 .... ....... 282 4. World Bank/IDA: Education Loans by GDP/Capita at Time of Loan, FY1963-1971 ...... ................. 283 5. Analysis of World Bank/IDA Education Lending, FY1963-1971 ...................................... 283 6. World Bank/IDA Education Projects: Student Places Provided or Improved and Boarding Places and Staff Housing, FY1963-1971 ....... .................. 284 7. World Bank/IDA Education Projects: Cost of Physical Facilities per Student ........ ....................... 285 8. World Bank/IDA Education Projects Approved as of July 1, 1971 . ...................................... 286 9. Comparative Education Indicators ...... .............. 288 10. Summary of World Bank Education Activities With Projection Through FY1976 ...... ............... 290 POPULATION PLANNING ......... ............................ 291 Introduction .............. .............................. 293 World Population Trends ........ ......................... 297 World Population Projections ....... ...................... 300 Economic Effects ........... .............................. 303 Family Planning Efforts ......... ........................... 307 The Bank's Program and Approach ...... .................... 317 Annexes 1. Population Projections, 1970-2100 ...... ............. 329 2. Specific or General Population Targets in 27 Countries . . 341 3. Contraceptive Methods ........ .................... 343 4. Evaluation of Family Planning Programs ..... .......... 351 5. Bilateral, Multilateral and Private Agencies .... ........ 361 6. Research . ......................................... 369 TOURISM . ................................................. 373 Review of the Sector ......... ............................. 375 World Bank Group Operations in Tourism ..... .............. 387 Other External Financial and Technical Assistance .... ......... 393 Future World Bank Group Operations ...... ................. 395 Annex Tables 1. Comparative Data on International Tourism in Selected Developing Countries, 1970 ..... ............ 400 2. Tourism Financing by the International Finance Corporation ........... ........................... 402 3. Commitments for Tourism Projects under World Bank Loans to Development Finance Companies ..... ....... 403 4. Direct World Bank Loans and International Development Association Credits for Tourism Projects, FY1971-1972 . .. 403 5. Projected Bank Group New Commitments for Tourism, FY1972-1976 . .................................... 403 URBANIZATION ............... ............................. 405 Introduction and Summary ........ ........................ 407 Dimensions of the Urbanization Problem ..... ............... 412 The National/Regional Setting ....... ...................... 425 The Problems of Urban Centers ....... ..................... 433 The Bank's Role ............. ............................ 457 Annexes 1. Background Data on Urbanization ....... ............ 475 2. Urban Land Policy .................................. 488 3. International Assistance in Urban Development ..... ... 503 4. Bank Lending for Urban Purposes ....... ............. 513 THE WORLD BANK GROUP The World Bank Group comprises the Bank itself, officially named the International Bank for Reconstruction and Development, and its two affiliates, the International Development Association and the International Finance Corporation. Each of the three institutions was established to fulfill a distinct function, but all are devoted to the same general objective-the provision of financial and other assist- ance for the economic development of member countries. The World Bank was founded at the Bretton Woods Monetary and Financial Conference in 1944 and began operations in 1946. It is a Specialized Agency of the United Nations, as are its two affiliates. The Bank's principal role is the provision of loans to developing countries for a wide variety of productive projects in the various fields de- scribed in this volume. At present, the Bank is the largest multilateral source of development financing in the world. By the end of June 1972, the Bank Group had committed nearly $23 billion in over 100 countries. The International Development Association (IDA) was established in September 1960. IDA performs the same function as the Bank and has the same staff, but its loans (known as credits) are on much easier terms and are made to the poorest member nations. The International Finance Corporation(IFC),establishedinJuly1956, finances most types of commercial enterprises through investments in equity, loans without governmental guarantee and underwriting com- mitments. Its mission is to supply venture capital to productive private enterprises, to stimulate the development of local capital markets, and to promote the international flow of private capital. AGRICULTURE Challenges Facing Agriculture ........ ......................... 3 Production .............. ............................... 4 New Technological Advances ....... ..................... 6 Exports . ................................................. 10 Employment ............. ............................... 13 Effects of Technological Change on Employment .... ........ 14 Income Distribution .......... ............................ 17 Some Major Policy Issues .......... ........................... 18 Making Better Use of the Price System ...... ................. 20 Non-Price Rationing ......... ........................... 21 Price Distortions and Unemployment ...... ............... 22 Policy Implications .......... ........................... 23 Price Distortions and Public Investment ..... ............... 24 Small Farm Sector ........... ............................. 25 Economies and Diseconomies of Scale ..... ............... 27 Adoption of Innovations ........ ........................ 29 Land Policy .............. ............................... 30 The Impact of New Technology ...... .................... 33 Implementation of Land Reforms ...... ................... 34 Past Bank Group Activities in Agriculture ...... .................. 36 Capital Transfer ............ .............................. 37 Direct Lending for Agriculture ...... ..................... 37 Lending in Indirect Support of Agriculture ..... ............ 37 Trends in Bank Lending for Agriculture ...... ................ 38 Diversification of Lending ....... ........................ 38 Irrigation Development .39 Agricultural Credit .41 Livestock .43 Integrated Smaliholder Development .44 Land Settlement .46 Estates, Seed Production and Forestry .46 Agricultural Industries .47 Fisheries .47 Training and Research .48 Geographical Distribution of Lending .49 Technical Assistance and Project Preparation .50 Economic Research Activities .53 Impact of the Bank's Activities .54 The Bank Group's Future Role ......... ......................... 57 Overall Agricultural Investment and Capital Transfers .... ...... 57 The Direction of Future Lending ....... ..................... 60 Project Content ........... ............................ 61 Geographical Distribution ........ ....................... 61 Agricultural Sector Analysis ........ ....................... 63 Annexes 1. Indices of Agricultural Production by Level of Development and by Region ............ ............................ 65 2. Normal Variations in Economic Structure with Level of Development ........... ........................... 66 3. Agricultural Output per Agricultural Worker and Factors Associated with Differences in Output, 19 Countries, 1960 ... 67 4. Compound Annual Rates of Growth in Total Agricultural Production, Total Agricultural Population, Agricultural Production per Farm Person, and Crop Areas per Farm Person, 25 Countries, Selected Periods 1950-1968 ..... ...... 68 5. Actual and Projected Capacity, Production and Consumption of Fertilizers in Developed and Developing Countries ....... 69 6. Area Devoted to High-Yielding Varieties in Relation to Area of All Varieties, 1970-1971 ....... ................... 69 7. Medium-term (1974) Forecast for Cereal Production, Utilization and Net Trade in Selected Developing Countries . 70 8. Share of Agricultural Primary Products in Export Earnings of Selected Developing Countries, Average 1967-1969 ...... 71 9. Indices of Value of Agricultural Exports and Imports, by Level of Development ........ ....................... 72 10. Composition of Agricultural Export Earnings of Developing Countries, 1955-1970 ...... .................. 73 11. Bank and IDA Lending by Area and Subsector for Various Periods ............. .......................... 74 12. Bank Group Investments in Agriculture-Related Industries ... 76 13. Average Size of Agricultural Project by Area and Subsector over Various Periods ....... ................... 77 14. Loans and Credits for Agriculture FY1969-1971: Total Cost and Number of Participating Farms ..... ........ 78 15. Distribution of Project Investment and Number of Participating Farms by Size of Project Investment Per Farm ... 80 16. Summary of World Bank's Activities in Agriculture, with Projections through FY1976 ....... ................. 81 AGRICULTURE SECTOR WORKING PAPER * This paper discusses the part that agriculture plays in achieving the key development goals of greater production and exports, more em- ployment, and a better distribution of income. Some policy issues of wide relevance for less developed countries are singled out for atten- tion. After a review of the Bank Group's past activities in agriculture, projections are made of its future work in the sector. The paper is not intended to be comprehensive in its coverage. In order to be concise, it touches only briefly, and sometimes not at all, on several important aspects, such as agricultural credit, rural develop- ment, rural-urban migration, nutrition, international commodity stabi- lization schemes and ecological problems. Some topics that are of relevance to agriculture are discussed in companion working papers -particularly those dealing with population planning, transportation and water supply. The balance between rural and urban development is considered in the paper on urbanization. The term World Bank Group as used in this paper refers to the International Bank for Reconstruction and Development (World Bank), the International Development Association (IDA) and the International Finance Corporation (IFC). References to the Bank include IDA, but not IFC. Money amounts are expressed in U.S. dollar equivalents. The Bank's fiscal year (FY) ends June 30. CHALLENGES FACING AGRICULTURE The importance of agriculture in the economic development of low-income countries has come to be recognized increasingly in re- cent years. A number of causes have contributed to this change. Rapid population growth has focused attention on the need to ex- pand food production. At the same time, the success that has been achieved in obtaining dramatic increases in crop yields through the use of new agricultural technology has demonstrated that agriculture offers excellent investment opportunities. This combination of cir- cumstances has meant that, for many countries, measures to raise agricultural production have become both more urgent and more feasible. The renewed interest in agricultural development has been accen- tuated by a degree of disillusionment with the outcome of industrial growth. In particular, it is recognized that industrial growth has not succeeded in preventing rising unemployment and continuing pov- erty for the great mass of people. The creation of employment oppor- tunities and the alleviation of poverty have thus tended to be raised to the status of important development goals in their own right. The challenge for agriculture is to combine an acceptable rate of growth of production with movement toward these goals. In dealing with the related problems of poverty and unemploy- ment, agriculture is of exceptional importance. It is the sector which provides employment for most-typically, 50% to 80%-of the work force, contains the majority of poor people, and is the birthplace of many of the urban poor. Food and fiber, the products of the agri- culture sector, are prominent among the goods which poor people demand in greater quantities as their incomes rise. Furthermore, in countries not endowed with exploitable minerals, the foreign ex- change earnings from agricultural exports are frequently of critical importance for development. There is growing confidence that the development and application of new technology can solve many of the production problems in agriculture. But increasing concern is being expressed about the dis- tribution of the benefits of economic growth. This reflects the fear that the technological revolution will itself have adverse distributional effects within the agriculture sector-in particular, that its benefits will be captured to a disproportionate extent by the landowners, and that labor will be "saved" to no purpose other than to swell the numbers of unemployed. If these consequences are to be avoided, much rethinking of the conventional wisdom is required. Production The performance of the agricultural sector is judged primarily by its ability to expand production of food and raw materials. Increased agricultural production is of critical importance when both popula- tion and income grow rapidly in low-income countries, particularly if the poorest segments of the population share in the income growth. For example, if the population grows by 3%, and per capita income also rises by 3%, the total demand for food could increase, typically, by about 5%. Rising incomes also generate a demand for a diet that is of higher quality and more varied, so that the pattern of agricul- tural production needs to be diversified. During the 1960s, agricultural production in developing countries increased at an average rate of 2.6% per year, while food production alone grew by 2.8% per year (Annex 1). These rates were little more than sufficient to keep pace with population growth. As a result, per capita production increased at annual rates of 0.16% for all products and 0.20% for food alone. In Latin America and Africa, agricultural production per head actually declined. Although, if judged against the 4 forecasts ofwidespread famine which were common in the mid-1960s, this performance may seem reasonably creditable, it is clear that agri- cultural production must accelerate if future demands are to be met. According to the projections underlying the Provisional Indicative World Plan for Agricultural Development (IWP) prepared by the U.N. Food and Agriculture Organization (FAO), the total population of developing countries will rise from 1,720 million in 1970 to 2,515 million by 19851. To meet the increase in domestic demand for food as populations grow and incomes rise, the IWP estimates that food production will have to expand by 4.3% per year for the period 1967-1985. The marketable portion will have to grow even more rapidly-by 5% to 6% annually-because of increasing urbaniza- tion. Even if the rise in incomes is smaller, the demand for food will increase rapidly, since the greater part (about 70%) of the increase will stem from population growth. The growth in agricultural production has come from three sources: expansion of acreage, increase in yields, and shifts to more valuable crops. Expansion of acreage has been more important in Latin America and Africa, while yield increases have been more im- portant in Asia. Although there are considerable areas of unused cultivable land in Latin America and Africa, and some in Asia, the expansion of arable and harvested area in developing countries is unlikely to exceed 0.7% and 1.3% per annum, respectively, over the period 1962 to 19852. Since these rates are less than the expected rate of population growth, an increasing proportion of future agri- cultural growth will have to come from increased yields. The shift to higher value crops involves a greater degree of farm and regional specialization in production; this calls for improved communication and marketing facilities within countries and for trade policies which allow international specialization to take place. It is necessary not only to increase agricultural production, but also to improve the productivity of labor in agriculture. The improve- ment in productivity is required to raise the incomes of those engaged in agriculture, and to meet the demand for agricultural products coming from the increasing proportion of the population engaged in non-farm activities. As Annex 2 shows, economic growth ordinarily involves a decline in the relative importance of agricultural produc- tion and of the agricultural work force, since other sectors expand more rapidly than agriculture. This means that the proportion of 'More recent projections by the United Nations raise these figures to 1,765 million and 2,675 million, respectively. 'FAO, Provisional Indicative World Plan for Agricultural Development (Rome, 1970), Vol. I, p. 39. 5 agricultural output that is marketed, rather than consumed on the farm, must rise. If the marketed portion does not increase commensurately with the increase in demand, food prices rise. Since food expenditures represent half or more of the total expenditures of low-income peo- ple, the effects can be very disruptive, both economically and politi- cally. Some of the possible consequences are inflationary wage demands which slow down industrial growth, expenditure of scarce foreign exchange on food imports, or resort to draconian-but fre- quently ineffective-measures to commandeeror ration food supplies. Agricultural production per man varies considerably among coun- tries but is low in most of the developing world (see Annex 3). There is a general tendency for low labor productivity to be associated with a high man/arable land ratio. But the main point conveyed by Annex 3 is that there is no simple explanation of low productivity in agriculture; it results from a complexof factors, such as poor natural resources, lack of accumulated capital and current inputs, technological backward- ness, poorly-developed human skills, and the social environment. Undoubtedly, many opportunities exist for increasing agricultural output through technological innovation involving relatively small amounts of investment per man and per unit of land. But identifying and exploiting these opportunities is not always easy. Nevertheless, a recent study (the results of which are given in Annex 4) showed that agricultural output per farm person increased in all but one of a group of developing countries, even though in some the crop area per person declined. The gains in productivity reflect partly an in- creased use of purchased inputs-and, indeed, it is this source which must be increasingly relied on in the future. If it is assumed that the use of fertilizers is a rough measure of the use of purchased inputs of all kinds, the table in Annex 5 gives some idea of the probable scope for increasing the use of these inputs. Although fertilizer consumption in developing countries is very low in comparison with developed countries, it is expected to double between 1967/68 and 1972. But even the doubling will mean that the application of fertilizer per unit of cultivated area will remain relatively low. The actual availability of modern inputs such as pesti- cides, fertilizer, engines and piping will, of course, be an important determinant of gains in productivity. New Technological Advances Considerable improvements in agricultural technology have been achieved in the developing countries over the past 20 to 30 years. Initially the emphasis was on cash crops, particularly export crops, 6 where some quite remarkable technological successes have been achieved. Of particular importance have been the breeding of high- yielding oil palm varieties; the use of vegetative propagation tech- niques in rubber and tea production to allow the properties of high-yielding clones to be adopted widely; and the development of cotton varieties and their adaptation to the requirements of indus- trial processes. Cultivation practices have been greatly improved and there has been wide acceptance of chemical fertilizers, pesticides and herbicides, often specifically developed for particular crops in particular environmental conditions. In livestock production, too, there have been notable technological advances, even under tropical and semi-tropical conditions: production has been increased by im- proving animal health control, using tropical legumes for pasture improvement, and upgrading local cattle with improved disease- and insect-tolerant breeds. More recently, there have been similar technological advances in the production of food crops under the tropical and semi-tropical conditions typical of many developing countries. Such advances are still largely restricted to wheat, rice and coarse grains grown under irrigation. The wheat and rice varieties of the "green revolution" are capable of high yields with the aid of fertilizers, pesticides and ade- quately controlled water supplies. They were produced by means of genetic and agronomic "engineering" carried out in international and national research centers, and were developed specifically for regions with an adequate supply of rainfall or irrigation water. Their wider use will require an expansion of reliably irrigated areas in various parts of the developing world. But more work needs to be done to produce still better and more widely adaptable varieties, particularly for dry-land farming. The "green revolution," impressive as it has been, has left many parts of the world and many millions of farmers unaffected. A much bigger research effort is needed for crops such as millets, legumes, oilseeds, vegetables, tubers and fodder, if the advances made with wheat, rice and maize are to be matched. Similarly, more research is needed on livestock production and disease control-especially in the humid tropics-and on water management in irrigation systems'. The use of new, high-yielding cereal varieties has spread with re- markable speed in Mexico and parts of south and south-east Asia- and, to a lesser extent, in the Middle East and North Africa. Their use in Asia, the Middle East and North Africa commenced as recently as 1965-66, yet, by 1970, they covered 10 million hectares, or22% of the 'The Bank's role in supporting such research is discussed on page 48. 7 wheat area of 14 major wheat-producing countries. In the same year, another 10 million hectares, representing 13% of the total rice area in 12 countries, were planted with high-yielding rice varieties1. Annex 6 gives the data for individual countries. It shows that adoption of the new varieties of wheat has proceeded furthest in Pakistan (46% of the total wheat acreage in 1970-71) and India (33%). Partly because less irrigated land is available, only 6% of the Turkish and 3% of the Iranian wheat areas were planted with the new varieties. Correspond- ing rates of adoption of the new rice varieties were 50% in the Philippines, 42% in West Pakistan, 19% in South Vietnam, 15% in India, and 11% in Indonesia. While the expansion in area of high-yielding varieties can be meas- ured, their precise effect on yields and production is more difficult to determine. In 1970-71, in the less-developed countries of Asia and the Middle East (excluding Mainland China), the estimated produc- tion of wheat was 71 million tons and that of rice (paddy) 163 million tons. These represented increases of 30% and 19%, respectively, over the tonnages produced in 1965. Not all of these increases are attrib- utable to the "green revolution," of course, even though 1965 may have been a somewhat better-than-average year. If one takes the total area planted to new varieties in the countries listed in Annex 62 in 1970-71, and uses estimated yield multipliers for the new varieties, it is possible to calculate roughly how much they may have contributed to production. For example, if it is assumed that use of the new varieties (in combination with complementary inputs), instead of the old, increases yields by 100% in the case of wheat, and 25% in the case of rice3, wheat output has increased by 8.3 million tons (or 22%), and rice output by4.6 million tons (3.25%) in 1970-71'. 'There seem to be two main reasons why the new rice varieties have not been accepted as readily as the new wheat varieties. One is that they are less palatable than the old varieties of rice. The second is that they are less readily adaptable to a wide range of environments than the new wheat varieties. 'Note that not all countries where high-yielding varieties have had an impact are included in this list. Notable omissions are Mexico for wheat, and Taiwan for rice. 'These percentages do not imply that varietal improvement has been markedly greater in wheat than in rice. The difference in favor of wheat reflects the fact that the new wheat varieties are grown almost exclusively under irrigation; under these conditions, yields are well above average in any case since, unlike rice, a substantial part of wheat output is produced under dry-land conditions. 'Availability of the new varieties has led to an expansion of plantings, particularly in the case of wheat, as well as higher yields per unit area. The above calculations attempt to catch only the yield effect. However, they probably do not underesti- mate the net effect of the new varieties on agricultural output. Although they do not credit the new varieties with increased output resulting from expansion of areas sown to wheat and rice, neither do they debit them with the reduced pro- duction of crops-such as cotton and pulses-that were displaced by wheat and rice. 8 These estimates are very rough, because the yield multipliers used are of uncertain validity. Furthermore, they do not take into account the contribution of modern inputs used in conjunction with tradi- tional crop varieties. Nevertheless it seems clear that the impact on total wheat production in these countries has been substantial, but that the effect on rice production has been relatively small in the aggregate, though quite significant in particular regions. One result of the increased cereal production-as the FAO me- dium-term forecast on cereals (Annex 7) shows-is a trend toward greater self-sufficiency in countries which have traditionally relied on imports for part of their foodgrain requirements. However, there is reason to believe that grain production will not continue to increase as rapidly as hitherto unless stronger efforts are made to spread the use of improved technology, to intensify research, and to provide complementary inputs, particularly water. Not only will increased plantings of the new varieties be restricted by the lack of suitable environments; but the spread of disease and pests may become a problem too. For instance, in the Philippines both the traditional and the new varieties of rice were attacked by a serious virus disease in 1971, necessitating a resumption of rice imports. Even in wheat and rice, therefore, there is a need for continuous research to make the new varieties better able to resist disease and pests. The availability of new technology was a necessary but not a suffi- cient condition for the "green revolution" to occur: changes in gov- ernment policies and priorities were required to strengthen the incentive to modernize. In India, for example, the shock of two poor monsoons helped lift agriculture to the top of the government's de- velopment priorities; greatly increased amounts of foreign exchange were allocated for importing fertilizer, better price incentives were provided, and research, extension and input distribution facilities were expanded and made more efficient. Agricultural production has been increased primarily by concen- trating capital investments, purchased inputs and the new technology on a small segment of agriculture. While this is largely because land and water resources are better in some regions, it also reflects the uneven distribution of entrepreneurship, capital availability and pub- lic investment policies. Concentration of resources where output increases are most readily obtainable is a sound strategy if the aim is to increase marketed output as quickly as possible. But the less urgent this need becomes, the greater is the attraction of a policy whereby the limited supply of capital is combined with expanded use of labor and new technical knowledge over a much larger proportion of farms. Such a strategy would help in a situation where production is con- 9 strained by lack of effective demand while a large proportion of the rural population continues to go hungry because it produces so little. Modernization on small farms will, however, require greater efforts to provide extension and credit services and possibly to develop tech- niques of production that are better adapted to the conditions pre- vailing in traditional agriculture. The adoption of new techniques is likely to have significant effects on the geographical pattern of production, and also on the product composition. Developments to date have increased the comparative advantage of irrigated areas in the production of foodgrains. As pro- duction of these basic necessities increases and their prices fall, the real incomes of consumers will rise and they will increasingly demand other agricultural products-meat, milk, fibers, fruits and vegetables -and agricultural resources previously committed to foodgrain pro- duction will become available for meeting this demand. If economic growth is to be promoted through greater diversification of agricul- tural production, a good deal of investment in research, extension and the marketing infrastructure will be required. Exports Just as agriculture dominates the economies of most low-income countries, so also it provides the largest portion of the foreign ex- change earnings that are so vital for their economic development. Although exports of minerals and manufactures have grown more rapidly over the last decade, agricultural primary products still ac- count for over 80% of the export earnings of at least eight countries, 50-80% in the case of at least nine countries, and 30-50% in the case of another eight or more (Annex 8)1. The major customers for the developing countries' agricultural exports are the industrialized coun- tries; consequently, agricultural production, income and employment for the majority of less developed countries are shaped in part by factors beyond their control. However, public policies in these coun- tries sometimes discourage agricultural exports by reducing producer incentives and discouraging the reinvestment of earnings. Despite some encouraging achievements, the record of less devel- oped countries in expanding agricultural exports has been generally disappointing. The unit value of such exports from both developed and developing countries has more or less stagnated over the last ten years. But, in terms of volume, agricultural exports from developing countries have grown far more slowly than those from developed areas. As a result, the total value of developing countries' agricultural 'Based on a sample of 36 countries accounting for 65% of less developed coun- tries' export earnings. in exports rose by only 30% between 1957-59 and 1970, compared to a growth three times as high in developed countries' exports (Annex 9). It is interesting to note that the more rapid growth in the latter case is accounted for in part by the lowering of trade barriers and expansion in trade among developed countries themselves. The FAO forecasts that the agricultural exports of developing coun- tries (excluding forestry and fish) will grow by about 3% annually over the next eight to ten years. In that case, the growth rate will be a little better than it has been during the past decade. It will need to be higher in order to make a greater impact on the employment problem in developing countries. The differences in labor and capital intensity in production indicate that if some agricultural production is shifted from the developed to the developing world, far more jobs would be created in the latter than would be lost in the former. Demand prospects for many of the typical agricultural products of developing countries, such as rubber, palm oil, groundnuts, coffee, tea, cocoa, sugar, cotton and tobacco, often appear to be discourag- ing. If considered over the medium period of, say, ten years, the scope for increased exports is often limited by growing competition among developing countries, competition from synthetic substitutes, and sometimes competition from subsidized agricultural production in developed countries. Yet in the longer run the prospect of population increases and higher living standards must result in a growing demand for some of the commodities where surpluses now appear to threaten. Uncertainties inherent in making price projections for the longer term cause difficulties in making investment decisions, especially for tree crops with long economic lives. While the export outlook for some agricultural products is not par- ticularly encouraging, for others it is quite good. The latter category includes meat, feedgrains, certain types of fish, fruits and vegetables, timber and paper products. Some of these commodities might be- come scarce unless developing countries exploit their production potential effectively. Another way in which developing countries could increase their export earnings is by doing more of the processing of raw materials themselves: meal and oil might be exported rather than oilseeds, cotton textiles rather than cotton lint, canned or frozen foods rather than fresh products. However, developments of this type require both better access to developed markets and increased investment in proc- essing capacity. It would also be helpful if developed countries elimi- nated some of the existing structure of protection and subsidies, and thus encouraged their producers to shift away from products that can be imported more economically from developing countries. Finally, 11 Income per Capita, 1967, and Percent of Labor Force in Agriculture, 1965, Developed and Less Developed Countries Income per Capita ($)' *United States 3,000 *Sweden Switzerland ll*Canada *Denmark Australia 2,000 U.K- I W. t France Belgium Germany Norway S Netherlands *Israel 0 Finland *Austria (OO *Italy W*Ireland Czechoslovakia- Japan- *Venezuela 801) _ USSROspa'n Argentina* Hungary* **Libya * Greece * Poland 600( Urga Urga Chile *Bulgaria *Jamaica 0 Panama Meic *Yugoslavia 40(1 *Costa Rica *Nicaragua Brazil Peru 0 * Malaysia eGuatemala lordans Colombia E El Salvador *Turkey EcuadorD incnRp Iraq* * DOmincnRp Taiwang //Gana Tun iaHonduras Senegal Morocco *Syria Liberia *Philippines Ceylone * Egypt *Bolivia Thailand.i 0S. Korea Pakistan Sudan. * Indonesia B0 India* Nigeria 60 Burma 0 4() ) 10 20 30 40 50 60 0) 8E) PERCENT OF LABOR FORCE ENMPLOYED IN AGRICULTURE IGr.ss Nat,onal Product per Capita, 1966 Dollars Source: Economic Progress of Agriculture in Developing Nations, 1950-68. (Foreign Agricultural Economic Report No. 59, U.S Department of Agriculture, Washington, 1970) 12 developing countries could stimulate trade among themselves by ex- panding their economies and adopting more liberal trade policies. Although agricultural exports account for a substantial part of total exports of many developing countries, they often represent a much smaller percentage of total agricultural production. In India, for in- stance, agricultural exports are over 40% of total exports but less than 3% of total agricultural output. A similar situation prevails in Pakistan, where the respective figures are 50% and 5%; in Thailand (80% and 25%) and in Ethiopia (98% and 10%). On the other hand, these per- centages are 90 and 50 for Argentina. A few countries may be able to sell a large part of the increase in their agricultural production in export markets. But most countries will have to rely mainly on the expansion of their domestic markets. Employment Developing countries face a pressing need to provide employment for a rapidly increasing population. In agriculture the problem is mainly one of underemployment, which is perhaps best defined as a condition in which irregular work of low productivity results in ex- tremely low levels of earnings and consumption. As mentioned earlier, and as indicated by the table in Annex 2, economic growth is ordinarily accompanied by a decline in the rela- tive importance of the agricultural work-force. Further evidence of this tendency-but also of the existence of large departures in indi- vidual countries from the general relationship-is provided by the figure on the opposite page. It would be unwarranted, however, to conclude that the task of employment creation must fall primarily to non-agricultural sectors of the economy. With rapid population growth, the absolute numbers of those available for employment in agriculture will increase, in many countries, for the foreseeable future. For example, as the table on the following page shows, the agricultural population of Asia will continue to increase until well into the twenty-first century, unless population growth is sharply reduced or the growth of urban employ- ment is greatly accelerated. Although the numbers employed in agriculture can be expected to increase in most developing countries, the rate of increase will differ among countries, depending mainly on the present size of their non- agricultural sectors. The latest FAO projections are that between 1970 and 1985 the agricultural population in East Africa will increase by 31% (from 80 million to 105 million); in South Asia by 24% (from 490 million to 606 million); and in Latin America by 14% (from 118 million to 135 million). 13 Agricultural Population Projections for Asia' Year in Size of which Agricultural Agricultural Assumed Rate of Growth of Population Reaches Population in Population Urbanization Peak Peak Year2 percent per annum million 2.8' 3.86 2052 2,790 2.5' 3.8 2025 1,310 2.3' 3.8 2011 950 2.8 5.0' 1993 820 2.5 5.0 1984 701 2.3 5.0 1979 640 'Excluding Communist countries and Japan. 2Agricultural population in 1962, the base period for these projections, was 583 million. By 1970 it had reached 701 million. 'U.N. high estimate. 'U.N. medium estimate. 'U.N. low estimate. 'IWP projected rate. "'High" rate. Source: FAO, Provisiona/ Indicative World Plan for Agricultural Development (Rome, 1970) Vol. 1, pp. 32-33. With few exceptions, agricultural production in developing coun- tries is carried out by labor-intensive methods: traditional agriculture provides work for many hands, even if earnings are miserably low. The demographic prospect depicted in the table above implies that in many countries sustenance and work will have to be found for many more people in the coming decades. Fortunately, modern tech- niques that increase output can be introduced into traditional agri- culture without changing its labor-intensive character. However, as is argued in detail later, if government policies and the structure of prices favor capital-intensive production techniques and if the small farm sector is neglected, agricultural growth will provide far fewer opportunities for productive employment than it could. Hence there is a need to re-examine existing policies and institutions to see whether they promote or discourage a pattern of development which utilizes labor resources. A policy of creating employment in rural areas should reduce rural- urban migration and the social and private costs associated with it. Furthermore, it is very difficult to reduce urban unemployment by creating more jobs in urban areas if underemployment and low earn- ings remain characteristic of agriculture. For, as the urban unem- ployed are absorbed into the work-force, they will simply tend to be replaced by new migrants from the countryside, attracted by the chance of securing a high-paying urban job. Effects of Technological Change on Employment The introduction of modern agricultural technology has had mixed effects on the demand for labor in Asia, and the net effect on employ- ment in regions where their use is spreading is not always clear. The high-yielding varieties of wheat and rice call for more intensive culti- vation. Moreover, the shorter growing season of the new varieties frequently permits double- or multiple-cropping, thus greatly increas- ing the number of cultural operations carried out in the course of the year. The use of improved seeds and fertilizers is therefore conducive to labor-intensive farming methods, and, of itself, tends to create more employment. The use of the new varieties, however, has been accompanied, for various reasons, by an increasing degree of farm mechanization. Not all mechanization has an adverse effect on employment; the mech- anization of operations that can be performed only slowly or inade- quately using hand methods-or which take place at times of peak labor demand when labor is, in fact, a scarce resource-is often nec- essary in order to break the bottlenecks hampering production. Some examples are the use of tractors for rapid land preparation so as to permit increased double-cropping, of mechanical threshers to handle the greatly increased harvest, and of motor pumps to lift irrigation water. In these cases more intensive cultivation frequently raises the total requirement for on-farm labor. However, once acquired for a particular purpose, some machines- and, in particular, tractors-can be used for other purposes as well, and it is usually economical so to use them. In addition, not all mech- anization that has occurred is of the sort that increases output. Some has been labor-saving in intent or in practice. For these reasons, and also because government policies in many countries often wittingly or unwittingly encourage indiscriminate mechanization, growing concern is being expressed about the extent to which mechanization is proceeding'. There is also a danger that the direct employment benefits of the new technology will, to some extent, be offset by changes in the cropping pattern. Areas planted to crops favored by technological advances tend to expand at the expense of other crops. If the former are less labor-intensive than the latter, the net effect on employment will be smaller; and, if the shifts in crop composition are large enough, employment could even decline. In South Asia some labor 'in an attempt to improve its knowledge of this matter, the Bank has arranged with the FAO/IBRD Cooperative Program for a pilot study of the employment and income distribution effects of the Muda Irrigation Project in Malaysia (which involved sub- stantial farm mechanization, including private machinery pools providing services to small farmers); and with some Indian research units for monitoring the employ- ment, income distribution and productivity effects of farm mechanization under two Bank-financed credit projects in India. displacement has occurred as wheat acreage has expanded at the expense of more labor-intensive crops such as cotton-a tendency that has been accentuated by price incentives working in the same direction. Looking to the future, however, the high-yielding grain varieties could contribute to employment by releasing land and water re- sources for producing other foods, such as vegetables and milk, where much more labor is required. Shifting away from monoculture with its seasonally-peaked demand for labor may also be the most practical solution to the problem of seasonal underemployment. The "green revolution" also generates off-farm employment, in activities connected with supplying farm inputs and consumption goods and services and marketing the larger harvests. In parts of India, for example, there has been a clear rise in the earnings of village craftsmen, such as blacksmiths and carpenters, and small workshop industries have come into being, producing such things as simple mechanical threshers. However, quantitative evidence regarding these secondary employment effects is not available. There is also considerable potential for creating employment by the construction and improvement of irrigation facilities, the productivity of which has been greatly enhanced as a result of the new agricultural techniques. In sum, the widespread adoption of the new, output-increasing agricultural techniques can significantly reduce unemployment and underemployment. However, if their full potential in this regard is to be realized, policies which favor almost exclusively the growth of a small capital-intensive "modern" agriculture sector will have to be modified. In particular, farm mechanization will have to be selective, the emphasis being on the type that is essential for increasing output per hectare. Of course, further development of yield-increasing tech- nologies for a wider range of crops and environments-rainfed as well as irrigated-is as essential for generating employment opportu- nities as it is to meet the demands of the market. As things stand, the fact that the new techniques are mainly appli- cable to irrigated agriculture means that their impact on employment is limited and is concentrated in particular regions. Because of low labor mobility, the expansion of output and employment in favored areas may be slowed by labor shortages, while chronic unemploy- ment exists in other parts of the same country. Nor do the new tech- niques necessarily lessen the seasonality of farm labor-requirements: in fact, they sometimes accentuate the seasonal peaks. Depending on the overall availability of labor, the solution may lie in selective mech- anization, in conducting some post-harvest processing, storing or marketing activities locally, in greater crop diversification, or in the -1 _ careful timing of programs for off-farm employment, such as rural public works. Income Distribution The progress of the agriculture sector-as the provider of subsist- ence and cash income to those engaged in it, and as producer of marketed foodstuffs and fibers-has a very direct bearing on the welfare of poor people in developing countries. The rural poor consist of many millions of subsistence farmers, small farmers, laborers and their families. Their poverty reflects the fact that their productivity is low and that they own few assets in the form of either land or capital equipment. Programs for raising agri- cultural production by using modern techniques benefit the urban poor, since they tend to hold down the prices of their basic neces- sities. But they do not necessarily help the rural poor. The effects on rural poverty depend on many factors, and are likely to be mixed. Initially, the benefits tend to go principally to the larger, more commercially-oriented farmers, who are better able to make the financial outlays required to purchase modern inputs and to as- sume the risks associated with trying something new. To secure acceptance, the new techniques are often subsidized. As they are accepted more widely, the burden of the subsidy weighs heavily on the exchequer, and it is sometimes reduced or removed just at the time when the smaller producers decide to adopt the techniques1. In addition, if the resulting production increases are not matched by rising demand, the prices of agricultural products tend to fall, so that those who have not managed to increase their production are left worse off than before. Similarly, if the new techniques are, on balance, labor-saving, the lot of landless laborers, tenant farmers and owner-operators who supplement their incomes with outside employment is likely to worsen. In general, the effect of the new technology on income dis- tribution will depend upon the pattern of adoption, the pattern of resource ownership, and on the net effect on the demand for labor. It will also depend upon existing redistributory devices, such as taxa- tion of income or land. Since the mass of the poor have few assets other than their capacity to work, they will be benefited by policies which increase the demand for labor; in this respect, the problems of poverty and unemployment are closely related. A development strategy which stresses paid em- ployment will benefit small farmers: more of them or members of 'In some cases, instead of removing the subsidy, governments have limited the sup- ply of the input. 17 their families will secure paid employment and there will be less underemployment of those who remain on the farm. But, in addition, specific programs designed to raise the productivity and the incomes of farmers and farm workers are needed. One way of doing this is by giving them knowledge of new produc- tion methods, together with the means of putting the new methods into practice-for example, better credit and marketing facilities. Another approach is to provide them with more land, through land settlement (if unused land is available) or, in areas where the pattern of land ownership is highly skewed, land redistribution programs. The additional land raises the farmer's labor productivity; moreover, since the land is usually sold on concessional terms, he also benefits in his role of landowner. Experience in a number of countries has shown that the second approach is unlikely to be successful unless it is accompanied by the first; that is, it is not sufficient to break up, through land reform, what is often a semi-feudal social structure and not put anything in its place. Similarly, in some situations, the first approach may make little headway without an admixture of the second, since the size and frag- mentation of land holdings and tenure arrangements may discourage the adoption of new production techniques. SOME MAJOR POLICY ISSUES It is proposed now to focus attention on some important policy issues which many developing countries must face as they strive to promote agricultural development. It is realized, of course, that be- cause of differences among countries in the physical, economic and social environment, the tasks facing governments will vary from country to country, as will the policy instruments that are feasible. However, a few of the issues that are of particular concern in many countries are singled out for attention here. Government activities in relation to the economy fall into three main categories: first, the creation and/or maintenance of an institu- tional framework that is conducive to social stability and economic progress; secondly, the general management of the economy; and thirdly, the provision of public services and investments, particularly for purposes unlikely to be performed by the private sector. In agri- culture, these public services and investments typically include re- search, technical extension services, general supervision of the cooperative sector, and provision of major infrastructure such as sur- face irrigation systems, systematic groundwater development and rural roads. 10 Governmental "management" of agriculture in most developing countries is weak-weaker than that of most other sectors of the economy. This is partly a result of past attitudes which gave low pri- ority to agriculture. Although attitudes have been changing, govern- ment activities affecting agriculture still suffer from insufficient funds and a lack of skilled manpower, and economic policies often militate against agricultural development. The concept of investment projects for agricultural development is still a fairly new one to some governments, and few are well equipped to identify, prepare and implement such projects. There is a reluc- tance to recognize that highly technical and specialized skills are as necessary in agriculture as they are in other sectors. For example, the fact that project management requires professional skills that might have to be hired from abroad is less readily appreciated in the case of agricultural projects than it is in the case of power and trans- port projects'. Resolution of the issues discussed below is the responsibility of the developing countries themselves. The Bank Group and other develop- ment agencies can assist through advice and objective commentary on the effects of existing and proposed policy options-for example, through the Bank Group's sector survey missions and in the course of project work. But the solution to one important problem will depend very much on policies adopted in developed countries. Agricultural exports from developing countries cannot increase significantly un- less developed countries grant wider market access and offer less encouragement to high-cost production by their own agriculture. This conclusion follows from the fact that roughly three-fourths of devel- oping countries' agricultural exports consist of commodities which compete directly with the same products or synthetic substitutes pro- duced in developed countries (see Annex 10). It is disturbing that despite the crucial importance of agriculture in most low-income countries, agricultural commodities remain the category where the least progress has been made after almost a quar- ter century of effort to achieve a worldwide reduction in trade bar- riers. A disposition to view the problem of agricultural development 'It is important to note that in the agricultural field there has been a significant improvement during the 1960s in the services which consultants can supply. Many consulting firms whose competence was formerly restricted to engineering have acquired agricultural expertise and in several cases have set up their own agricul- ture departments. A considerable number of new consulting firms have come into existence and it is now possible to hire suitable consultants for most agricultural work which may need to be done. This gives governments and development agen- cies much more scope and choice when they wish to undertake studies or when they need help in the execution of projects. 10 in a broader perspective is necessary in the interests of both the developed and the developing countries. Making Better Use of the Price System A considerable body of evidence is now available to demonstrate that peasant farmers respond to economic incentives. For example, the supply of many peasant-produced crops has been shown to have responded to changes in their relative prices. Also, it has been ob- served that small farmers are not indifferent to the income-increasing opportunities afforded by the adoption of new technology with a high pay-off. Traditional agriculture is recognized as being a good deal more efficient than it was once thought to be. It follows that the price system can play an important role in allo- cating resources and products in the agricultural sector, and between it and the rest of the economy. If price signals are distorted so that they fail to reflect relative scarcities, resources may be misallocated, to the detriment of production, growth and employment. This is not to say that all distortions are bad-some, for example, may be justified on the ground that they improve the distribution of income; nor is it to say that the resulting misallocations are necessarily serious. How- ever, it does mean that the possibility should be borne in mind that substantial misallocations may be occurring; price, trade and fiscal policies should be examined in that light. Agricultural prices are often the subject of a good deal of govern- ment intervention designed to serve a variety of political, social and economic ends, and the resulting structure of relative prices may not encourage the efficient use of resources. Adjusting prices to bring them into line with relative scarcities and trading opportunities is a difficult task which calls for considerable political courage. It also calls for perceptive and informed economic analysis: in a situation where there are many discrepancies between the actual and an ideal price structure, it is not always apparent whether a particular dis- crepancy is harmful or, by offsetting another distortion, is beneficial. The task is further complicated by the existence of dynamic factors which cause production and trading opportunities to change, and by uncertainty regarding future trends-for example, in the market pros- pects for export commodities. A few years ago, there was a general tendency for the price systems of developing countries to discourage agricultural investment and production. However, in the meantime, a number of countries have substantially revised, or are in the process of revising, their policies affecting agriculture. As a result, the situation is now much more mixed, and agricultural production in several countries is being ac- tively encouraged through price support programs, investment incen- tives and so on. But while price systems that are systematically biased against agri- cultural production are now less common, the problem of price dis- tortions within the agricultural sector remains. This problem may, indeed, be more serious than it was, owing to the tendency of gov- ernment intervention to be directed at specific products, or specific inputs, and also to a failure to adjust prices in directions indicated by technical advances. For example, the policy pursued by many coun- tries of supporting foodgrain prices at levels above world prices de- serves reconsideration in view of the fact that, as a result of the new agricultural technology, unexpected marketable surpluses of grain have been generated. These surpluses give rise to a host of problems, including the strain on government budgets when surpluses are acquired at high support prices; loss of production of other crops, including export crops, because of diversion of resources to grain production; problems of storing grain stocks which cannot be moved at the high prices; and eventually, depression of world grain prices as former importers attain self-sufficiency or become exporters. The problem basically is to es- tablish within the overall structure of market prices a level of price support for grains which appropriately reflects the decline in costs of production. This is a difficult task technically, and even more so politically. Retention of high support prices is sometimes urged as being nec- essary to provide incentives for small farmers, even though small farmers are least likely to have marketable surpluses, and therefore least likely to benefit from the price supports. Similar arguments are invoked to justify the granting or retention of input subsidies. Price supports and subsidies are generally an inefficient way of assisting small farmers since, if payments are proportional to the volume of production, the benefits go mainly to well-off farmers. Non-Price Rationing In many cases, key resources necessary for agricultural develop- ment are allocated not so much by price, as by non-price rationing devices. Much of the credit available for agriculture goes to those who can provide acceptable collateral, i.e., to those with land or other forms of transferable wealth. Small farmers and landless laborers derive little or no benefit from low interest rates, low-cost inputs, cheap power rates, etc., when they cannot take advantage of them because they lack collateral or complementary resources. Ironically, such concessions are usually introduced supposedly to benefit these groups. The problem has been recognized, and the Bank is trying to 11 find ways of dealing with it. But continuing efforts are required to devise allocative mechanisms that avoid the inherent biases associ- ated with the traditional forms of subsidy within prevailing institu- tional frameworks. Irrigation water is almost universally underpriced and allocated by physical rationing. Rules and practices governing the use of water often serve neither efficiency nor equity very satisfactorily, and their inflexibility hinders adjustments to changing production techniques and cropping patterns. Raising water charges to reflect opportunity costs would promote the more effective use of water and at the same time generate additional public revenues which could be used for other development projects. It is possible to devise water pricing systems that promote both equity and efficiency-for example, through discriminatory pricing, and by making water rights transfer- able, so that an allocation of rights on grounds of equity need not prevent an efficient allocation of water. Price Distortions and Unemployment Several types of price distortions which are commonly present in developing countries encourage producers to substitute capital for labor to a greater extent than is warranted by their relative scarcities. The fixing of interest rates at artificially low levels, trade policies that underprice imported capital equipment, such as farm machinery, and minimum wage and other social welfare legislation that raises the price of labor above its supply price, are cases in point. These distor- tions-since they do nothing to increase the aggregate supply of capital relative to that of labor-introduce rationing problems such as those mentioned above. Entrepreneurs who are lucky enough to have access to cheap sources of credit or foreign exchange, or who are subject to minimum wage legislation, have an incentive to organize their economic activi- ties in a capital-intensive manner. The labor that may be displaced either becomes unemployed or finds employment in those sectors- which may be traditional agriculture-that are not favored by access to cheap capital and in which the relative prices of labor and capital thus tend to be distorted in the opposite direction'. The effect is THigh urban wage rates induce unemployment not only through the displacement of labor by machines, but in a more direct manner. A large discrepancy between urban wage rates and rural earnings-provided it is not explicable in terms of higher skill requirements, high living costs, or similar factors-attracts rural people to cities in search of high-paying jobs. The number of such job seekers depends upon the urban-rural wage differential and on how they perceive their chances of getting a job. Creating more urban jobs may do little to reduce the total of urban unemploy- ment, since the pool of job seekers is likely to be replenished by more migrants from the countryside. It can be reduced only by narrowing the wage differential. therefore to accentuate the dualistic nature of the economy, i.e., to heighten the contrast between a technologically modern sector, with high capital-output ratios and labor productivity, and a traditional sector using labor-intensive techniques and providing extremely low labor incomes. The misallocation of resources restrains both total out- put and the demand for labor. The policies just described are economy-wide in their impact. Be- ing the main traditional sector, agriculture is affected if industrial employment does not expand as fast as it could: more people remain in the traditional sectors, resulting in lower labor incomes, under- employment and unemployment. But such policies also foster the growth of dualism within agriculture. As was pointed out above, it is the large commercial farmers who have the easiest access to cheap credit and cheap inputs. They are also frequently subject to minimum wage legislation-whereas traditional farmers are not, or can evade its provisions-and this has a direct effect on such decisions as replac- ing labor by machinery or chemicals (e.g., weed killers) and cultivat- ing labor-intensive crops. The present state of knowledge permits only impressionistic judg- ments concerning the effect of price distortions on the choice of technology; quantitative estimates are lacking. Nevertheless, the cir- cumstantial evidence is too significant to be ignored. Furthermore, the use of capital-intensive methods of production is fostered by the ability of developing countries to borrow capital-intensive technol- ogy from developed countries. Often the developing country's needs would be better served by some intermediate type of technology, but the incentives to develop it are generally lacking1. Policy Implications The best way of avoiding the resource misallocations associated with price distortions is to remove the distortions. This may involve the removal of taxes, subsidies, etc., or, in some cases, the imposition of "corrective" taxes, subsidies, etc. Thus countries should be encour- aged to adopt realistic foreign exchange rates or, where appropriate, offset overvaluation with appropriate rates of import duties or special taxes; to adopt interest rates that adequately reflect the cost of capital; and, whenever they consider it essential to grant subsidies to agricul- tural production, to do it in such a way as to direct these to specific 'it is necessary to encourage research into labor-intensive agricultural production methods, as part of the increasing interest in agricultural research generally. When suitable opportunities arise, it should be possible to support applied research in this field. In the transportation sector, the Bank has recently begun a study of the substitution of labor for equipment in road construction. The results may have con- siderable significance for agriculture in terms of labor-intensive construction of minor agricultural roads and irrigation infrastructure. high priority programs or enterprises in line with the optimal use of resources. Measures such as these could lead to political difficulties. This is, if anything, more true of correcting distortions in the price of labor, since it is not politically realistic to revoke existing minimum wage laws or to subsidize labor use directly, except perhaps to a limited extent-for example, by subsidizing the labor cost of local public works. The best that can be done, perhaps, is to make the maximum prac- tical use of the small farmer as a production unit. The small farmer (defined here as one who uses principally family labor) will tend to displace labor by other factors of production later than would other agricultural producers, since the price he places on his own and his family's labor will represent its opportunity cost, as he sees it. Unfor- tunately, there are practical limitations to expansion of small-farm production (see "Small Farm Sector" below). It is also theoretically possible to offset price distortions in the labor field by deliberately raising the price of labor-substituting inputs; but such action, unless generally applied to all sectors, would tend to discourage agricul- tural production. The use of additional taxes or subsidies to correct price distortions is limited by the political difficulties which governments face in rais- ing levels of taxation or public spending. Hence, close attention should be given to existing fiscal arrangements, to see if corrective taxes or subsidies can be substituted for distorting ones, within the given limits of governmental budgets. Price Distortions and Public Investment Price distortions particularly affect private investment decisions. They are, therefore, important in agriculture, where there are a large number of private producers. In the public sector, investment deci- sions need be less influenced by prices, since the desire to maximize profits or minimize costs is moderated by social considerations. In- deed, the influence of price distortions on public investment planning can be removed by the use of techniques such as shadow pricing, whereby prices intended to reflect the true relative scarcities of prod- ucts and resources are used to assess the costs and benefits of investments. There are some practical difficulties with this approach, however. A basic difficulty is that since many price distortions are the result of government policies, a high degree of sophistication is required of governments if they are to negate these policies in their public invest- ment activities. Another difficulty is thatwhile selecting and executing investment projects on the basis of shadow prices reduces the social costs or increases the social benefits, it has the opposite effect on money costs and benefits, which depend on the prevailing market prices. Governments facing severe budgetary constraints may, therefore, prefer to select projects and construction methods which are cheap in, terms of ruling prices. They may be reluctant to launch, say, labor- intensive public works programs the opportunity cost of which might be low because of unemployment in the economy, but which might be quite costly on a cash basis'. Project selection on the basis of shadow prices is likely-given a fixed budgetary constraint-to result in a smaller works program, but production elsewhere in the econ- omy will suffer less since the program will tend to use resources, such as labor, that are underutilized in the private sector. Total output is thus likely to be greater than if the larger works program were imple- mented; furthermore, the use of labor-intensive works will improve the distribution of income. In some cases the price distortions in the public sector may be more severe than in the private sector, since many public-sector agencies enjoy special concessions, such as duty-free imports, favorable ex- change rates, and interest-free capital. The price distortions are sometimes so great that although particu- lar agricultural projects are financially attractive, they are neverthe- less unsound for the economy as a whole. The Bank, therefore, requires that the projects it finances should provide a satisfactory rate of return to the economy, taking shadow prices into account. While this procedure can weed out some clearly undesirable projects, it does not assure the selection of the projects with the highest priority. This is particularly the case when labor resources are abundant. The Bank is, therefore, increasing its sector work in order to help borrowers identify priority investments. It is playing a more active role in the identification and preparation of projects, so that relative re- source endowments are considered at an early stage. The Bank's early involvement in project identification also presents the oppor- tunity for introducing income distribution and employment as impor- tant criteria in the project selection process. Small Farm Sector A major policy issue in almost every developing country is how to raise the incomes of the rural poor-small farmers, landless laborers, 'In some instances, the budgetary cost of labor-intensive public works projects has been reduced by persuading local people to donate their labor services, or work for small rewards; and by drafting people, e.g., members of the armed forces, to work on them. the unemployed, and their families. These groups represent a large proportion of both the total and the rural population; they constitute the poorest segment of society, and they have been largely bypassed by the prdgress of the past decade. Policies and techniques for raising their incomes and productivity have not yet been devised for appli- cation on a massive scale. Yet development cannot have much mean- ing if it does not include the alleviation of rural poverty. The size of the problem is immense. About 700-800 million-about one-third of the total population of the developing world'-are eco- nomically deprived rural people. The gap between them and the rest of the population will continue to grow unless determined efforts are made to better their lot. There is considerable potential in the small farm sector for reducing underemployment, alleviating poverty and increasing agricultural production. The marketable production of individual units is small, since most small farmers live at or near the subsistence level, but their combined output constitutes an important, and sometimes dominant, share of total agricultural production. Because productivity is low, incomes are low even though they may be supplemented by small farmers and their families undertaking off-farm work. Changes in technology and access to services would boost productivity and in- comes, and serve to reduce underemployment. Even marginal in- creases in productivity in the small farm sector would make a substantial contribution to total output, though the total growth in output is constrained by the growth of effective demand. There has been limited success in bringing the benefits of develop- ment to this economically-disadvantaged group. Several governments have initiated programs to combat poverty in rural areas; some ex- amples of projects designed to help the small farm sector in which the Bank has participated are described later in this paper. The scope of the problem is enormous in terms of the number of families involved and the capital and institutional services required. Moreover, the rural poor have the highest illiteracy rate, the worst health conditions and the poorest access to markets, all of which make innovation and investment more difficult. The average invest- ment per farm family in a selected group of recent Bank-financed projects was $1,100. This investment level cannot be sustained in a world where there are more than 100 million small farm families. The total investment in agriculture in developing countries is estimated at $7,000-10,000 million per year, which would be equivalent to less than $100 per small farm family if all the investment was for them. The analogy of the housing problem in developing countries may 'Excluding Mainland China. 26 be instructive. In that area, too, the capital cost of even minimum housing units far exceeds the available resources. The emerging solu- tion is a "sites and services" approach, whereby governments provide the infrastructure but the urban poor have to construct their own dwellings. A comparable approach-substituting idle labor for capi- tal-is indicated for the rural poor. However, even the provision of extension, credit and other services to enable small farmers to increase their productivity requires massive resources. The total supply of trained personnel is grossly inadequate; moreover, only a small proportion of the staff available at present provides services for the rural poor. In addition, the rural poor are less well supplied with other public services, such as education, health, electricity and water. Since growth in agricultural production is unlikely to exceed 4-5% per annum, and the rural population is expected to increase at about 11/2% per annum for some decades, production per head in agricul- ture cannot be expected to increase by more than about 3-31/2 % per year. The increases in net income will be less since more purchased inputs will be required to boost production. But these are average figures. Unless new measures are devised, the increases will be con- centrated among the larger farm units and the real incomes of the rural poor will change very little. There are no ready-made, generally-applicable solutions to the problem. If maximum impact is to be achieved, programs will have to be devised which encourage initiative and self-help among small farmers and increase their productivity. However, increases in pri- mary production alone cannot provide sufficient opportunities for increasing the incomes of small farmers and others among the rural poor. It is proposed to outline here some of the difficulties and opportu- nities in the development of the small farm sector. While the problem is clear, the framing and implementation of programs to assist small farmers remain a major challenge to the ingenuity of those engaged in agricultural development. Economies and Diseconomies of Scale Many agricultural operations can be performed quite efficiently on a very small scale. For example, from the standpoint of crop biology there is no reason for yields to change according to size of field or of farm. However, some inputs require a certain minimum scale of op- erations if they are to be used efficiently. This is true of machinery of various types and of certain fixed investments, such as tubewells. Similarly, a certain minimum size of herd or flock is needed for effi- cient livestock breeding. 27 It is often possible for small farmers to share the services of these inputs by the creation of machinery pools, by contract hire, by setting up public or cooperative tubewell systems serving several farms, and by group ranching. However, such arrangements create organiza- tional and managerial problems, and there have been many failures. Miniaturization of machinery is also possible, but usually at the cost of some loss of efficiency. In general, then, small farmers are at some disadvantage because of technological economies of scale, but usu- ally the disadvantage is slight. On the other hand, financial economies of scale in agriculture are often quite significant. For example, the small farmer suffers, some- times considerably, from being able to purchase only small quantities of inputs, and to sell only small quantities of produce at a time. He also finds it hard to obtain credit on reasonable terms: he usually has little security to offer against loans and finds it difficult to make an initial down payment in cash. Banking institutions tend to find it too expensive to administer loans to small farmers because the amount of each loan is small. Since it is difficult to get credit on reasonable terms for storing their output, small farmers usually are forced to sell their produce at harvest time when prices are normally low. The ratio of labor to land and capital on small farms tends to be high. Off-farm employment opportunities are often lacking, limited or inconvenient to exploit, so that family labor tends to be used on the farm, even if its marginal productivity is low. This tendency is reinforced by the willingness of small farmers to work hard, under the spur of poverty. The result is that labor-intensive production tech- niques tend to be used, and labor productivity is low. However, the productivity of land in small farms is frequently as high as, or even higher than, in large farms'. Provided that he is not too handicapped 'A number of studies have shown that the value of output per hectare, on average, is higher on small farms than on large farms. If this is because large farms tend to occupy the poorer land, or because small farms concentrate on labor-intensive high-value crops, the finding would be of little significance for agricultural policy: after all, subdividing large farms would not raise their natural fertility, nor could all land be devoted to intensive horticulture. However, when careful investigations have been made in which land quality and product-mix have been held constant, some difference in favor of small farms often remains, though in some cases it dis- appears. There is thus some evidence that small farmers use land more efficiently. However, the relevance of this evidence may be questionable since most such studies refer to earlier periods when the application of modern technology was not common practice. A priori reasoning suggests-and experience tends to confirm-that yields per hectare may be greater on small farms for some particularly labor-intensive crops (such as fruits and vegetables and tobacco), but higher on large farms for crops (such as sugar, rubber and oil palms) where sophisticated technical management is required. Yields per unit area can be an ambiguous guide to intensity of land use, however, since as commonly quoted they do not take account of double-cropping or land left fallow. 28 by other disadvantages-in particular, as discussed below, by re- stricted access to modern technology-the small farmer can, through hard work and careful management, use land efficiently. Adoption of Innovations Small farmers tend to be slower to adopt innovations than large farmers. Such behavior does not imply that the profit motive is lack- ing; rather, it arises from the objective circumstances of the situation. Most innovations involve risks and a small farmer can ill-afford to take additional risks when the annual crop is all that stands between his family and starvation. The tactic of adopting the innovation on a small experimental scale, to see if it works, is hardly available to him when the scale of his whole operation is so small: it is far more rational for him to wait and see how others fare. New technologies also usually require increased expenditure on cash inputs, such as improved seeds, fertilizers and weedicides. The small farmer cannot always afford such expenditures, and his access to credit is limited. If he does secure credit it is likely to be from a moneylender and to carry a high interest charge, so that he will want to be reasonably sure of a high pay-off before he borrows. For these reasons, large and medium-sized farmers are at an ad- vantage in a period of rapid technological change. Some evidence from India illustrates the point. Before the "green revolution," small farms commonly showed higher yields per hectare, but subsequently the reverse relationship has been observed, owing to the larger farmer's greater propensity to adopt the new yield-increasing meth- ods. The large farmer will have a permanent advantage if there is a continuing flow of potential innovations from research and develop- ment activity-even though the new technology itself is neutral to scale. Extension services, credit agencies and similar institutions find it easier to deal with a few large farms than with many small ones. Apart from the fact that administrative costs are lower, large farmers tend to be more receptive to advice and more creditworthy; they are also less likely to be isolated by social barriers or by illiteracy. While large farmers can play an important role in demonstrating the effectiveness of new techniques, deliberate efforts are required if governments are to reach small farmers through extension services and credit agencies. Policies which neglect the small-farm sector tend to widen income disparities, sharpen social tensions in rural areas, and do little to assist the bulk of the rural population. In a situation of population pressure and widespread unemployment, small farms have the not 29 inconsiderable social virtue of providing more employment per unit area than large farms. They are also capable of farming their land intensively. If they are offered the knowledge and the means to use new techniques, the social advantages of small-scale farming can be gained without harming production. However, a substantial effort is required to plan and implement schemes for developing the small farm sector. It is seldom useful to offer only credit without also providing complementary facilities such as extension services and marketing infrastructure (including roads and physical facilities as well as marketing institutions). Such integrated schemes are thus apt to draw heavily on the financial and manpower resources of governments; and it is easy to succumb to the temptation to provide less than the minimum necessary for suc- cess (sometimes, on the other hand, such schemes are provided with an excessive share of scarce resources in order to push them through). Land Policy The pattern of land ownership and the contractual and customary relationships between landlords and tenants can have a considerable influence on production, employment and income distribution in agriculture. In developing countries, land represents a much higher proportion of total wealth than in developed countries, and inegali- tarian patterns of land ownership are a major source of income in- equality. Furthermore, the owners of land usually possess political and economic power which can be exercised in ways that harm the interests of the bulk of the rural people. Land reforms aimed at securing a more equal distribution of land have been prosecuted successfully in a number of countries, though in some cases at the cost of an initial drop in agricultural production. Some land reform measures have been little more than political gestures and have been ineffective, or even counterproductive. Whether measures designed to break up large holdings-such as land distribution, progressive land taxes, or the imposition of ceil- ings on the size of individual holdings-will be undertaken depends on what society considers equitable, and on the balance of political forces. However, the need for productive efficiency and employment creation also has to be considered. Although a landholding is not necessarily the same thing as a farm operating unit, the pattern of land ownership does affect the way in which land is used, and does so in two main ways. First, because the amalgamation or sub-division of owned parcels of land to form operating units involves transaction and coordination 30 costs', there is a tendency for landholdings and farm operating units to coincide. If the economies or diseconomies of scale are strong, operating units will tend to become of optimum size, despite the transaction and coordination costs. But the economies of scale gen- erally are not very strong. It is the pattern of land ownership, there- fore, that largely determines the size of operating units. Size, in turn, influences the character of farm operations. With small holdings, the agriculture is likely to be more labor-intensive, with lower out- put per man but probably no less output per acre. It will also require more publicly-provided credit and extension services but probably use less purchased inputs. Second, the incentive to improve the land and use it efficiently depends in part on whether the person who cultivates the land also owns it; and, if not, on the contractual arrangements between the owner and the operator.' Ownership and operation will tend to be divorced if the size of the owned parcel of land is very different from the optimum-sized operating unit. (Absentee ownership also implies the separation of operation from ownership.) Under tenancy arrange- ments there is frequently little security of tenure, nor are there guar- antees that the tenant will be compensated for the improvements he may make on the farm. His incentive to invest in farm improve- ments is thus blunted. Similarly, the tenant's or manager's incentive to farm efficiently is reduced if he does not share fully in the rewards of efficient operation. These deficiencies in contractual relationships are usually not sim- ply the result of landlords and tenants being thoughtless, but reflect the fact that there are difficulties and costs involved in devising ten- ancy contracts that preserve all the incentives. Tenancy reforms which give the tenant greater scope for exercising entrepreneurship usually reduce the landlord's entrepreneurial role correspondingly, so that the gain in efficiency is not clear-cut; there may, in fact, be a loss. Tenancy is a device whereby landlord and tenant combine their resources in a common enterprise: there are bound to be conflicts of interest and some blunting of incentives. Owner-operated farming is often favored because it avoids the need to reconcile the interests of owners and operators through 'These include the cost of negotiating contractual arrangements, each party's un- certainty regarding the competence and good faith of the other, and, in general, all the monetary, psychic and time costs involved in the association of two or more persons in a single enterprise. These costs are the basic source of the disincentives to efficiency which inhere in many rental arrangements and which are discussed in the next paragraph. 'Communal ownership of land is not discussed in this paragraph. However, it should be noted that in some cases it provides sufficient incentives to cultivators. 31 rental and other contractual arrangements. It is also favored because, with the rent component of farm income accruing to the operator, a more egalitarian pattern of income distribution is generally pro- moted. The inclination to provide land to the largest possible number of small farmers means that the size of the operating unit tends to become as small as is compatible with reasonably efficient operation and the provision of a socially-acceptable minimum standard of liv- ing for the operators. More recently, as the implications of demographic changes have become clearer, the principal virtue seen in small-scale family farm- ing is that it can provide employment and, to a lesser extent, can economize on purchased inputs. If small holdings tend to be more socially-productive than large holdings-in that they produce as great an output per unit area using more of the abundant and less of the scarce resources-the conflict which many have perceived between an equitable distribution of land and efficiency in farming may be largely illusory. Belief in the existence of this supposed conflict was often bol- stered by an exaggerated idea of the economies of scale that could be realized in agriculture. Some governments therefore favored the establishment of large-scale cooperative or collectively-owned farm enterprises. However, the management of such enterprises has often been weak. Current approaches favor retaining owner-operated pro- duction units but assisting them through cooperative, collective or state-owned organizations that can realize economies of scale and exercise bargaining advantages in the provision of off-farm and some specialized on-farm services. Whatever size of farm it is desired to promote, it is not merely a matter of breaking up large holdings; some farms are so small that they could never support a family, even if modern techniques are used to increase production. Such units can be viable if suitable opportunities exist for off-farm employment but, where they do not, land reform measures are required to bring farm sizes up to some minimum level. This can sometimes be accomplished by combining small farms with land that is being redistributed: in Tunisia, for example, land taken over from colons was available for redistribution. But since it is politically difficult to enlarge holdings by dispossessing some small- holders (unless they can be induced to leave to take up other jobs), often the best that governments can do is to try to prevent the situa- tion from getting worse. Inheritance laws which prohibit the sub- division of land below a certain size of holding can help in this respect. Small farmers suffer from a number of disadvantages, of which perhaps the most serious are meager financial resources, restricted access to credit, and the consequent reluctance to adopt new tech- niques. Land redistribution therefore needs to be accompanied by the extension of credit, marketing and technical advisory services to small farmers. Often these services or functions were previously performed by landlords. If they are not provided, the impact of land reform on production, and possibly also on the welfare of the new smallholders, is likely to be adverse. The Impact of New Technology New agricultural technology which increases output per unit area is essentially land-substituting in character, since it enables a given output to be produced from fewer hectares of land. Its widespread adoption will tend ultimately to diminish the relative importance of agricultural land and the share of agricultural revenue attributable to land. On a global basis this tendency does exist. But in regions that have been favored by the new technology, land values have risen. For example, the introduction of the new high-yielding varie- ties has increased the value of wheat land in many parts of Asia. However, in areas where the new. technology has not been applied, the value of wheat land will tend to fall-since the overall increase in wheat production will tend to depress wheat prices. The danger therefore is that in regions where the new technology is being adopted, income inequalities may worsen-the tendency being re- inforced by the fact that the larger farmers are usually the first to adopt the new techniques. Far from reducing social tensions in rural areas, the spread of the new technology is likely to sharpen them, and lead to greater demand for the implementation of measures, such as land reform, for the redistribution of income and wealth. Use of the new technology may often be accommodated only with difficulty by existing tenancy arrangements. Although the latter are not without their deficiencies, they are often reasonably well adapted to the conditions of traditional agriculture. But they may need extensive revision if the new technology is to be adopted and its benefits equitably shared. For example, under crop-share tenancy -which, as compared with cash tenancy, or the use of hired labor, serves the purpose of dividing risks between the landlord and the tenant-each party has an incentive to supply too little of the inputs for which he is responsible. In practice, it appears that this tendency is often held in check so that each tends to adhere to customary norms of performance: at any rate, there is little evidence that share- rented farms are operated much less efficiently. However, new technology, particularly since it involves increased 33 use of purchased inputs, will require a new set of commitments from landlord and tenant, and perhaps also changes in the customary shares into which produce is divided. Until the revisions are made, the incentives to adopt new techniques that depend on the use of purchased inputs may be seriously impaired. Furthermore, it appears that rather than entering into revised tenancy agreements, landlords in areas favored by the new tech- nology have been tempted to let their tenancy agreements lapse and to farm the land themselves, using hired labor and machinery. Several considerations probably explain this behavior: it may be easier for landowners to introduce new techniques by directly super- vising hired workers; being uncertain about the yield increases that can be attained, they may have found it difficult to negotiate revised tenancy agreements; and legal protection of tenants-or the threat of it-may have prevented the negotiation of contracts acceptable to the landlord. Another possible explanation is simply that modern agriculture is intellectually more engaging and financially more re- warding than traditional farming. There is a good deal of fragmentation of landholdings in some countries, especially in South Asia. Under traditional agriculture and with underemployment of labor, fragmentation probably does little to constrain production, and reduces risks to some extent. The effects of fragmentation are more serious under modern agricultural con- ditions since it increases the costs of distributing inputs, of appraising credit requests, and of arranging for custom services. It becomes highly inconvenient when labor is in short supply, as now happens in some areas at harvest time. The inconvenience of fragmentation is greatest in irrigated areas, since it raises the costs of distributing water (both directly and through the loss of cultivable land to additional canals). Also, labor shortages at harvest time are more likely to occur in areas served by irrigation. Consolidation of land is difficult to arrange in practice; farmers are usually reluctant to participate voluntarily because they fear that the land they receive may be less valuable than the land they give up1. Implementation of Land Reforms The historical record shows a tendency for land reforms to be enacted in law but not successfully-or only partially or slowly- 'Voluntary participation may be more readily encouraged if consolidation is accom- panied by land improvement works, such as irrigation or land levelling within exist- ing irrigation schemes. Since all participants are likely to benefit, each may be less concerned-so the argument runs-as to whether he receives precisely the same share of improved land as he had of the original unimproved land. 34 implemented in practice. Where the program has been vigorously prosecuted, the results have sometimes been adverse, in that agri- cultural output has fallen and the condition of those whom it was intended to benefit has not been improved, or has even deteriorated. The record is, in a sense, unfair to land reform in that it has often been implemented in times of great social tension, and against strong opposition,-by radical governments. But even in a favorable environ- ment, successful implementation requires a good deal of political courage and administrative finesse. Land reform is expensive, both in terms of the fiscal cost of com- pensating the former owners, and of the human and material re- sources required to administer it properly and to provide services to the newly-created small holders or cooperative enterprises. It also suffers from perverse announcement effects: owners faced with the threat of losing their land-usually compensation is incomplete, so that a degree of expropriation is involved-cease to improve it and allow existing facilities to deteriorate; and laws which promise ten- ants security of tenure lead landlords to dismiss their tenants and replace them by hired laborers and/or machinery. Thus a difficult dilemma arises if the adverse announcement effects are to be mini- mized: the redistribution of land should be done rapidly, but a piece- meal approach is more in keeping with the limited fiscal and administrative resources of governments. Attempts to bring about land reforms have sometimes been frus- trated by vested interests, and governments may feel they have to proceed cautiously because production might be disrupted in the initial stages. In some cases governments, bowing to the pressure for land reform, but lacking the will to take effective action, have en- acted token legislation. However, such political expediency usually worsens the existing situation, as landlords take evasive or defensive actions to the detriment of both production and the employment opportunities of tenants and rural workers.1 It is clear that agricultural development cannot do all it might to improve rural life if the distribution of land ownership is highly skewed. Despite the political and administrative difficulties involved, governments should pursue land policies which help disperse the benefits of agricultural progress widely rather than allowing them to 'A study of the impact of land reform legislation on two Indian villages showed that although laws were framed to improve the position of tenants (by limiting the amount of rent payable, and by giving them greater security of tenure and the right to purchase land) the results left them worse off. Owners were able to evade the legislation by various means, including the replacement of tenants by hired laborers. It is interesting to note that a new tenurial relationship-a form of quasi-tenancy (but not legally classifiable as tenancy)-was devised to meet the situation. 35 go to only a small segment of the population. Steps should at least be taken to tax the higher rental incomes arising from public invest- ments, and to reinvest the proceeds in infrastructure so as to increase the income-earning opportunities of wider strata of the population. If this is not done, rising land values will make it increasingly difficult to redistribute the land later. Although the Bank's overall influence on these matters is bound to be slight, it can take explicit account, in its appraisal work, of whether the benefits of a project are likely to be widely dispersed or captured by only a few individuals. The burden of financing the compensation paid to owners under land redistribution programs is the responsibility of the governments concerned, but the Bank can assist in financing other programs designed to raise the productivity of those to whom the land is distributed. PAST BANK GROUP ACTIVITIES IN AGRICULTURE The Bank Group's activities in the agricultural sector take various forms. The most obvious and easily measured is direct lending for agricultural projects. But a substantial proportion of the lending in other sectors-such as transportation, electric power, education and industry-also benefits agriculture indirectly. Analyses are made of the agricultural sectors of the economies of borrowing countries. Technical assistance is provided mainly, but not exclusively, through the preparation, appraisal, supervision and post-evaluation of proj- ects. Finally, support is given for economic and agricultural research. The agricultural sector is among the most difficult of any to "man- age." It is thus a sector in which aid donors, including the Bank, have particular difficulty in agreeing with governments on satisfactory project lending programs. Decision-making at the farm level is in the hands of a very large number of small operators. The political repercussions of agricultural policies are usually wide, particularly in the case of food prices. The Bank does not claim to have answers to the multifarious prob- lems posed by agricultural development. The patterns of past agri- cultural lending described here are not a precise guide to the future. A great deal of innovation will be required, which may in time lead to considerable deviations from present patterns, and from the pro- jections about future lending in the section entitled "The Bank Group's Future Role." 36 Capital Transfer Direct Lending for Agriculture The Bank's direct lending for agriculture has accelerated rapidly in recent years. By June 30, 1971, it had reached a cumulative total of about $2,500 million. A breakdown by subsector, area and period is given in Annex 11. Until the early 1960's the importance of agriculture in overall eco- nomic development was not fully recognized. This tended to limit the scope for Bank operations in the sector. Thus, over the period FY1948-63, the Bank's lending for agriculture amounted to only $628 million for 71 projects; that sum represented 8.5% of its total lend- ing. In 1963, the Bank decided to increase its support for agricultural development. The increase was facilitated by the fact that the Inter- national Development Association (IDA) had been established in 1960; with IDA, it became possible to expand activities in countries which were particularly poor and in which agricultural development was frequently of the highest priority. The effort to increase support for agriculture was initially ham- pered by a shortage of well-prepared projects. The Bank began in- creasingly to provide assistance for the identification and preparation of projects. As a result, its direct lending for agricultural development over the five years, FY1964-68, amounted to almost as much as dur- ing the previous 16 years: about $600 million for 46 projects. The proportion of agriculture in the Bank's total lending rose to 12%. In 1968, the Bank decided to accelerate its agricultural lending further. The target adopted was that over the five years, FY1969-73, it should total about $2,400 million, or four times as much as in the previous five years, FY1964-68. By the end of FY1971, 60% (or three years) of the time had elapsed, and around 50% of the target had been reached. About $1,200 million had been committed for 96 projects, which was almost equal to the total agricultural lending during the previous 21 years. In the period FY1969-71, the share of agriculture in total lending rose to almost 19%. Enough projects are now being processed to enable the FY1969-73 target to be met. Lending in Indirect Support of Agriculture A considerable amount of Bank Group lending in other sectors supports agriculture indirectly. While there are a number of con- ceptual difficulties in preparing precise data, some orders of magni- tude can be given. Of particular importance in this context are the investments in road transportation, especially in countries that are predominantly agricul- tural and where much of the traffic is therefore related to agriculture. '7 Over the past five years, the Bank has invested about $1,350 million in the highway sector. A large part of this investment is helping to finance the construction and improvement of some 30,000 km of roads. Over two-thirds of these roads, in terms both of length and of cost, are directly benefiting the agricultural sector. Thus, * about 2,000 km are low-cost roads which are virtually part of agricultural development schemes; * about 5,400 km are rural roads where the benefits to agriculture represent an important part of their economic justification; and * about 14,400 km are roads where the benefits to agriculture have been identified although not separately estimated. Investments in other modes of transport, such as railways, have also helped agriculture considerably. But in their case the benefits cannot be so readily assessed in sectoral terms. The Bank Group has invested significant amounts in agriculture- related industries. Such industries include those that use locally- produced raw materials like cotton or forest products, as well as those that produce agricultural inputs like fertilizer for local use. By the end of FY1971, the Bank and IDA had lent about $340 million directly for such industries. The International Finance Corporation (IFC) had invested another $157 million. Over the period FY1968-71, the Bank Group as a whole channeled a further $213 million into agriculture-related industries through local development finance companies. A sub-sectoral breakdown of these investments is given in Annex 12. Many of the educational facilities supported by Bank lending are in rural areas. In addition, since 1968 about $40 million has been specif- ically lent to support agricultural education, training and research. Trends in Bank Lending for Agriculture Diversification of Lending The most notable trend in the Bank's lending for agriculture, par- ticularly since the early 1960's, has been the diversification beyond basic irrigation infrastructure into on-farm activities, technical serv- ices and related rural development (such as rural road construction, processing industries and, so far to a minor extent, education and health services). Increasing emphasis is being given to rain-fed, rather than irrigated, agriculture. The Bank is becoming more heavily in- volved in storage, marketing, seed multiplication, forestry and fish- eries projects. Individual projects of all types are becoming more comprehensive in nature, the stress being on general rural development. 38 The volume of lending for irrigation projects continues to grow. The annual average has risen from $23 million during FY1948-63, to $53 million in FY1964-68 and $140 million in FY1969-71 (Annex 11). These figures reflect the priority that continues to be given to irriga- tion projects in those parts of the world (such as East and South Asia, the Middle East and North Africa) where resource conditions require it. On the other hand, the share of the Bank's agricultural lending devoted to irrigation projects has declined steadily. It has fallen from nearly 60% in FY1948-63, to 44% in FY1964-68 and 34% in FY1969- 71. The decline is even more marked when it is considered that recent irrigation projects have increasingly involved items other than irriga- tion infrastructure. The diversification beyond basic irrigation infrastructure results from: (a) the need to follow-up investments in irrigation infrastructure with credit for on-farm development, so that farmers can take fuller advantage of the water made available to them; (b) the greater need for the multiplication of improved seeds, and for increasing storage and processing capacity, particularly in Asia, as a result of the "green revolution"; (c) the good export prospects for beef and fresh fruit and vege- tables; (d) the availability since 1960 of IDA funds, allowing investments in many of the poorer countries which typically depend heavily on agriculture. In many of them, particularly in Africa, irrigation has so far had a relatively low priority. Investment in African agriculture has involved mostly small farmers requiring projects specifically designed to overcome the constraints on small farm development'; and (e) population growth has encouraged settlement of new lands, which has often become possible as a result of other developments, such as disease control. Irrigation Development Through the 1950's the Bank's lending for irrigation was primarily for the construction of dams and main distributary canals. Govern- ments, farmers and local entrepreneurs were responsible for financ- ing, without further Bank assistance, the complementary investments required to take advantage of the improved water supply2. However, it was found that, in many cases, although progress on the major 'For example, all nine agricultural projects in Africa financed by the Bank in FY1971 involved the small farm sector. 'For example, field ditches and drains; land levelling; input supplies; extension serv- ices; and additional processing and marketing facilities (including roads and farm tracks). 39 construction works was generally satisfactory, the complementary developments were delayed. The economic benefits of the capital- intensive works, therefore, were either not fully realized or were realized rather late. This was often attributable to a lack of coordina- tion between various agencies involved in the development of irri- gated agriculture and to a lack of appropriate financing. In recent years, therefore, the Bank has given particular attention to these complementary developments during project preparation and appraisal. It has sought to ensure that the planning of agricultural and ancillary developments has been carried out, and that suitable arrangements have been made for implementation, including those required for proper coordination. In some cases, especially those where the construction and consequently the disbursement periods are long, the Bank has financed irrigation development in phases or, where appropriate, has financed only the major civil works. But the trend is towards paying equal attention to all aspects of the development and use of water resources, and financing the agri- cultural and ancillary developments as well as the basic irrigation infrastructure. Some recent irrigation projects have thus provided for the construction of the complete distribution system down to the field level; the construction of roads and tracks; the strengthening of applied research, demonstration facilities and agricultural extension services; the provision of credit to entrepreneurs, cooperatives and farmers for input supply, on-farm investments and storage and proc- essing facilities; and training of farmers as well as of operations personnel. The broader approach has helped to make it possible to realize more quickly the full economic benefits of the capital works. But often, many other problems have remained, the resolution of which has called for changes in government policies. It is still frequently the case that the organization of government services within irriga- tion projects is not suitably coordinated. This is particularly true of coordination between the entities directly responsible for infrastruc- ture construction and those responsible for agricultural extension services and applied research; and between the latter and agricultural development banks. Similarly, governments need to attach much more importance to the efficient control of scarce water. The introduction of double- cropping-which often forms one of the principal economic justi- fications for the irrigation infrastructure-generally continues to be disappointingly slow, especially where large numbers of small farm- ers are involved. Some governments still insist on making the water available over as large an area as possible, in order to reach the maxi- A (n mum number of farmers. As a result, individual farmers may be left with too little water for intensive cultivation, and there is the danger that the soil may deteriorate. If cropping intensities remain relatively low, the opportunities for fuller, round-the-year employment also suffer. The scope for increasing cropping intensities by systematic and fully integrated development of surface and groundwater re- sources deserves greater attention. The Bank frequently has encountered situations where dams have been constructed, but developments "below the dam" have still to come, leading to a growing number of projects for the improvement and completion of existing systems. In some cases, the projects are for the rehabilitation and redesigning of very old systems; in these, modern methods of water control and drainage are introduced to increase efficiency and thus increase cropping intensity. In other cases, the projects are for completing distribution and drainage sys- tems which have not been developed to their full potential. In yet other cases, the principal objective is to increase the supply of water through an existing distribution system, converting what might have been designed originally as a drought-protection system into one allowing intensive irrigation and double-cropping. Here, systematic and integrated ground water development has an especially impor- tant role to play. The types of projects described in the preceding paragraph gener- ally involve relatively little capital investment, but bring high returns, and are thus of high priority. One reason why governments do not give them priority more frequently is that they involve farmers who have already received some benefits from public investments (e.g., in the original system), whereas governments often prefer to give priority to new schemes which generally are more prestigious, even though such schemes are likely to be more capital-intensive, yield benefits only in the more distant future and bring lower economic returns. Agricultural Credit Reference has already been made to the diversification of the Bank's agricultural financing into support for on-farm investments and storage, marketing and processing, usually provided in terms of agricultural credit facilities. Ultimate borrowers have included indi- vidual farmers, groups of farmers, cooperatives and public-sector entities. It is not feasible for the Bank itself to invest directly in a large number of small enterprises. So Bank funds provided in support of credit facilities are usually channeled through intermediary insti- tutions within the borrowing countries. 41 These institutions range from agricultural development banks to those parts of the commercial banking system involved in extending medium and long-term credit facilities. Sometimes, especially if there are no suitable intermediaries, it is necessary to establish channels of credit for on-farm development within project authorities, often newly created; the latter may be involved simultaneously in giving technical assistance to participating farmers and in constructing infrastructure. The share of Bank funds lent in support of agricultural credit has grown very rapidly. It has risen from an annual average of $2 million (5% of Bank lending for agriculture) over the period FY1948-63 to about $56 million (47%) in FY1964-68 and about $258 million (63%) in FY1969-71. The range of investments financed by the Bank through credit institutions covers practically all aspects other than the transfer of existing resources such as land. The principal items include: on-farm land development (groundwater development, land levelling, field ditches and drains, land clearing); other on-farm investments (tree crop planting, pasture improvement, livestock purchase, farm ma- chinery purchase, building construction, fencing, on-farm processing and storage facilities); incremental working capital needs for initial purchases of additional inputs (such as fertilizers, pesticides and fungicides); and off-farm developments (processing and storage enterprises, fishing vessels and gear). About half of the Bank's financing of private-sector agricultural investments has been for projects where the provision of medium or long-term credit facilities is the sole or principal element1. Of course, this is only possible where the other ingredients essential for agricul- tural development are already available or are to be provided from other sources. To be successful, such projects depend on active de- mand for the credit facilities being extended, which means that the policy climate has to be conducive to private-sector agricultural investment. Other essential prerequisites for effective credit programs are tech- nical advisory services backed up by applied research and demon- stration, a well functioning transportation infrastructure, the avail- ability of inputs and a dealer network, including service facilities. On-farm development needs to be backed up by suitable storage, processing and marketing facilities, the establishment of which de- pends on the scope for increasing marketable surpluses. Where any of these essential prerequisites for effective credit pro- 'These are the so-called "credit projects" of Annexes 11 and 13. 42 grams are missing or under-developed, the Bank only finances "credit" within the context of a comprehensive integrated project. For example, "credit" forms a part, sometimes a major part, of prac- tically all lending for integrated rural development and settlement projects. "Institution building" has been an integral and extremely impor- tant aspect of most lending in support of agricultural credit programs. The Bank has actively participated in the creation of new credit insti- tutions, and in the improvement and rehabilitation of many already in existence. Of particular importance has been the adoption of de- velopment-oriented lending policies. Existing institutions frequently stress lending against adequate collateral, thus favoring those in possession of assets while not providing facilities to those who offer promising opportunities but inadequate security. In such circumstances, the Bank has stressed the need for the in- stitutions to rely more on an expert technical appraisal and financial assessment of the incremental earning capacity of a proposed private- sector investment, rather than on the ability of the sub-borrower to provide mortgages and other security. This is particularly important in lending to small farmers, especially tenant farmers and others who have no land to mortgage. The reluctance of many credit institutions to move far enough in this direction is still a major constraint in channeling Bank financing to small farmers, although there is no evidence that the average repayment record of small farmers is any worse than that of larger farmers. Similarly, the Bank has stressed the need for credit institutions to adopt terms of lending (down payments, grace periods and repay- ment periods) appropriate to the investments being financed and the financial condition of the potential sub-borrower. Interest rates are frequently subsidized by governments in order to reduce the debt- servicing burdens of the poor farmer. However, because access to institutional credit facilities is frequently limited by collateral require- ments as well as by local vested interests, it is usually the wealthier farmers who benefit from such subsidies. The poor farmer is left to the mercy of the moneylender and the exorbitant terms he imposes. For these reasons, the Bank has paid increasing attention to the adop- tion of realistic interest rate structures and other terms of lending. Livestock For the most part, Bank lending in support of livestock develop- ment has taken the form of specialized credit programs, although other forms have also been used. Such lending has risen from an annual average of $500,000 (1% of Bank Group lending for agricul- ture) in the period FY1948-63 to $91 million (22%) in FY1969-711. The rise principally reflects the good export prospects for beef. Within the overall trend, certain other trends are discernible. Although meat export remains a major objective, a growing number of projects are designed principally to meet local demand. The scope of livestock projects is being steadily widened. Most of the early projects involved existing private ranch operations in the temperate zones, almost exclusively for beef cattle and sheep. But more recent projects have moved increasingly into: tropical and semi-tropical zones2; pigs and poultry for local consumption (often to replace beef, as the price of the latter has risen with the increase in export demand); livestock operations as part of mixed farming operations, including smallholder participation"; dairy operations producing for local consumption, usually also involving dairy-beef; secondary industries such as slaughter-houses and meat-packing plants4; and state-owned ranching operations in some African coun- tries where this form of organization is preferred by governments and where privately owned ranching has not been established. Various institutional forms have been tried to bring these develop- ments to the smaller producer. The institutional forms include the establishment of group ranching based on an existing tribal structure, cooperative ranching and ranch corporations owned by smallholders. In all livestock operations financed by the Bank, the essential ele- ments have been improved management, the transfer of appropriate technology (particularly grassland improvement), training of local technical service cadres, the strengthening of local administering organizations and the application of sound economic and financial lending policies by credit institutions. Integrated Smaliholder Development Smallholder development frequently takes place within projects for regional rural development, irrigation and settlement. Sometimes 'These figures, for the most part, are included in the global figures for the credit component of Bank lending mentioned in the sub-section on "Agricultural Credit" above. 2This move into tropical and semi-tropical zones has been made practical by the development of new techniques, particularly improved animal health control, use of tropical legumes for pasture improvement and upbreeding of local cattle with improved cattle which are less susceptible to disease and insects. 3Beef-fattening and milk production have been found to be practical for smallholder operation (as opposed to breeding, which usually demands larger herds, especially under marginal range land conditions), although milk production by smallholders demands an effective extension service and marketing infrastructure. 'In view of stiffer sanitary regulations in the main importing countries, increasing amounts of beef exports are in the form of precooked meats and almost all beef exports now have to be bone-out. it is possible to assist smallholders through the provision of credit alone. But in most cases-especially in Africa-smallholder develop- ment has to beof an integrated nature. In otherwords, it has to provide for such items as technical advice and marketing facilities (including roads), as well as credit for on-farm development and input purchase. Such integrated smallholder development programs are relatively recent (dating mainly from the early 1960's); and patterns of organi- zation and management are not well established in many countries. The Bank's lending has, therefore, involved a large degree of in- novation. While some trends are beginning to emerge, not enough is known yet about smallholder motivation in different environments to make any firm judgments. It is likely that the period of innovation will continue for some time, and will be increasingly based on the applied economic research sponsored by the Bank and the work of the proposed Rural Development Unit'. Existing tenurial arrange- ments may create substantial difficulties in smallholder development. The Bank has, therefore, found it necessary at times to insist on cadastral surveys, and increased permanency in usufruct and/or ten- ure rights, before financial support is provided. Two principal strategies of integrated smallholder development have emerged. The first concentrates attention on a specific cash crop of high potential as a focal point for rural development. An example is the Kenya smallholder tea project which provided credit and exten- sion services for smallholders scattered throughout the tea-growing zones, together with complementary tea-processing factories and access roads. For many cash crops, however, the application of this strategy is limited by market prospects, even if particular projects are successful and the climate and soils are suitable over wide areas. The second strategy is to concentrate attention on the overall de- velopment of a specific geographical area which has a high potential and is of manageable size. The activities supported include produc- tion both for subsistence and for a marketable surplus, and such aspects as soil conservation, local marketing facilities for food sur- pluses, input supplies, roads and social services. Recent examples of such a strategy are the Lilongwe and Karonga projects in Malawi, the Wolamo project in Ethiopia and the Casamance project in Senegal. The principal danger in this strategy is that too many scarce re- sources, particularly of trained manpower, may be concentrated on a small area. It may become difficult to apply the strategy on a large scale later because of budgetary and manpower constraints. Thus there is a continuing attempt to determine the minimum package of inputs and services which can make a substantial impact on a 'See pages 60-61. EI large number of widely distributed smallholders farming under vary- ing conditions. Within both strategies, it is advantageous for small farmers to group together in some way to obtain economies of scale in procur- ing inputs and marketing their produce. This end has been achieved in a variety of ways in accordance with local traditions and outlook, but no definitive trend has been established. In certain cases, com- mercial or state-owned estates have formed nuclei for smallholder development, providing smallholders with essential services. In other cases, the service cooperative has been the favored solution. Where cooperatives are unpopular, possibly because of earlier failures, or where they are still in their infancy, semi-autonomous state agencies have sometimes been found to work well. One of the keys to success in all these organizational forms is the quality of the management. Land Settlement The settlement of new lands has been supported by the Bank when- ever appropriate opportunities have arisen. However, there are not many parts of the world with large areas of suitable and unexploited land. The settlement of new lands has high priority, therefore, only in some countries. The trend within settlement projects financed by the Bank has been towards reducing the "public" input and relying more on the settlers' own efforts. Otherwise, in countries where the principal initial pur- pose of settlement is to provide large numbers of families with at least a subsistence income, public budgetary constraints would se- verely restrict the number of settlers that could be accommodated. Recent settlement projects (e.g., in Colombia, Papua/New Guinea, and Tanzania) thus concentrate on the provision of low-cost access roads, technical advisory services and credit for on-farm develop- ment, and construction of drinking water sources. Estates, Seed Production and Forestry Some countries continue to seek Bank finance for private and public sector tree crop estates. A start has been made in a small way with the financing of integrated projects covering seed multiplica- tion, processing and marketing. This field of activity should become important as the benefits of modern agriculture are spread more widely. Another sub-sector which has so far received little Bank lending, but which is expected to grow in importance, is forestry (though forest industries have been extensively helped within the industrial context)'. The demand prospects for forest products are favorable. 'See Annex 12. A'; Bank involvement, however, is being restricted at present by in- appropriate forest development policies in many countries where viable forest projects have been identified. For example, the need to conserve forest resources may be interpreted as a need to restrict exploitation absolutely, rather than as a need for rational exploita- tion and resource management through selective cutting and re- planting. A further constraint is that the development periods for forest projects are usually very long. However, new varieties of fast-growing species of tropical woods, particularly pines and eucalypts, are suit- able for pulp, paper and chipboard industries and provide a promis- ing base for future projects. Agricultural Industries The demand for agriculture-related industries, such as food proc- essing and fertilizer production, is growing. In the case of smaller projects, the most appropriate way of channeling Bank Group funds to such industries is usually through intermediate industrial financ- ing institutions in the borrowers' countries. In the case of the larger projects, such as fertilizer plants, the most appropriate way may be to make industrial, rather than agricultural, loans. Recent Bank Group industrial investments of this type are summarized in Annex 12. Lending for certain other types of agriculture-related industries is more appropriately handled within the context of general agricul- tural lending. Projects of this type include private and public-sector grain storage operations, the provision of public wholesale grain markets and the expansion and improvement of grain and feed mill- ing facilities, and crop processing, especially as part of integrated schemes for production, processing and marketing. Other projects for agricultural industries which have recently been financed-and which possibly point to future trends-involve fresh fruits and vegetables. In such projects it is necessary to integrate investments in on-farm production, sorting and packing facilities and transportation (particularly refrigerated trucks and ships). The amount of such lending has so far been very small; it has averaged about $18 million per annum over the three years FY1969-71, and represents about 4% of total agricultural lending during this period. But it is tending to rise, particularly since marketable surpluses of grains in Asia and elsewhere are increasing. Fisheries Although fisheries projects so far account for a very small part of the Bank's agricultural lending-about $24 million since FY1964, 47 representing about 1% of total agricultural lending-support for the development of fisheries is growing and some trends are discernible. The earliest projects involved relatively large and sophisticated opera- tions based on tuna fish for export. But the scope for further lending of this type is limited by the very competitive nature of the industry. More recent projects have concentrated on smaller-sized opera- tions, based principally on shrimp (mainly for export) and food fish (mainly for local consumption). Expansion of Bank lending for the latter is hampered, however, by difficulties in ensuring the repayment of credits granted to small fishermen in this high-risk industry. An- other trend is towards making projects more comprehensive by including shore facilities (wharves, boat repair, ice-making and proc- essing facilities) and technical assistance, as well as the supply of boats and fishing gear. The Bank is currently working on its first project involving fresh-water fish farming, a sub-sector which may be expected to grow in importance. Training and Research The widening scope of projects is evident in the tendency to in- clude elements of training and applied research whenever necessary. Moreover, as the importance of these elements has been increasingly recognized, agricultural education, training and research have fea- tured more prominently in the Bank's education lending1. The Bank has recently made its first loan exclusively for agricultural research activities (to Spain, at the end of FY1971). Several other countries have meanwhile approached the Bank for technical advice on the formulation of national agricultural research projects. The need to strengthen the international agricultural research pro- gram has been discussed in the earlier section entitled "Challenges Facing Agriculture." To assist in meeting this need, the Bank has taken the lead in co-sponsoring, along with FAO and UNDP, the Consultative Group on International Agricultural Research in order to mobilize finance to continue and expand the work of existing international research centers and to establish new ones. The Bank provides the Secretariat of the Consultative Group. For technical advice, the Consultative Group has established a Technical Advisory Committee (TAC) consisting of 12 distinguished scientists; the Secre- tariat of the Committee is located in FAO. Within this context, the Bank is contributing to the financial re- quirements in 1972 of three centers. The first two are the already existing Centro Internacional de Mejoramiento de Maiz y Trigo 'See page 38. 48 (CIMMYT)', located in Mexico, and the Centro Internacional de Papas (CIP)2, located in Peru. The third center-the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)-is currently being established; it will be located in India. Geographical Distribution of Lending There have been some trends in the geographical distribution of agricultural lending. These are indicated in detail in Annex 11 and are summarized below: FY1948-63 FY1964-68 FY1969-71 Operations Amount Operations Amount Operations Amount Area No. % $ million % No. {M $ million % No. 1% $ million 0/ Sub-Sahara Africa 7 10 60 9 8 17 44 7 32 33 153 12 Asia and Pacific 21 30 255 41 16 35 295 49 35 37 588 48 Europe and Australia 15 21 176 29 - - - - 6 6 119 10 Middle East and North Africa 4 5 42 6 5 11 52 9 4 4 116 10 Central and South America 24 34 95 15 17 37 209 35 19 20 250 20 Totals 71 100 628 100 46 100 600 100 96 100 1,226 100 Average Amount per Operation ($ million) 8.8 13.1 12.8 Comparisons between FY1948-63 and later periods are distorted by lending in the former period to countries which no longer re- ceive Bank financing. But comparing FY1964-68 with FY1969-71, the share of agricultural lending to Sub-Sahara Africa increased from about 7% to about 12%. Agricultural lending to Southern Europe was resumed in the latter period after a total absence in the former. The share of agricultural lending to Asia, the Middle East and North Africa remained about the same in the two periods, although all these regions showed increases in absolute amounts. On the other hand, the share of lending to Central and South America declined from about 35% to about 20%. This decline is wholly attributable to the South America region, which alone shows a fall in absolute lend- ing for agriculture between FY1964-68 and FY1969-71. The fall partly 'International Center for Maize and Wheat Improvement. 2International Potato Center. A Cn reflects expanded agricultural lending by the Inter-American De- velopment Bank. The average size of agricultural loans increased substantially from $8.8 million in FY1948-63 to $13.1 million in FY1964-68, but fell mar- ginally to about $12.8 million in FY1969-71. There are substantial variations by region and sub-sector (see Annex 13). Thus, lending to Sub-Sahara Africa has averaged only $5 million per operation, while most other regions average $10-20 million. The low average for Sub- Sahara Africa reflects the small size of individual countries, the rela- tively recent introduction of the concept of agricultural investment projects and the relatively small capital investments required for the technologies being introduced. Thus the region's share (33%) of the total number of agricultural projects processed by the Bank is almost three times as high as its share of amounts committed (12%). Since Bank staff time absorbed by an agricultural lending operation is largely independent of the amount lent, Sub-Sahara Africa's large share of the number of agricultural projects reflects a similar large share of Bank staff time which, in recent years, has about equalled that devoted to Asia. This fact contrasts sharply with the relative popu- lation of the two areas (240 million and 980 million, respectively, the latter excluding Japan, Mainland China and Oceania). As another in- dication of contrast, each man-year of Bank staff time devoted to agricultural project appraisal in FY1969-71 led, on average, to $3.4 million of lending and to assisting some 10,000 farmers in Sub-Sahara Africa; whereas for Asia the figures were $12 million and 40,000 farmers, respectively. This reflects the effort the Bank is making to serve the smaller and poorer countries among its membership. Technical Assistance and Project Preparation The technical assistance provided by the Bank for agriculture has increased with the volume of lending, although there has been a conscious attempt to avoid competition with other sources of such assistance. Previously, technical assistance for agriculture was con- fined largely to broad advice on planning and organization, as part either of general survey missions or of special agricultural missions sent in response to specific requests from borrowers. As the emphasis on agriculture increased, the need to assist developing countries in the identification, preparation and implementation of projects be- came more apparent. The establishment of the FAO/IBRD Cooperative Program in 1964 and the posting of permanent missions in Eastern and Western Africa were specific responses to the need for assistance in project identi- fication and preparation. Resident missions in India, Pakistan and Colombia were also, at least in part, engaged in similar activities, and the resident mission in Indonesia has given particular emphasis to project identification and preparation work. A further trend has been to support project-connected technical assistance-project manage- ment, applied research and further project preparation-under loans or credits, in cases where UNDP has been unable to finance it. In addition, in the course of loan supervision, the Bank has offered advice not only on implementing the project in question, but also on developing new projects. In relatively few cases is the Bank the executing agency for UNDP agricultural projects (e.g., the Yaque del Norte irrigation project in the Dominican Republic, the Khanabad irrigation project and the agricultural credit project in Afghanistan, and the Indonesian Sugar Survey). Nearly always, UNDP looks to FAO as the executing agency for its agricultural projects. In some cases, the Bank has made technical assistance grants for feasibility studies (e.g., for the Jengka Triangle project development in Malaysia). On UNDP pre-investment projects which are likely to be of interest to the Bank, and for which FAO is the executing agency, there is close collaboration between the three agencies. The Bank comments on the terms of reference and plans of operations for UNDP projects, and makes its views known to FAO during the course of project execution. The proposals emerging from these pre-investment proj- ects thus tend to have a better focus on investment content and im- plementation plans when they become ready for appraisal. Activities such as these have made it possible to assist in identify- ing and preparing promising investment opportunities in sufficient numbers to meet the needs of the Bank's projected agricultural lend- ing program. But "numbers" alone are not enough; there is also the need for improvement in "quality." This is particularly so if projects are to be selected not only because they increase production, but also because they help broader development objectives such as em- ployment creation and income distribution. One difficulty is that many developing countries do not have a clear and specific set of development objectives against which a potential project's impact can be measured. The inadequacies of their planning organizations also sometimes mean that too many policy matters are left to the agencies providing the project prepara- tion assistance. Within the Bank, a principal handicap is that it needs greater in-depth knowledge of the agricultural sectors of its borrow- ers, and of the development policies affecting them, than it presently has. Partly for this reason, the Bank has recently instituted a series of in-depth agricultural sector survey missions'. 'See "Agricultural Sector Analysis," page 63. 1I The Bank has undertaken several major studies in the agricultural field. These include the Special Study of the Water and Power Re- sources of West Pakistan, the East Pakistan Agriculture and Water Development Program, a Study of African Agriculture, the Gezira Study in the Sudan, an agriculture and irrigation related manpower survey in Morocco, an agricultural credit study in Malaysia, a review of the Master Plan for Medjerda Basin Development in Tunisia and a review of the programs of Colombia's agrarian reform agency, INCORA. The Bank participated in the study of Nigerian Agricul- ture sponsored by the U.S. Agency for International Development (USAID), and later developed the blueprint for a federal agricultural credit system. The Bank has also become involved in direct assistance to its borrowers for project management. In lending for agriculture, the weakness of project management often requires the establishment of new public project authorities supported by foreign expertise to manage the project until such time as local staff can be trained to take over. In providing this assistance, great care has to be exercised so as not to stifle local initiative and to ensure that the management training is carried out effectively. To assist in meeting this problem, the Bank established the Agri- cultural Development Service (ADS) in East Africa in 1965 to provide a corps of project managers to help governments in the execution of projects. In addition, a number of project managers have been re- cruited, especially for livestock projects in Latin America, Africa and Europe-through direct hire by the borrowers, through arrangements with UNDP or the Organization of American States (in the Western Hemisphere), under bilateral aid or, as a last resort, as seconded per- sonnel from the Bank. Through its Economic Development Institute (EDI), the Bank has been helping governments to improve the capacity of their staff to undertake project analyses in the agricultural and other sectors. The EDI offers courses for qualified officials in developing countries, both in Washington and overseas. The number of courses offered each year is increasing, with a higher proportion conducted overseas and/or in languages other than English. Thus, the EDI has scheduled seven Agricultural Project Courses in FY1973, more than in any previous year. They will be conducted in English, French and Spanish, and four of them will be overseas. Increasingly, the EDI expects to focus its efforts on helping member- countries to develop their own capacity to teach project analysis. For this purpose, it expects more of its efforts to be directed towards cooperation with local and regional institutions. The preparation of teaching materials based on Bank experience for circulation outside the EDI will also assume greater importance. Since 1964, the contribution made by FAO to the Bank's activities has expanded considerably, particularly in relation to technical assist- ance. The FAO/IBRD Cooperative Program came into being in mid- 1964 with a team of 12 men in Rome. By December 1971, the number had grown to 52. The team is multi-disciplinary, consisting of agri- culturalists, irrigation specialists, economists and others working together on project problems. Most of the work under the Cooperative Program has been con- cerned with the identification and preparation of projects to be financed by the Bank. At present, FAO staff are participating in some phase of project preparation in the case of about 40% of the projects in the Bank's agricultural pipeline; they are also providing assistance in the preparation of agricultural education projects. The Bank has also called on FAO increasingly to participate in country economic missions and in the appraisal and supervision of projects. Economic Research Activities The Bank's economic research activities have two principal aims: to establish a factual or conceptual basis for policy, and to provide general support to operations in the sector. A brief account of the economic research relating to agriculture which is in progress illus- trates the current focus. A study is being made of the effects on employment and productivity of the use of tractors in India; this re- sults from the Bank's initial projects to finance mechanization there. Irrigation rehabilitation projects in Mexico have been re-appraised. Similar re-evaluations of two irrigation projects in India and a credit project in West Pakistan are in progress. The farmers' response to new varieties of rice is being investigated in Bangladesh. A study of rural development in East Africa is being prepared. It will be closely linked with the in-depth agricultural sector surveys that are being carried out in a number of countries and with reviews of a number of ongoing projects. The aim is to improve understand- ing of smallholder motivation and rural development generally, and to assist in determining the minimum package which may be needed in the future for smallholder development projects covering a wider area. Other studies which are being conducted or prepared and which are closely related to the Bank's role in rural development include: a study in India of substitution between labor and capital in road- building; a likely extension of this study to the construction of irriga- 53 tion works; studies of village electrification and water supply; an evaluation of existing experience in non-formal education for rural development; and the construction of programming models of the agricultural sectors of Mexico, Brazil and Portugal. Some of these studies simply explore the opportunities for raising the productivity of familiar types of agricultural projects-by learn- ing from the experience of past investments and by seeking to im- prove the theory and practice of measuring project costs and benefits. There is also an "entrepreneurial" aspect-in exploring opportunities for the Bank to support new types of projects. Outside of this formal part of its economic research program, the Bank is endeavoring to expand its research-oriented program of monitoring project perform- ance. So far, a systematic data-gathering and evaluation procedure has been built into only a few projects, such as the livestock projects in Mexico, Uganda and Uruguay and the Kadana irrigation scheme in India. So much agricultural economic research of relevance to develop- ing countries is undertaken throughout the world that a good deal of attention is devoted within the Bank to reviewing and interpreting studies conducted elsewhere. This is an increasingly important source of new ideas for development policy. The content of the Bank's own research program is determined by very specific policy needs or by the advantages of working with its own data. Nevertheless, much of it is conducted outside. For ex- ample, the Indian mechanization study referred to above is being carried out by two local universities, and many of the other studies are conducted with the help of consultants or cooperating agencies in the developing countries. The Bank has arranged with the FAO/ IBRD Cooperative Program for a pilot study of the employment and income distribution effects of the Muda irrigation project in Malaysia, which includes substantial farm mechanization (including private farm machinery pools providing services to small farmers). Impact of the Bank's Activities Although Bank lending has been increasing rapidly, it still repre- sents only about 1-2% of the total investment (domestic and foreign) in agriculture in developing countries'. The impact of the lending is, however, greater than might be implied by such figures. This is be- cause the capital transfer is in the form of foreign exchange provided on relatively favorable terms (particularly in the case of IDA credits). 'The basis for this estimate is given at the beginning of the next section entitled "The Bank Group's Future Role." 54 Furthermore, the Bank carefully considers the development value and appropriate design of the projects it finances. The overall impact thus extends beyond the capital transfer. For example, the Bank has helped in the introduction or extension of better technologies and in the improvement and strengthening of institutions. Projects have led to improvements not only in farming practices, but frequently also in government policies. Some ways in which the Bank's agricultural project work has had an impact far beyond the project being financed are indicated below: (a) introduction or expansion of new techniques such as in the livestock industry', bulk handling of grains, improved water control in irrigation schemes, land levelling, low cost settlement methods; and detailed consideration of the appropriate technology (hand versus mechanical methods of forest clearing; pros and cons of on- farm mechanization); (b) institution building (particularly agricultural credit institu- tions, agencies handling smallholder development projects, coopera- tives and marketing boards); the strengthening of management; and improved coordination between the various entities involved in agricultural development (particularly those responsible for construc- tion and agricultural development in irrigation projects); (c) improvements in credit policies2, and the adjustment of in- terest rates to reflect the scarcity of capital, thereby encouraging labor-intensive rather than capital-intensive methods of production, and promoting resource mobilization through increased savings; (d) greater attention to particular aspects of project preparation which are frequently neglected (the effects of irrigation on various types of soils, land classification and drainage; fisheries resource esti- mation; marketing infrastructure and institutions; crop quality con- trol standards); (e) changes in taxation and other fiscal measures such as water charges (price distortions in the economy may be preventing the development of a particular activity which would be economically advantageous or may be encouraging an activity which is not, as dis- cussed in the earlier section on "Some Major Policy Issues"); and (f) introduction of project monitoring systems, so that both bor- rowers and the Bank can learn more from their experience. 'See pages 43-44. °See page 43. To a considerable degree, the Bank's influence for change depends on the extent to which it is supported by the sort of detailed knowl- edge that comes from sector surveys and project activities. Such sector surveys are a fairly recent development, but are scheduled to be expanded greatly in the coming years'. Because the data are lacking, it is difficult to estimate the likely impact on employment and income distribution of any but the most recent Bank investments in agriculture. But these aspects are now being regularly investigated during the identification, preparation and appraisal of projects. An analysis can, however, be made of the numbers of farm families directly benefited by Bank agricultural in- vestments over the past three years (FY1969-71) and the average investment per farm. A list of the projects involved, with the perti- nent data, is presented in Annex 14. In approximately 73 of these projects, it was possible to estimate the numbers of farms that could be expected to benefit-about 2 million farms in all, involving a population of about 10 million. The average investment per farm was about $1,100, of which the Bank financed about half. However, the average investment per farm varied widely from one project to another. It ranged from a low of $24 in Niger and $50 in Burundi, to a high of about $120,000 in one case in Iran. A more detailed breakdown is given in Annex 15. Roughly 80% of all the farms involved were assisted by projects where the average investment per farm was less than the overall average of $1,100. This indicates that a large number of "small" farms benefited from such projects. On the other hand, over 80% of the total investment and over 70% of the operations involved proj- ects in which the average investment per farm was above the overall average of $1,100. Less than 1% of the participating farms benefited from projects where the average investment per farm exceeded $21,000. This category accounted for about 20°h of Bank support (in terms of both total project costs and the number of operations). The remaining 80% went to projects in which the average investment per farm was in the range of $20 to $21,000. The skewed nature of the distribution of the investments implied by these figures reflects the generally skewed distribution of farm sizes referred to in the earlier section entitled "Some Major Policy Issues." It also reflects the ability of the operators of larger farms to undertake investments efficiently, whereas the operators of small farms are usually in greater need of short-term input financing and technical assistance which is being provided under a number of bilateral programs. 'See sub-section on "Agricultural Sector Analysis" below. THE BANK GROUP'S FUTURE ROLE Agricultural investment in most developing countries will continue to receive high priority. The financing of agricultural projects can thus be a useful way of increasing transfers of capital. However, bilateral lending for agricultural projects is comparatively small. There is thus a substantial role to be played by the multilateral aid agencies such as the Bank. Over the period FY1974-78, the Bank's agricultural lending pro- gram is projected at about $4,400 million (compared with $2,400 million in FY1969-73 and $600 million in FY1964-68). This section considers the lending program in the perspective of the total invest- ment in agriculture in the developing countries and of the other agricultural project aid (bilateral and multilateral) likely to be avail- able. It also discusses what the future trends in the content of the Bank's agricultural lending program are likely to be, in the light of the discussions in the preceding sections, and the need for the Bank to improve its agricultural sector analyses. Overall Agricultural Investment and Capital Transfers No systematic attempt has been made to estimate the financial resources needed to maintain the momentum gained as a result of the "green revolution." However, some orders of magnitude can be derived on the assumption that present ratios between agricultural investment and agricultural GNP will continue to hold true. In the 30 developing countries for which the Bank has relevant data, it appears that this ratio averages about 9%'. The figure com- pares with the 7% contained in FAO's Indicative World Plan, which FAO suggests should be a sufficient ratio to sustain a growth in agri- cultural output of 4% per annum. Agricultural GNP is projected to grow at about 5% per annum in current prices, or from about $100,000-110,000 million in 1969 to about $140,000-155,000 million in 1976. This implies that overall investment in agriculture in devel- oping countries, which may have been of the order of $7,000-10,000 million in 1969, should increase to about $10,000-14,000 million by the mid-1970's. 'The figure of 9% is used in this paper for want of a better one. The results derived from it (i.e., the estimates of agricultural investment in the developing countries) should be treated with caution. Firstly, the data on investments in the 30 countries are often probably unreliable and, in any case, refer only to a single year. Secondly, the ratio for individual countries varies widely around the average 9% figure, from less than 1% to almost 50%; and only 12 of these individual country ratios lie within the range 6-12%. Thus, it is possible that widening the statistical base to more countries would substantially alter the result. 57 In comparison with the total agricultural investment in develop- ing countries, the external assistance provided by developed coun- tries and multilateral institutions such as the Bank Group for agri- cultural projects has been small. Commitments of such assistance for the period 1966 through 1970 appear to have averaged about $630 million annually. Of these, about $220 million have been in the form of bilateral assistance from DAC member countries1, $170 million from the regional development banks-the Inter-American Develop- ment Bank (IADB) and the Asian Development Bank (AsDB)-and $240 million from the World Bank2. The trend in bilateral capital assistance for agriculture has recently been downward (it has fallen from $245 million in 1966 to $160 mil- lion in 1969); but there was a considerable rise in 1970 to some $300 million. The regional development banks are generally increas- ing their commitments to agriculture. IADB's commitments reached $236 million in 1970, although they fell sharply to $93 million in 1971. AsDB's commitments started at $24 million in 1969, rose to $31 million in 1970 and then declined to $28 million in 1971. The rapid increase in the World Bank's commitments has already been noted'. In FY1971 they were about $420 million, representing about 40% of the current annual level of such commitments of about $1,000 million from all sources. Considering that, with a rapidly in- creasing rate of commitments, disbursements lag considerably behind commitments, the World Bank's contribution to overall agricultural investment in the developing countries at present is probably of the order of 1-2%. Future trends in bilateral capital assistance for agricultural projects are difficult to foresee. Although recently such assistance has been declining, there are indications that some DAC member countries intend to reverse the trend in the future, as evidenced by the recovery in 1970. If the need for food aid decreases as the "green revolution" spreads, donor countries may have to consider increasing their direct project assistance for agriculture. 'in addition, bilateral food aid is currently running at about $1,500 million per an- num; a substantial proportion of the technical assistance (also running at about $1,500 million per annum) is directed to the agricultural sector; some project aid which was for unspecified purposes at the time of commitment ($235 million in 1969) undoubtedly went to agricultural projects in practice; and aid to finance current imports ($884 million in 1969), general purpose loans ($164 million in 1969) and budget support ($260 million in 1969) can generally be expected to involve the agricultural sector to some extent. 2Agricultural investments by the European Development Fund (FED) are not included in these computations. 'in the section on "Past Bank Group Activities in Agriculture." However, one constraint may be that the foreign exchange com- ponent of agricultural projects is generally low. Another may be that it will take time for the donor countries to build the staff required to appraise agricultural projects. The future trend in agricultural lending by the regional development banks is also difficult to deter- mine, although it seems likely to be upward, perhaps reaching a level of about $350-450 million by the mid-1970's. Thus the scope for expanding the World Bank's lending in the 1970's is clear. It is proposed to raise the annual rate of such lending from the target of around $500 million for about 40 projects in FY1972 to about $750-850 million for about 60 projects by the mid- 1970's and to about $1,000 million for about 70 projects by 1978. This means that the Bank Group's share of external assistance for agricul- tural projects would probably rise to around 50% by the mid-1970's, assuming that bilateral aid can be raised to about $350-450 million per annum'. The World Bank is thus expected to remain the main external source of agricultural project financing. Its contribution to the over- all investment in agriculture in developing countries would still be relatively small, but would rise from the present 1-2% to about 4-5%. Part of the increase would result from the fact that the lag in dis- bursements is expected to be less as the rate of growth of new com- mitments slows down. Recent experience indicates that sufficient promising agricultural projects will be identified and prepared as a basis for the Bank's pro- jected agricultural lending program. The various sources of project preparation assistance are now well geared up to this task although more manpower will be needed to achieve the targets, to improve the quality of preparation and to equip the sources of preparation to handle increasingly complex and comprehensive projects, including non-agricultural rural development components. This situation, which is an improvement over the position in the early 1960's when the lack of prepared projects constrained very much the expansion of agricultural lending, has been brought about in large measure by the establishment of the FAO/IBRD Cooperative Program, the work of the Bank's overseas missions and the provisions for preparation of follow-up projects in loans and credits, which have been described earlier. Other factors contributing to this improve- ment have been the closer collaboration between the Bank and FAO on UNDP pre-investment studies in the field of agriculture, the in- 'Even if bilateral capital project assistance remains at the pre-1970 level of about $150-250 million, the Bank Group's share would remain below 60% of total official capital project assistance to agriculture. 59 creasing capabilities of local institutions and consulting firms to carry out project preparation and continuous bilateral technical assistance for project investigation. The Direction of Future Lending The direction of the Bank's future lending for agriculture will re- flect the progress made in resolving many of the fundamental ques- tions raised earlier in this paper about agriculture's role in the overall development process. For example, will the agricultural products of the developing countries be given better access to markets in the developed coun- tries? How might protein deficiencies in many traditional and low- income diets be avoided? What constitutes the essential minimum of factors necessary to stimulate broad-based rural development, taking into account the widely varying stages of development reached by different developing countries? Can appropriate intermediate technologies be devised which will make use of abundant supplies of unskilled labor without sacrificing the overall rates of growth which are fundamental to development? What are the trade-offs, if any, between economic growth and more equal distribution of incomes? In the developing countries, changes in government policies will affect agricultural investment plans, and the direction of Bank sup- port for such plans. Removal or reduction of price distortions could stimulate agricultural development and affect the form it takes. Of fundamental importance will be the attitudes of governments on agrarian reforms generally and on the priority to be given to programs to assist small farmers in particular. The Bank will continue to invest in projects supporting economically sound agrarian reform programs (including land reform, where appropriate), although it does not par- ticipate in the financing of land transfers. At a time when ideas, technologies and socio-economic priorities are changing rapidly, it is essential that the Bank maintain a flexible attitude in its agricultural lending policies and keep abreast of changes. This paper has already referred to the Bank's support for research in agricultural technologies and agricultural economics. The need for improved knowledge of the agricultural policies of borrowers is discussed below'. The Bank has recently established contacts with, and provided budgetary support for, the UN Protein Advisory Group. Within the 'See "Agricultural Sector Analysis." Bank, a Rural Development Unit is being set up; it will concern itself with all aspects of rural development with a view to identifying agri- culture's changing role in the urban-rural polarization as it occurs at different stages of development. The work of this group will have a strong operational bias, and should assist the Bank in giving greater emphasis to the small farm sector. Project Content In view of all these factors, it is difficult to predict the longer-term trends in the project content of the Bank's agricultural lending pro- gram. But in the shorter term, many of the trends that have become apparent in recent lending may be expected to continue. These include: (a) a continuing expansion in the scope of projects to cover all investments required to achieve general development objectives in the rural areas concerned; (b) consistent with governments' ability to provide adequate resources (particularly of qualified manpower), and building on the substantial experience of recent years, an extension of smallholder development schemes (including settlement of landless people) to support larger numbers of small farmers, providing them with a meaningful incremental income at the lowest possible cost per farm; (c) an increase in the number of projects aimed at producing food crops, livestock and fish for local consumption; (d) an increase in the number of projects to handle the expected increases in marketable surpluses of food products (e.g., marketing, processing, storage and transportation projects) as well as the pro- vision of facilities for input distribution and related service industries; (e) basic irrigation infrastructure lending, potentially supporting productive rural works programs, will probably continue to expand in absolute terms although its share of Bank agricultural lending is likely to decline further as increasing emphasis is placed on on-farm investments and agricultural industries; arid (f) as individual developing countries become increasingly con- scious of the need to provide employment opportunities and to dis- tribute incomes more equitably, an increasing emphasis on small farmers and projects designed to be labor-intensive. Geographical Distribution The aggregate of the agricultural components of the Bank's current country-by-country operations programs for the period FY1972-76 exceeds by far its capacity to appraise projects over that period. However, no other basis exists at present for analyzing the geographi- cal distribution pattern of future agricultural lending. The data are summarized below in relative terms, compared with two preceding periods: Projected FY1964-68 FY1969-71 FY1972-76 No. of No. of No. of Projects Amount Projects Amount Projects Amount (percentage of total) Sub-Sahara Africa 17 7 33 12 35 17 Asia and the Pacific* 35 49 37 48 29 39 Europe, Middle East and North Africa 11 9 10 20 15 19 Central and South America 37 35 20 20 21 25 100 100 100 100 100 100 *The FY1972-76 figures for Asia and the Pacific understate the probable position since they do not include lending to Bangladesh or Pakistan, for which projections are not presently available. The increased rate of lending is also matched by an increase in the number of countries in which the Bank has made, and is expected to make, agricultural investments: Number of Countries FYI 972-76 FY1964-68 FY1969-71 (Projected) Sub-Sahara Africa 5 18 33 Asia and the Pacific 7 9 13 Europe, Middle East and North Africa 3 6 16 Central and South America 12 10 21 27 43 83 The rapid expansion of agricultural lending in the recent past and that projected for the five years FY1974-78 will further tax the capacity of receiving countries successfully to implement agricultural develop- ment projects supported by the Bank. Reference has already been made to the need for the Bank and others to provide assistance for agricultural project management. The Bank's project supervision re- sponsibilities will increase even more rapidly and become more com- plex as the agricultural portfolio is further diversified both in types of projects and geographical distribution; as larger amounts are com- mitted for projects with significant policy reform, rural development and institution building components; and as training programs asso- ciated with agricultural project lending receive increased emphasis and in turn require more attention during their implementation. The effectiveness with which projects are implemented not only deter- mines in large measure the benefits the receiving country will attain, but also the rate at which funds committed will be used. Continued expansion of the Bank's lending for agriculture will thus require greatly increased efforts to meet these responsibilities. Agricultural Sector Analysis Although most of the agricultural projects at present in the pipe- line have been selected without a systematic analysis of the entire agricultural sector, the judgments of the past have probably not been far wrong. This is partly because the choices were rather obvious and were limited by the lack of well-prepared projects. But while the Bank's appraisal procedures ruled out the financing of projects with low economic returns, the projects selected may not always have ranked highest in terms of sector priorities or development goals such as employment creation. With the projected expansion of lending, it is essential that the Bank should improve its knowledge of the agricultural sector in the countries in which it operates. Greater understanding of the policy framework within which a project would have to be carried out should help also to improve its design and execution. Until quite recently, the Bank's knowledge of the agricultural sec- tor in a particular country was derived largely from two sources. The first was represented by the contacts established in the process of preparing, appraising and supervising projects (sometimes expanded to include subsector reviews, such as the Uruguayan livestock sur- vey). The second source was the agricultural sector work done in the course of economic missions. But experience has shown that although project-related sector knowledge can be valuable, its focus seldom extends beyond the subsector in which the project is located. Economic mission work, on the other hand, emphasizes the macro-economic aspects; while it is important in providing guidance on intersectoral priorities, it usually cannot provide the depth of sector analysis that is now re- quired to support agricultural lending on the projected scale. To improve its sector knowledge, the Bank has to depend increas- ingly on in-depth sector surveys, conducted by expert staff to cover all major aspects of so diverse a sector as agriculture. Such sector surveys can produce guidelines for future lending, particularly in relation to government policies which affect agriculture. They are expected to have a major impact on the selection of investment priorities (and thus also on the selection of projects for Bank financ- 63 ing). They should also facilitate appraisal work and the preinvestment work done by others. Ten such sector surveys (covering a variety of countries in Eastern and Western Africa, South Asia, East Asia, Europe and South America) have been carried out. One more is scheduled for FY1972 (making five in all for that year). The first five surveys have recently been completed through the stage where the findings have been discussed with governments. Another two reports have been prepared and are scheduled for discussion in FY1972. The remaining three reports are under preparation. The completed sector surveys have clearly helped to identify priorities and further steps in project preparation, and have been well received by governments. In one case, the survey report has led to a reorientation of the government's irrigation and rural development strategy for an im- portant but depressed region, and to a higher priority for storage and transportation. In another case, the survey has resulted in an exten- sive revision of irrigation policies and a renewed interest in reorgani- zation of government services to agriculture. A third country has begun serious attempts to reorganize its agriculture along lines rec- ommended in the sector survey report, and to take steps toward land tax reform and more economical use of available lands. In the other two cases where the results of the survey have been discussed with governments, a positive contribution has been made to the countries' preparations for forthcoming development plans. In all these cases, there have been continuing discussions between the governments and the Bank on major issues, as well as further project preparation activities, based on the results of the sector sur- veys. Of course, not all recommendations made by sector survey mis- sions are accepted by governments. The results of sector survey missions have so far been encouraging. It is proposed to expand coverage to other developing countries as rapidly as possible, taking into account the competing demands on the Bank's staff resources. It is planned that, by FY1975, sector survey missions will be mounted at the rate of ten per annum. At that rate, sector surveys would be completed for the prospective 25 largest agricultural borrowers by FY1975, and for the prospective 35 next largest by FY1980. rC4 Indices of Agricultural Production by Level of Development and by Region Annual Growth Rate TOTAL PRODUCTION 1948-52 1953-57 1958/60- ALL AGRICULTURAL Average Average 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970(1) 1968/70 PRODUCTS (1952-56 average = 100) (Percent) Developed Countries(2) 89 101 106 114 116 114 114 119 124 124 125 128 133 137 136 136 1.75 Developing Countries 86 103 109 114 118 121 125 129 132 135 138 138 144 148 154 158 2.60 Latin America 87 103 111 117 118 120 128 128 132 134 143 140 148 147 153 158 2.58 Far East(3) 87 103 107 111 117 121 127 128 132 136 134 135 141 148 155 160 2.86 Near East(4) 82 104 115 118 122 124 123 135 137 141 145 148 153 159 163 163 2.91 Africa(5' 85 102 107 110 113 121 116 124 128 132 134 134 138 142 147 148 2.42 FOOD PRODUCTS ONLY Developed Countries(2M 89 101 108 115 117 115 115 120 125 125 127 132 138 142 141 141 2.02 Developing Countries 86 103 109 113 116 120 124 127 131 135 136 137 143 149 154 159 2.84 Latin America 87 103 111 116 114 117 124 126 131 137 141 142 151 152 157 166 3.20 Far East(3' 87 103 107 112 118 123 127 129 132 137 134 135 141 149 156 162 2.84 Near East(4) 82 104 115 118 121 122 123 133 135 137 141 145 150 155 159 159 2.73 Africa(5) 86 102 106 108 111 117 113 121 125 127 128 128 132 137 140 141 2.20 PER CAPITA PRODUCTION ALL AGRICULTURAL PRODUCTS Developed Countries(2) 93 100 102 108 108 106 105 108 111 110 109 111 114 117 115 113 0.69 Developing Countries 94 101 102 103 104 105 106 106 106 106 105 103 104 105 106 106 0.16 Latin America 97 100 103 105 102 102 105 103 103 101 105 100 103 99 100 101 -0.30 Far East(3a 95 101 101 102 104 106 108 107 107 107 103 102 103 106 108 109 0.35 Near East(4) 91 102 107 106 107 106 103 110 109 109 109 108 109 110 110 107 0.16 Africa(5) 94 100 100 100 101 105 98 103 104 104 103 100 101 101 102 100 -0.09 FOOD PRODUCTS ONLY Developed Countries(2 93 100 103 109 109 107 105 109 112 111 111 114 119 121 118 117 0.92 Developing Countries 94 101 102 103 103 104 105 105 105 106 104 102 104 105 105 106 0.20 Latin America 97 101 103 104 100 YY 102 101 102 103 104 102 105 102 103 105 0.23 Far East(3) 94 101 101 103 106 107 109 107 107 108 104 102 103 107 108 110 0.28 Near East(4) 90 102 107 107 106 105 103 108 107 106 106 106 107 108 107 104 0.03 > Africa)5u 94 100 99 98 99 102 96 100 101 100 99 96 97 97 97 95 -0.35 = 'l Preliminary estimates .2) Includes Japan, S. Africa and Israel and excludes U.S.S.R. and Eastern Europe (3) Excludes Japan (4) Excludes Israel (5) Excludes S. Africa x Source: The State effoodandAgricuIturu, 1971, (FAO, Rome: 1971) Annex 2 Normal Variations in Economic Structure with Level of Development Level of GNP per capita (in 1964 USS) Structural Characteristic $50 $100 $200 $300 $400 Percent Share of GDP of Primary production 58 46 36 30 27 Industry 7 13 19 23 25 Services 30 35 38 39 40 Utilities 5 6 7 8 8 Percentage of labor force in Primary production 75 68 58 50 44 Industry 4 10 16 21 24 Services and utilities 21 22 26 29 32 Urban population, as % of total population 7 20 34 41 46 Exports of primary products, as % of total exports 89 78 68 61 56 Source: Hollis B. Chenery, "Growth and Structural Change", Finance and Development, September 1971. 66 Agricultural Output per Agricultural Worker and Factors Associated with Differences in Output, 19 Countries, 1960 Rank of Country in Urban Miles of Agricultural Agricultural Total Land Arable I nfant Agricultural Fertilizer Population as Road per Output per Gross Output per Capita Land per Mortality Workers per Used per a Percentage 1,000 Hectare of Domestic per Farm of Total Agricultural Literacy Rate Hectare of Hectare of of Total Sq. Miles of Arable Product Country Worker Population Worker Rate per 1,000 Arable Land Arable Land Population Land Area Land per Capita Group I Dollars Hectares Hectares Percent Number Number Kilograms Percent Rank Dollars Dollars Israel 1,825 0.9 3.3 96 32.0 0.31 80.5 77.3 3 557 905 Argentina 1,080 12.5 13.1 86 59.6 .07 n.a. 67.0 16 78 465 Spain 656 1.6 4.4 87 51.6 .23 31.6 n.a. 7 150 372 Poland 616 1.0 2.4 95 74.7 .41 49.0 48.1 2 252 538 Chile 547 9.1 9.3 80 118.0 .11 17.0 67.2 12 59 405 Colombia 531 7.7 1.9 62 100.0 .51 n.a. n.a. 18 270 248 Venezuela 500 12.5 3.2 52 64.1 .30 3.8 66.1 17 150 650 n Japan 402 0.4 0.4 98 37.7 2.39 303.7 63.5 1 961 337 Greece 391 1.6 1.9 80 41.4 .52 38.0 42.5 5 205 297 Mexico 369 5.6 4.1 65 77.7 .30 9.4 50.7 11 110 321 Average 692 5.3 4.5 79 65.7 0.52 66.6 60.3 9 279 454 Group 11 Egypt 365 3.7 0.6 20 130.1 1.76 87.0 37.7 15 643 155 Turkey 326 2.7 2.6 39 n.a. .39 1.5 37.8 13 127 254 Yugoslavia 250 1.4 1.8 77 98.5 .57 28.0 n.a. 4 141 179 Brazil 229 11.1 1.4 49 n.a. .45 13.0 45.1 14 104 145 Taiwan 228 0.3 0.6 54 34.2 2.10 203.8 59.5 6 477 97 Pakistan 182 1.0 1.5 19 n.a. .73 3.2 n.a. 10 133 64 Philippines 181 1.0 1.2 75 82.6 .77 12.5 42.7 9 139 113 India 114 0.7 1.2 24 145.9 .80 2.3 17.9 8 91 70 Thailand 94 1.9 0.9 68 54.8 1.13 2.3 11.8 19 106 84 D Average 222 2.6 1.3 47 91.0 0.97 39.3 36.1 11 218 129 p S Source: Changes in Agricult ore in .26 Developing Nat fions, 1S48-1983(U.S. Department of Agriculture, Foreign Agricultural Economic Report No. 27, 1965). W Annex 4 Compound Annual Rates of Growth in Total Agricultural Production, Total Agricultural Population, Agricultural Production per Farm Person, and Crop Area per Farm Person, 25 Countries, Selected Periods, 1950-68 Total Total Agricultural Crop Country and Agricultural Agricultural Production Area per Region Production Population per Farm Farm 1950-68(1) 1950-65(2) Person Person Percent Latin America Costa Rica 4.2 2.3 1.9 2.1 Guatemala 5.0 3.4 1.6 - .2 Jamaica 2.7 1.3 1.4 1.6 Mexico 5.1 1.7 3.4 1.4 Panama 3.6 2.1 1.5 .4 Argentina 2.0 1.9 .1 - .8 Brazil 3.8 1.3 2.5 3.1 Colombia 3.3 1.3 2.0 .5 Paraguay 2.4 1.8 .6 n.a. Uruguay -.1 .8 -.9 -2.8 Venezuela 5.1 2.1 3.0 1.8 Asia and Africa Cyprus 4.9 - .8 5.7 n.a. Iran 2.7 1.2 1.5 n.a. Israel 9.3 2.2 7.1 n.a. Turkey 3.4 2.8 .6 - .6 Ceylon 2.9 1.2 1.7 .3 Indonesia 2.3 1.5 .8 .8 Malaysia 4.1 3.1 1.0 -1.5 Thailand 4.5 2.9 1.6 .6 Egypt 2.4 2.2 .2 -1.2 India 2.6 2.1 .5 - .8 Japan 3.3 -4.5 7.8 3.7 South Korea 3.7 .3 3.4 1.2 Philippines 3.7 2.0 1.7 .7 Taiwan 4.5 2.4 2.1 -1.7 (1) Data cover a shorter period than 1950-68 in some instances. (2) Estimates based on data from several sources. Data on economically active population in agriculture were used for Ceylon, Costa Rica, Egypt, Indonesia, Malaysia and Thailand. Source: Economic Progress of Agriculture in Oeveloping Nations, 1950-1968 (U.S. Department of Agriculture, Economic Research Review, 1970). cQ0 Annex 5 Actual and Projected Capacity, Production and Consumption of Fertilizers(') in Developed and Developing Countries 1967,68 1972 Gross Potential Projected Capacity Production Consumptior Capacitys Producti,o)3) Consumption (t) (2) (31 (4) (5) (6) Million metric tons Latin America 1.6 0,8 2.1 4.0 2.5 3.5-4.0 West, Central and East Atrica - 0.3 0.7 0.6 0.5-0.6 Near East and North Africa 1.7 0.6 1.0 3.5 2.4 1.8- 2.0 Far East (excluding Japan) 1.9 1.2 3.5 5.3 3.3 7.0-8.0 ot which: India (1.1) (0.6) (1.8) (3.3) (2.0) (4.0-4.5) Pakistan (0.2) (0.1) (0.3) (0.7) (0.5) (0.6-0.8) Total developing countries 5.2 2.6 6.9 13.5 8.8 13-15 Total developed countries(4i 63.0 53.8 46.1 108.4 78.6 65 World Total 68.2 56.4 53.0 121.9 87.4 78-80 'i In termso nutrient content (N, PrO, and KeO). Wi Gross capaory includes plants in operation and plants and luctories under constrootion. contracted, or in an advanced stage of planmnig ir 1967, 68, Ior the U.5.5.R , production goals have seen used. 11 Potential production has been derived fram gross capacity by taking into account closure ot old plants, reduced capacity utilioaboo during the first pear of operation of new plants, losses in manutacturing and non.agricultural use of primary plant nutrients. Including Eastern Europe and the U.S S.R. source: The State afel,d eodAgitevtaae, 1P70, (FAO, Rnme 1970). Annex 6 Area Devoted to High-Yielding Varieties (HYV) in Relation to Area of All Varieties, 1970/71 (Prelinsinary; subject to revision) Area-0) Proportion HYV All Varieties HYV -acres- -percent- WHEAT Atfghaistan 574,300 (7,329,500)(1) (7.8) Algeria 346,300 7,400,000(2) 4.7 India 14,559,000 44,211,100 32.9 Iran 321,200 10,378,000 3.1 Iraq 309,000 5,023,500 6.2 Jordan 300 543,600 0.1 Lebanon 8,600 150,700 5.7 Morocco 49,400 1,230,600 4.0 Nepal 242,700 (959,000)(3) (25.3) Pakistan (E) 24,000 311,000 7.7 Pakistan (W) 7,288,000 14,976,000 48.7 Syria 94,000 930,000 10.1 Tunisia 255,000 1,816,200 14.0 Turkey 1,184,000 20,262,200 5.8 RICE Burma 496,300 12,295,000 4.0 Ceylon 73,000 (1,609,000)(3) (4.5) India 13,593,000 92,494,500 14.7 Indonesia 2,303,400 20,353,600 11.3 Laos 132,500 (1,900,000)(i) (7.0) Malaysia (W) 327,100 1,336,000 24.5 Nepal 167,600 (2,900,000)(3) (5.8) Pakistan (E) 1,137,000 24,494,000 3.3 Pakistan (W) 1,548,000 3,715,000 41.7 Philippines 3,868,100 7,691,300 50.3 Thailand (400,000)(2) 18,779,600 (2.1) Vietnam (S) 1,240,300 6,423,000 19.3 (X) Escluding Communist Asia. I Particularly rough eshmate. 1969 70 area. source maporas end Pt,oriags eo High- Yielding Ventieo of Wheat od tke in the dean Deeloped NOelen. By Dana Dalrymple (Foreign Economic Development Service, U.S. Depaitment oa Agriculture in Cooperation Mith U.0 Agency tor I steroatioral Development, Washinrton., D.C.) 69 Medium-Term (1974) Forecast for Cereal Production, Utilization and Net Trade in Selected Developing Countries Production Apparent Domestic Utilization Net Trade Actual Forecast Actual Forecast Actual Forecast 1965-67 1965-67 1965-67 Average 1970(l) 1974 Average 1970(') 1974 Average 1970(1) 1974 NET IMPORTERS Thousand metric tons Far East Ceylon 680 1,064 1,364 1,634 2,054 2,034 + 954 + 990 + 670 China (Taiwan) 2,244 2,295 2,410 2,456 3,529 3,696 + 212 + 1,234 + 1,286 India 68,627 91,309 109,400 77,582 94,454 110,000 + 8,955 + 3,145 + 600 Korea, Republic of 5,937 6,788 7,624 6,598 8,764 9,964 + 661 + 1,976 + 2,340 Malaysia (West) 881 933 1,320 1,532 1,655 1,684 + 651 + 722 + 364 Pakistan 17,557 22,814 26,072 19,025 24,500 26,530 + 1,468 + 1,686 + 458 Total Countries Listed 95,926 125,203 148,190 108,827 134,956 153,908 +12,901 + 9,753 + 5,718 Near East Iran 5,168 6,358 7,245 5,426 6,707 7,385 + 258 + 349 + 140 Iraq 1,892 2,152 2,940 1,926 2,265 3,190 + 34 + 113 + 250 Egypt 5,782 6,642 8,210 7,908 7,293 9,610 + 2,126 + 651 + 1,400 Total Countries Listed 12,842 15,152 18,395 15,260 16,265 20,185 + 2,418 + 1,113 + 1,790 [3 Latin America Brazil 17,355 21,993 23,855 18,958 22,307 23,250 + 1,603 + 314 - 605 Chile 1,789 1,771 2,357 2,195 2,251 2,750 + 406 + 480 + 393 Total Countries Listed 19,144 23,764 26,212 21,153 24,558 26,000 + 2,009 + 794 - 212 Africa Algeria 1,383 1,598 2,172 1,920 2,057 2,735 + 537 + 459 + 563 Ghana 527 487 573 638 624 689 + 111 + 137 + 116 Morocco 2,779 4,301 4,519 3,354 4,287 4,902 + 575 - 14 + 383 Senegal 663 672 952 898 951 1,342 + 235 + 279 + 390 Tunisia 524 655 515 786 1,228 1,202 + 262 + 573 + 687 Total Countries Listed 5,876 7,713 8,731 7,596 9,147 10,870 + 1,720 + 1,434 + 2,139 Grand Total 133,788 171,832 201,528 152,836 184,926 210,963 +19,048 +13,094 + 9,435 NET EXPORTERS Argentina 16,781 19,166 25,750 7,407 9,287 13,285 - 9,374 - 9,879 -12,465 Mexico 12,276 14,150 17,780 10,727 13,366 16,480 - 1,549 - 784 - 1,300 Thailand 8,713 10,368 12,215 6,020 7,828 8,605 - 2,693 - 2,540 - 3,610 Total Countries Listed 37,770 43,684 55,745 24,154 30,481 38,370 -13,616 -13,203 -17,375 n NOTE: Net imports are indicated by a (+) sign; net exports by a (-) sign. S (1) Preliminary estimates Source: The State of Food and Agriculture, 1971, (FAO, Rome: 1971) Annex 8 Share of Agricultural Primary Products in Export Earnings of Selected Developing Countries, Average 1967-69 Total Share of Country Export Agricultural Earnings Products(') (US5 million) (percent) 1. Over 80% Ethiopia 106.7 99.1 Ceylon 327.6 98.8 Turkey 518.0 90.7 Uganda 186.7 90.4 Argentina 1,481.5 87.6 Ghana 295.4 83.9 Brazil 1,948.8 82.4 Morocco 453.1 81.7 I. 50-80% Thailand 680.7 78.3 Colombia 525.3 77.8 Tanzania 223.6 76.6 Kenya 162.2 75.3 Philippines 890.7 73.3 Mexico 1,277.6 66.0 Malaysia 1,111.9 60.0 Nigeria 713.4 58.8 Tunisia 157.6 53.4 Ill. 30-50% Pakistan 682.2 47.8 Singapore 1,320.1 45.6 Peru 833.8 45.0 Spain 1,621.8 41.7 India 1,758.0 38.0 China (Taiwan) 826.8 33.8 Indonesia 717.9 31.2 Yugoslavia 1,330.1 30.1 (1) SITC 0, 1, 2 (-28), 4. Source: Yearbook ofInternational Trade Statistics. 1969 (United Nations>. 71 Indices of Value of Agricultural Exports and Imports, by Level of Development 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970'( (1957-59 Average=100 Exports Developed countries(2) 84 95 106 94 100 110 118 119 136 153 151 159 158 156 164 191 Developing countries 102 102 102 98 100 102 101 102 111 117 118 116 111 114 119 130 Latin America 104 106 105 99 96 99 100 103 112 121 125 123 117 127 128 143 Far East(3) 108 101 101 93 106 109 103 102 112 111 109 106 101 100 103 108 Near East(4 95 101 110 91 99 102 96 101 113 114 123 127 123 129 135 138 Africa(') 94 95 96 104 100 100 100 99 107 119 113 113 108 116 117 136 Imports Developed countries(2) 95 100 99 97 104 107 110 116 118 119 124 130 129 134 137 142 Developing countries 83 91 100 98 102 116 120 122 129 138 144 154 162 163 154 - Latin America 101 91 99 102 99 103 108 115 125 140 136 147 144 153 145 - Far East(3) 74 90 103 98 98 121 117 117 133 143 145 158 173 169 165 185 Near East4) 77 87 95 94 111 124 137 138 141 149 169 171 174 173 155 - Africa(5) 87 97 100 95 105 119 131 128 112 116 131 136 151 151 140 - (') Preliminary estimates 12) Includes Japan, Israel and South Africa and excludes U.S.S.R. and Eastern Europe (3) Excludes Japan and Mainland China (4) Excludes Israel (s) Excludes South Africa Source: The State of Food and Agricilt/ure, 1971. (FAO, Rome: 1971) rD Annex 10 Composition of Agricultural Export Earnings of Developing Countries, 1955-70 Competing Products(') Non- Basic Raw Sub- Competing Foods Materials Total Products(2) Total (percent) 1955 36.5 31.2 67.7 32.3 100.0 1960-62 (average) 43.0 29.5 72.5 27.5 100.0 1965-67 (average) 47.1 26.2 73.3 26.7 100.0 1970(3) 50.5 23.5 74.0 26.0 100.0 (X) Agricultural commodities produced in direct competition by both developed and developing countries. In the case of agricultural raw materials (e.g., cotton, rubber, hides and skins), the competition taken into account includes that between natural products and synthetic substitutes also. (2) Coffee, cocoa, tea, bananas, spices, etc. (3) Estimates. Source: The State oi Feud andAgriculture, 1970. p. 178, (FAO, Rome: 1970: 7 Bank and IDA Lending for Agriculture by Area and Subsector for Various Periods (Amounts in US$ million) Irrigation, Storage, Drainage, Processing, Flood Control Livestock Credit Marketing Number Amount Number Amount Number Amount Number Amount FY1948-63 Eastern Africa 3 48.0 - - - - - - Western Africa - - - - - - - _ East Asia & Pacific 7 55.4 South Asia 11 182.5 Europe & Australia - - _ _ _ _ _ _ Middle East & North Africa 3 38.0 - - - - 1 3.9 Central America & Caribbean 3 27.5 - - - - 2 .8 South America 1 18.0 1 7.0 3 15.0 - - Totals 28 369.4 1 7.0 3 15.0 3 4.7 % of totals 39.4 58.8 1.4 1.1 4.2 2.4 4.2 0.8 FY1964-68 Eastern Africa - - - - 2 8.6 - - Western Africa - - - - - - - - East Asia & Pacific 4 103.0 - - 1 5.0 - - South Asia 5 103.0 - - 2 37.0 1 19.2 Europe & Australia - - - - - - - Middle East & North Africa - - - - 5 51.5 - - Central America & Caribbean 2 44.0 - - 2 28.0 - - South America 2 15.5 9 101.8 1 15.0 - - Totals 13 265.5 9 101.8 13 145.1 1 19.2 % of totals 28.3 44.2 19.6 16.9 28.3 24.1 2.1 3.2 FY1969-71 Eastern Africa 1 5.0 5 13.2 - - - - Western Africa - - - - 2 10.1 - - East Asia & Pacific 5 117.0 1 7.0 1 12.5 2 21.8 South Asia 8 143.2 - - 8 188.4 1 6.0 Europe & Australia 2 52.0 3 42.2 - - 1 25.0 Middle East & North Africa 3 96.0 - - 1 20.0 - - Central America & Caribbean 1 5.0 6 153.8 1 3.7 - - South America - - 8 56.9 1 17.0 - - Totals 20 418.2 23 273.1 14 251.7 4 52.8 % of totals 20.8 34.1 24.0 22.3 14.6 20.5 4.2 4.3 FY1948-71 Eastern Africa 4 53.0 5 13.2 2 8.6 - - Western Africa - - - - 2 10.1 - - East Asia & Pacific 16 275.4 1 7.0 2 17.5 2 21.8 South Asia 24 428.7 - - 10 225.4 2 25.2 Europe & Australia 2 52.0 3 42.2 -- - 1 25.0 Middle East & North Africa 6 134.0 - - 6 71.5 1 3.9 Central America & Caribbean 6 76.5 6 153.8 3 31.7 2 .8 South America 3 33.5 18 165.7 5 47.0 - - Totals 61 1,053.1 33 381.9 30 411.8 8 76.7 % of totals 28.6 42.9 15.5 15.6 14.1 16.8 3.8 3.1 7A Annex 11 General Agriculture Forestry Fishery Total Number Amount Number Amount Number Amount Number Amount % of total 4 12.2 - - - - 7 60.2 9.4 - - - - - - 7 55.4 8.7 3 17.1 - - - - 14 199.6 32.2 12 170.2 3 5.8 - - 15 176.0 28.6 - - - - _ _ 4 41.9 6.5 6 9.0 - - - - 11 37.3 5.8 8 17.8 - - - - 13 57.8 8.9 33 226.3 3 5.8 - - 71 628.2 100 46.5 36.0 4.2 0.9 - - 100 100 5 18.0 - - - - 7 26.6 4.4 1 18.0 - - - - 1 18.0 3.0 1 14.0 - - 2 13.7 8 135.7 22.6 - - - - - - 8 159.2 26.5 - - - - _ _ 5 51.5 8.6 1 5.0 - - - - 5 77.0 12.8 - - - _ - - 12 132.3 22.0 8 55.0 - - 2 13.7 46 600.3 100 17.4 9.2 - - 9.2 2.3 100 100 9 44.4 2 7.9 - - 17 70.5 5.7 12 71.1 - - 1 1.3 15 82.5 6.7 6 67.5 1 8.5 1 3.5 17 237.8 19.5 1 13.0 - - - - 18 350.6 28.6 - - - - - - 6 119.2 9.7 - - - - - - 4 116.0 9.5 _ - - - - - 8 162.5 13.3 1 8.1 - - 1 5.3 11 87.3 7.1 29 204.1 3 16.4 3 10.1 96 1,226.4 100 30.2 16.6 3.1 1.3 3.1 0.8 100 100 18 74.6 2 7.9 - - 31 157.3 6.4 13 89.1 - - 1 1.3 16 100.5 4.1 7 81.5 1 8.5 3 17.2 32 428.9 17.5 4 30.1 - - - - 40 709.4 28.9 12 170.2 3 5.8 - - 21 295.2 12.0 - - - - - - 13 209.4 8.5 7 14.0 - - - - 24 276.8 11.3 9 25.9 - - 1 5.3 36 277.4 11.3 70 485.4 6 22.2 5 23.8 213 2,454.9 100 32.9 19.8 2.8 0.9 2.3 1.0 100 100 71i Annex 12 Bank Group Investments in Agriculture-Related Industries(') Direct Investments By Investments Channelled Through Lotal industrial Bank/IDA IFC Development Banks(2) (USS million) Subsector Agricultural Production - - 5 Food Processing and Beverages 34 25 34 Textile Manufacturing 13 19 113 Tobacco Processing - - 7 Leather and Leather Goods - - 2 Wood-based Industries 173 60 32 Rubber Products - - 11 Sub-total 220 104 204 Fertilizer, Pesticides and other Chemicals 117 53 6 Farm Machinery 2 - 2 Miscellaneous Products for Agricultural Use - - 1 Sub-total 119 53 9 Grand Total 339 157 213 (1) Those industries based on locally-produced agricultural raw materials or supplying local agriculture; through FY1971. (2) FY1968-71 only. 7 C Annex 13 Average Size of Agricultural Project by Area and Subsector over Various Periods (US$ million) FY1948-63 FY1964-68 FY1969-71 FY1948-71 A. By Subsector Irrigation, Drainage and Flood Control 13.2 20.4 20.9 17.3 Livestock 7.0 11.0 11.9 11.6 Credit 5.0 11.2 18.0 13.7 Storage, Processing and Marketing 1.5 19.0 13.2 9.6 General Agriculture 6.8 7.0 7.0 6.9 Forestry 2.0 - 5.5 3.7 Fisheries - 6.8 3.4 4.8 B. By Area Eastern Africa 8.6 3.8 4.1 5.1 Western Africa -- 18.0(1) 5.5 6.3 East Asia and Pacific 7.9 17.0 14.0 13.4 South Asia 14.3 19.9 19.5 17.7 Europe and Australia 11.7 - 19.9 14.1 Middle East and North Africa 10.5 10.3 29.0 16.1 Central America and Caribbean 3.4 15.4 20.3 11.5 South America 4.4 11.0 7.9 7.7 C. Overall Average 8.8 13.1 12.8 11.5 C') A single project. Annex 14 Loans and Credits for Agriculture, FY1969-71(') Page 1 Total Cost and Number of Participating Farms Loan or Credit Total Number of Project Participating Number Amount(2) Cost Farms(') (S million) (S million) FY1969 Ecuador Fisheries 555 5.3 6.6 - Sudan Mechanized Farming 556 5.0 8.0 140 Guyana Sea Walls 559 5.0 7.5 - Zambia Forestry 562 5.3 11.1 - Senegal Agriculture Credit 584/140 6.0 24.1 90,000 Malagasy Livestock 585 2.8 4.2 - Turkey Seyhan Irrigation 587/143 24.0 63.0 4,800 Cameroon Oil Palm 593 7.9 14.1 900 Iran Dez Irrigation 594 30.0 74.3 3,700 Korea Irrigation 600 45.0 89.9 50,000 Philippines Agriculture Credit 607 12.5 25.0 6,500 Mexico Livestock and Credit 610 65.0 200.0 4,600 Ivory Coast Oil Palm 611/12/13 17.1 29.1 5,500 India Seeds 614 13.0 22.4 400 Paraguay Livestock 620/156 8.6 15.5 450 Colombia Agriculture Credit 624 17.0 42.5 2,500 Zambia Livestock 627 2.5 5.8 - Indonesia Irrigation Rehabilitation 127 5.0 8.8 110,000 Kenya Livestock 129 3.6 11.4 - Uganda Livestock 130 3.0 5.1 56 Tanzania Livestock 132 1.3 2.0 - Pakistan Consultants 136 2.0 6.6 Papua New Guinea Agriculture Development 137 1.5 3.3 580 Dahomey Oil Palm 144 4.6 9.6 4,000 Burundi Coffee Smallholders 147 1.8 2.1 44,000 Indonesia Estates 1 155 16.0 32.0 - Pakistan ADB III 157 30.0 47.7 18,000 FY1970 Spain Livestock 633 25.0 52.0 1,050 Philippines Irrigation 637 34.0 67.5 17,000 Kenya Forestry 641 2.6 4.0 Morocco Sebou Irrigation 643 46.0 109.5 15,000 Colombia Livestock 651 18.3 44.1 970 Ceylon Mahaweli Irrigation 653/174 18.7 50.0 40,000 Iran ADFI Credit 662 6.5 16.0 131 Malaysia Jengka ll 672 13.0 28.6 3,000 Malaysia Forestry 673 8.5 12.4 1,200 Zambia Commercial Farming 685 5.5 11.1 300 Ivory Coast Cocoa 686 7.5 13.6 22,400 Uruguay Livestock 698 6.3 13.1 1,000 Ghana Fisheries 163 1.3 2.3 - Ceylon Drainage 168 2.5 4.1 10,000 Ethiopia Wolamo Agriculture Dev. 169 3.5 5.1 7,050 Bolivia Livestock 171 1.4 2.3 110 Ecuador Livestock 173 1.5 2.5 60 Papua New Guinea Agriculture Development 175 5.0 8.8 1,020 India Irrigation 176 35.0 66.7 176,000 Honduras Livestock 179 2.6 5.2 135 Egypt Drainage 181 26.0 147.0 250,000 Pakistan I rrigation 184 13.0 22.7 60,000 Ethiopia Agriculture Development 188 3.1 4.4 - India Gujarat Credit 191 35.0 67.0 26,840 -7 0 Annex 14 Page 2 Loan or Credit Total Number of Project Participating Number Amount() Cost Farms(') (S million) (S million) Indonesia Estates 11 194 17.0 31.7 - Indonesia Irrigation Rehab. 11 195 18.5 37.0 372,000 Afghanistan Agriculture Credit 202 5.0 7.2 7,000 India Punjab Credit 203 27.5 40.0 8,000 Ghana Cocoa 205 8.5 15.6 5,000 Niger Agriculture Credit 207 0.6 0.9 25,000 Pakistan Tubewells 208 14.0 45.0 45,000 Pakistan Irrig. Engineering S-8 0.8 1.4 - FY1971 Israel Agriculture Credit 709 20.0 49.0 7,000 Jamaica Agriculture Credit 719 3.7 8.0 140 Philippines Rice Processing 720 14.3 21.0 - Guatemala Livestock 722 4.0 7.8 300 Colombia Caqueta Land Settlement 739 8.1 21.6 6,300 Mexico Livestock 111 747 75.0 192.5 10,900 Greece Groundwater Development 754 25.0 50.0 6,000 Ivory Coast Palm Oil and Coconuts 759/760 7.0 17.6 2,300 Turkey Fruit and Vegetables 762/257 25.0 44.0 4,500 Nigeria Western State Cocoa 764 7.2 11.6 8,000 Spain Personnel and Training 768 12.7 28.2 - Uruguay Livestock III Supplement 773 4.0 21.4 3,000 Indonesia Fisheries 211 3.5 4.3 - Uganda Tobacco 212 4.0 7.3 6,000 Malagasy Alaotra Irrigation 214 5.0 8.2 2,900 Tanzania Tobacco 217 9.0 14.7 15,000 Indonesia Irrigation Rehab. III 220 14.5 29.1 260,000 Guyana Livestock 221 2.2 4.4 - Ecuador Livestock III 222 10.0 19.7 575 Upper Volta Cotton 225 6.2 9.4 46,000 India Andhra Pradesh Credit 226 24.4 45.0 40,000 India Agro-aviation 230 6.0 8.8 - Korea Livestock 234 7.0 12.6 700 Turkey Livestock 236 4.5 7.5 300 Mauritius Tea Development 239 5.2 7.0 3,730 Malawi Lilongwe 11 244 7.3 8.6 52,000 Dominican Republic Livestock 245 5.0 9.0 260 Afghanistan Khanabad Irrigation 248 5.0 6.0 6,000 India Haryana Credit 249 25.0 44.5 20,000 India Tamil Nadu Credit 250 35.0 62.3 50,000 Senegal Casamance Rice 252 3.7 4.8 5,000 Senegal Terres Neuves 254 1.4 1.6 550 Indonesia Tea 259 15.0 24.9 - Bolivia Livestock 261 6.8 11.0 3,000 Indonesia Seeds 246 7.5 12.0 - ') Excludes a multipurpose protect in Yugoslavia and a supplemental loan for the Tarbela Dam, Pakistan. (2) Net of cancellations. (O) Where no number is given, the concept of "participating farms" is not applicable to the project. Number of projects: 94 Total cost: $2,500 million Bank and IDA financing: $1,200 million Number of participating farmers: 2 million (in 73 projects costing $2,300 million, involving $1,100 million of Bank and IDA financing). 7Q Annex 15 Distribution of Project Investment and Numbers of Participating Farms by Size of Project Investment per Farm' (For those 73 projects listed in Annex 14 for which the concept of "participating farms" is applicable.) A. Distribution of Project Investment Amount of project investment in approximate 20% groupings, ranked by project investment per farm. PROJECT INVESTMENTS PER FARM NUMBER OF NUMBER AVERAGE RANGE PARTICIPATING OF ($ million) $ $ FARMS PROJECTS 436.2 43,721 20,910-122,140 9,977 17 468.5 15,513 9,780- 20,080 30,200 9 482.4 5,729 3,430- 9,530 84,200 14 467.3 1,699 1,120- 3,120 275,020 14 456.6 285 20- 1,030 1,602,450 19 (All projects) 2,311.0 1,154 20-122,140 2,001,847 73 B. Distribution of Numbers of Participating Farms Numbers of participating AMOUNT farms in approximate 20% PROJECT INVESTMENTS PER FARM OF PROJECT NUMBER groupings, ranked by project AVERAGE RANGE INVESTMENT OF investment per farm. $ S ($ million) PROJECTS 399,397 4,643 1,120-122,140 1,854.4 54 367,450 674 410- 1,030 247.5 9 372,000 330 200- 380 122.9 4 312,000 121 110- 160 37.7 2 551,000 88 20- 100 48.5 4 (All projects) 2,001,847 1,154 20-122,140 2,311.0 73 (') Project investment per farm is defined as the total investment in a project, divided by the number of farms partici- pating in it. The data thus ignore the distributional pattern of investment per farm within individual projects. Annex 16 Summary of World Bank's Activities in Agriculture, with Projections through FY1976 Actual Program ActualM) Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector Surveys - 3 3 5 7 - 18 41(2) Commitments(US$ million) 366 414 446 510 690 600 2,400 3,630 % of Total Bank and IDA 21 19 18 18 24 12 20 22 Number of Countries 24 23 27 35 40 28 74 83 Lending Operations (No.) 28 32 36 42 53 46 190 273 % of Total Bank and IDA 25 27 28 29 32 17 28 32 Projects under Supervision at End of Fiscal Year 87 103 120 141 170 40(3) 124(0) 189(3) (') Including scheduled for FY1972-73. (') Includes 5 repeat missions. (3) Annual average over period. 01 INDUSTRY Industrialization and Development ....... ...................... 85 Stages of Industrialization ........ ......................... 87 "Non-industrial" Countries ....... ....................... 88 "Industrializing" Countries ....... ....................... 88 "Semi-Industrialized" Countries ...... .................... 89 "Industrialized" Countries ....... ........................ 90 Obstacles to Industrialization ....... ....................... 91 Shortage of Skilled Labor, Management, Entrepreneurship ... . 91 Shortage of Capital Resources ...... ..................... 92 Inadequate Domestic Markets ...... ..................... 92 Balance of Payments Difficulties ...... ................... 93 Limited Access to Markets ....... ....................... 93 Inadequate Infrastructure ....... ........................ 96 Social Aspects of Industrialization ...... .................... 96 Employment ............ .............................. 96 Income Distribution ......... ........................... 97 Location and Environment ....... ....................... 98 Links with Other Sectors ........ ........................ 98 Strategies and Policies ......... ........................... 99 Manufacturing and the World Bank Group ...... ................. 102 Main Features of Past Operations ...... ..................... 102 Sectoral and Geographical Distribution ..... ............... 104 Other Principal Characteristics ...... ..................... 106 Projected Activities, FY1972-1976 ...... ..................... 107 Financing of Manufacturing Operations ..... ............... 108 Industrial Sector Studies ........ ........................ 112 Research .............. ............................... 113 Technical Assistance ......... .......................... 114 Project Appraisal and Supervision ...... .................. 114 The Role of the Bank Group ....... ........................ 118 Need for Concentration of Financing Effort ..... ........... 118 Assistance to "Non-industrial" Countries ..... ............. 119 Improvements in Policy Framework ...... ................. 119 An Outward Orientation for Manufacturing ..... ........... 121 A Promotional Role in Industry ...... ..................... 121 Other Issues ............ .............................. 124 Cooperation with Other Agencies ...... .................. 124 Conclusion ................ 125 p 22 Annexes 1. Some Indicators of Industrial Development ..... ........... 126 2. Characteristics of Bank Group Manufacturing Projects by Organizational Unit, FY1967-1971 ..... ........ 129 3. Bank Group Manufacturing Operations by Sector, FY1967-1971 130 4. Pattern of Bank Group Manufacturing Operations in Selected CoLuntries by Organizational Unit, FY1967-1971 .... ........ 131 5. Bank Group Manufacturing Operations by Organizational Unit and Major Recipient, FY1967-1971 ..... .............. 132 6. Bank Group Manufacturing Commitments by Regions and Countries, FY1967-1971 ....... ...................... 134 7. Bank Group Commitments to Industrial DFCs by Regions and Countries, FY1967-1971 ....... ...................... 136 8. IFC Manufacturing Commitments by Regions and Countries, FY1967-1971 ........ ........................ 137 9. Direct Bank-IDA Manufacturing Commitments by Regions and Countries, FY1967-1971 ..... .............. 139 84 INDUSTRY SECTOR WORKING PAPER * The first part of this paper reviews the stage of industrialization now reached in developing countries, examines the lessons of experience, and explores the principal obstacles such countries are likely to face in advancing their industrial development in the 1970s. It then con- siders some social aspects of industrial development, and concludes with a brief survey of strategies and policies. The second part begins with a review of past Bank Group financing of manufacturing industry, with particular emphasis on the period FY1967-71. Projected Bank Group operations in manufacturing for FY1972-76 are then discussed. A final section examines the main tasks and issues faced by the World Bank Group as it extends its capability for assisting industrial development. The term World Bank Group as used in this paper refers to the International Bank for Reconstruction and Development (World Bank), the International Development Association (IDA) and the Interna- tional Finance Corporation (IFC). Money amounts are in U.S. dollar equivalents. The Bank's fiscal year (FY) ends June 30. INDUSTRIALIZATION AND DEVELOPMENT Industrialization is a key factor in the economic development of most countries. It offers prospects of a growing availability of manu- factured goods, increased employment, improved balance of pay- ments, and greater efficiency and modernization throughout the economy. Industrialization is characterized by technological inno- vation, the development of managerial and entrepreneurial talent, and improvements in technical skills, which lead to rising productiv- ity. Improvements in living standards create a growing and highly elastic demand for manufactured goods. Consequently, industry does not face the same market constraints which affect agriculture, and presents the prospects of a more rapid rate of growth than might otherwise be possible. When promoted along efficient lines, industrial- ization holds out the possibility of easing balance of payments prob- lems through import substitution and the diversification of exports. or The extent to which industrialization fulfills these expectations and contributes to economic development depends on the strategies and policies which countries pursue. If industrialization is given excessive incentives and is not integrated with overall economic development, it may hinder rather than stimulate economic growth. Industrialization is not necessarily synonymous with economic growth, nor can it be expected to propel a nation from technological obsolescence to material prosperity and full employment overnight. Such expectations can lead only to disappointment. Industrialization, like other aspects of development, is a slow and complex process. The impetus to industrialization in today's developing countries stems from the difficulties of colonial and semi-colonial countries in the 1920s, when terms of trade for primary products began to deterio- rate and employment problems were experienced in some countries. The depression indelibly underlined these difficulties, and World War II added new ones by cutting off many predominantly agricul- tural countries from supplies of manufactured products. The prices of manufactures began to rise in the late 1940s and, after the boom engendered by the Korean War, the terms of trade seemed once again to be turning against primary products. At the same time, develop- ment demands aggravated balance of payments problems, and the per capita income gap between agricultural and industrialized coun- tries appeared to be widening. In many cases, balance of payments pressures led to the emergence of import and exchange restrictions, and to a protected market for local industry. In general, industrialization was easiest with respect to products for which a market had already been created by imports. Import substitution with high protection thus became the principal industrialization strategy. It was frequently supported by various credit and fiscal incentives, as well as protective tariffs. Such incentives were designed primarily to attract the foreign investment sought by many countries. Governments not only sup- ported industrialization, but sought to control its course to ensure that the available resources were used to their countries' best advan- tage. This often led to complex restrictions on entry into industry, directly and through the control of imports of capital goods and other industrial inputs. Social implications of industrialization, such as the structure of ownership and levels of employment, were largely ignored. Japan and Hong Kong, with their emphasis on production for exports, were virtually the only exceptions to the import replacement trend until the 1960s, when a number of other developing countries-notably the Republic of China, Republic of Korea, Singapore and Israel- began to concentrate on industrialization via export markets. Many more countries are now seeking to re-orient their industrialization efforts to include exports. The employment and income distribution aspects of industrialization are also receiving more attention in many countries. Stages of Industrialization Developing countries have had a great variety of experience in industrialization, depending upon population size, natural resource endowments and geographical location, as well as upon the strategies they have pursued. In discussing their present situation and experi- ence, it is convenient to classify countries according to their levels of industrial development. One approximate measure of this is the proportion of gross value added in manufacturing to gross value added in commodity' produc- tion. The classification of countries used here and in Annex 1 is based primarily on this approach, although adjustments are made in the case of some countries with large internal markets or other special characteristics. In this classification, countries with less than 20% manufacturing in total commodity production are defined as "non-industrial," those with 20% to 40% are regarded as "industrializing," those with 40% to 60% as "semi-industrialized," and those with more than 60% as "industrialized." An adjustment has been made for large countries such as India, Nigeria and Pakistan, where the internal market is large enough for industry to be relatively highly developed although its contribution to total commodity production is still low. Conversely, in certain other countries, e.g., Argentina, Israel, Mexico and Portugal, the classifica- tion takes into account the fact that the relatively high estimated proportion of manufacturing (61 %) reflects in part the effect of pro- tection in raising prices in the manufacturing sector. It should be emphasized that these stages of industrialization blend into each other and are significant only in an illustrative sense. The various levels of industrialization are not necessarily indicative of the level of overall economic development. The degree of export orien- tation and other characteristics, such as the relative importance of heavy as against light industry, also vary as countries move to higher levels of per capita income and industrialization. 'For purposes of this classification, "commodities" are taken to mean the products of agriculture, mining and manufacturing, and of utilities producing power, gas and water. "Non-industrial" Countries These are countries which are still taking their first steps towards industrialization. Principal opportunities in the early stages usually lie in import replacement in the least technically complicated branches of industry or those based on domestic natural resources. Conse- quently the incipient manufacturing sector in these countries is en- gaged mainly in the import substitution of non-durable consumer goods, such as processed foods and leather products, or in the pro- duction of simple producer goods such as cement. These first steps in the industrialization process are very difficult, as such economies are generally deficient in the basic skills and entre- preneurship which industry demands. Many countries at this stage are tempted to use high protection to accelerate their industrial growth, but this usually leads to inefficient development. Since most manufacturing consists of simple processing activities, even low rates of protection for the final product can result in extremely high protec- tion on a value added basis. Costs are raised in other sectors and eco- nomic growth can be slowed down. Most "non-industrial" countries follow the traditional pattern, de- pending for most of their foreign exchange earnings on specialization in primary exports. In general, new industries oriented toward the export market can develop a comparative advantage only after skills and entrepreneurship have been acquired through production for the domestic market. "Industrializing" Countries These are countries that have made considerable advance in the production of consumer goods. They are moving from assembly oper- ations to manufacturing proper, and into import substitution in inter- mediate goods such as packaging and construction materials and, to some extent, even in capital goods (e.g., pumps, agricultural imple- ments, etc.). Skills and entrepreneurship are being acquired through domestic production, but the injection of intricate technology and capital intensiveness, and the need for economies of scale in produc- tion, make this a more complex stage of industrial development. Many "industrializing" countries have experienced difficulties in making the transition from the simple consumer goods import substi- tution stage. Frequently, balance of payments difficulties have aggra- vated these problems, especially where industrial protection has tended to raise costs of primary production, restricting export growth. Countries with large internal markets can frequently achieve econ- omies of scale and improved efficiency in industry through produc- tion for the domestic market, and thus move to an advanced stage of industrialization. Brazil and Mexico are examples. Those with rela- tively small markets face greater difficulties. Some that have rich natural resources may be able to specialize in primary exports and move to higher levels of per capita income; Malaysia is in this category. Small countries which lack primary natural resources, however, are forced to adopt a deliberate export-oriented industrial strategy at an early stage of their development in order to ensure sustained growth and overcome their balance of payments difficulties; this pattern was followed by Hong Kong and Singapore. In general, all "industrializing" countries, large and small, stand to gain by putting greater emphasis on outward-looking industrial poli- cies which promote internal efficiency and keep the manufacturing sector in line with the outside world. Outward-looking policies do not mean mere emphasis on exports; import substitution if efficiently carried out can be equally "outward-looking." Several developing countries which initially relied on a policy of import substitution have begun to move into the export field. Pakistan began to encourage exports from the mid-1960s; others such as Iran and Cost Rica have begun to look in this direction and have already had a measure of success. "Semi-industrialized" Countries These are countries in which the manufacturing base has been deepened and diversified from assembly and finishing operations to the production, in varying degrees, of most product groups, including capital goods. Basically, most of them have continued to be oriented toward import substitution as they have deepened industrial produc- tion into more sophisticated intermediate and capital goods, and this has often involved a distortion in the allocation of their resources. Recently, however, a number of countries such as Brazil, Colombia, India and Mexico have begun to stress exports of manufactures. Rates of industrial growth in "semi-industrialized" countries have generally ranged from six to eight percent. In some cases industry has stagnated, unable to overcome the difficulties of low productivity and high costs, with their limiting effect on the demand for industrial goods. Greater efficiency and export orientation are issues of major concern for most "semi-industrialized" countries, but the export of manufactures on any substantial scale normally requires a reorienta- tion of policy, including changes in the structure of protection and export incentives. Among the relatively more industrialized developing countries are some, such as the Republic of China, Singapore and the Republic of Korea, which were forced into manufactured exports because of a lack of primary resources and limited domestic markets. There is Aq much to be learned, however, from their industrialization experience. They based their industrialization efforts initially on products that benefited most from low labor costs and required limited amounts of capital and technology. For example, they were able to become bases for the assembly of electronic components for large international corporations. Since then Singapore, and to a lesser extent the Re- public of China, have been able to move to more capital-intensive export products with a higher skill content. Israel adopted this strat- egy from the start. These small industry-oriented countries enjoyed relatively high rates of industrial output and employment growth during the 1960s. It should be noted, however, that even an export-oriented indus- trialization strategy can encounter difficulties if unaccompanied by balanced development of other economic sectors. Aside from the possibility of market restrictions in the developed countries against the importation of manufactured goods, such a strategy can inhibit the growth of agriculture and rural incomes and hence the develop- ment of domestic markets for manufactures. These are problems now being faced in the Republic of Korea. "Industrialized" Countries These are mostly developed countries which have attained high levels of productivity and where industry makes a large contribution to production, exports and employment. Some developing countries, such as Mexico, are approaching the "industrialized" category, al- though they still have a relatively low level of per capita income. The policy framework in "industrialized" countries varies from the relatively strong inward orientation of countries like Australia and the United States to emphasis, as in the small Scandinavian countries, on the achievement of high productivity through specialization in pro- duction and liberal trade policies. Japan is a rather special case; it has relied on import substitution behind high trade barriers as the basis for achieving a relatively high level of productivity and export orien- tation. The emphasis to be given to industrialization in development strat- egy, and the resources to be allocated for this purpose, are generally a function of the stage of economic development. Over the last two decades, manufacturing industry has been the most rapidly growing major sector in the developing countries, and its contribution to over- all growth has increased with its rising share of the Gross National Product (GNP)'. 'Despite this growth, manufacturing remains concentrated in the developed coun- tries. The share of developing countries in total manufacturing production of today's World Bank members (7% to 8%) did not change over the period 1937-1967. QA A number of developing countries have built up relatively broad and diversified industrial sectors. Some, such as Brazil, the Republic of China, Colombia, Iran an.d the Republic of Korea, have made rapid progress in industrialization, beginning virtually from a "non-indus- trial" stage. In still others, such as Spain, India and Yugoslavia, which after World War II could have been characterized as "industrializ- ing," the manufacturing sector has grown greatly in size and sophisti- cation as they have advanced to "semi-industrialized" status. Many of the "semi-industrialized" and "industrializing" countries have already made large strides in infrastructure development. Indus- trialization now holds out for them the promise of rapid gains in do- mestic production to meet their changing demand patterns; in the progressive injection of modern technology into their economies; and, most critically, in the availability of foreign exchange through exports of manufactured goods to meet the increasing import re- quirements of their further economic development. Industrialization in the 1970s has become a central issue of development policy for important segments of the developing world. Obstacles to Industrialization Countries seeking to increase their rate of industrial growth are faced with various economic and non-economic obstacles. These vary from country to country. Some general patterns of constraint are evident, however, which are broadly related to the levels of industrial development: * Shortage of Skilled Labor, Management, Entrepreneurship For developing countries, and in particular for the "non-indus- trial" countries, a lack of skills is frequently the greatest obstacle to industrial development. A serious shortage of skilled and even semi-skilled labor is commonly experienced simultaneously with a superabundance of unskilled labor. There is also usually a critical shortage of entrepreneurship, so that manufacturing opportunities often fail to be recognized and exploited. To resolve this problem, a variety of approaches have been tried and are possible. Local development banks can undertake promotional activity to assist medium and large scale industries. Specialized organizations are required to assist small scale industries in matters of finance, tech- nology, management and accounting. Direct foreign investment can also effect an important transfer of management skills and technology, especially through joint ven- tures. It may also be feasible in many cases to acquire such know-how under contractural arrangements, especially if they provide profit sharing incentives for the sources of technical and management skills. At times, governments have intervened directly in industry to ensure that important industrial projects are not neglected when pri- vate initiative is not forthcoming. These are typically large or complex projects, or those in which the social rate of return appears to be much greater than the pri.vate return. Assistance from governments is frequently important in the area of technical education and training, to help overcome manpower deficiencies in industry. The key ele- ments of entrepreneurial and management skills may be fostered by the promotion of business schools and management institutes. Shortage of Capital Resources Capital resources are generally insufficient in developing coun- tries, and there are difficulties in mobilizing them for industrial invest- ment. Foreign investment can often make a significant contribution to resource availability, but domestic savings finance by far the largest share of industrial investment in developing countries. As countries industrialize, the shortage of capital tends to become less severe because of capital accumulaton within the industrial sector. This fre- quently occurs, however, only at the cost of an excessive concentra- tion of wealth and power. The pursuit of rapid industrial growth together with a broad dis- tribution of industrial ownership requires, among other things, the development of local financial institutions and money and capital markets. The "non-industrial" countries need the establishment and spread of commercial and development banks, and other financial institutions which concentrate on the mobilization of savings. As the use of financial assets grows in the economy and a more diversified industrial base emerges, creating a demand for specialized financial services, more sophisticated types of financial intermediaries dealing in short and long term securities may be needed. Foreign investment can often make an important contribution in this special- ized field. * Inadequate Domestic Markets Developing countries beginning to industrialize generally have limited domestic markets. Without exceptional entrepreneurial re- sources and sufficient incentives, their industries do not attempt to break into export markets. Small markets lead to low levels of pro- duction and a conflict between the need for an efficient scale of production and internal competition, and often result in low effi- ciency and high prices. Countries at low levels of industrial development can do much to avoid such difficulties by selecting an industry mix which will avoid high cost industries. This usually means emphasizing mass consump- tion goods such as food products, textiles and building materials, rather than luxury and semi-luxury consumer durables. Fostering effi- ciency in the production of such articles, so that they may be sold at low prices which the bulk of the population can afford, helps to cre- ate broad and expanding markets. The multiplier effects of incomes generated in the production of these articles will further expand consumption, employment and incomes. As countries become more industrialized and develop an efficient industrial base, their industries can more readily enter export markets and thus begin to break out of market and scale constraints. * Balance of Payments Difficulties The need to combine export production with efficient import substitution is underlined by the balance of payments difficulties con- fronted by most developing countries. Often a policy of import sub- stitution fails to reduce the volume of imports, but leads instead to a change in their composition from manufactured consumer goods to intermediate products and raw materials. Furthermore, the resulting dependence of domestic industries upon the availability of imported inputs makes it much more difficult for the authorities to control the balance of payments without adversely affecting industrial plant utili- zation and employment. An import substitution strategy without regard to internal effi- ciency also has a detrimental effect on exports because of the result- ant higher costs of protected local inputs, often accompanied by increasingly over-valued currencies. As already noted, there have been exceptions; a number of "industrializing" and 'semi-industrial- ized" countries have devised incentives which make exports as profit- able as domestic sales, and sometimes even more so. Although most countries, especially those in the "industrializ- ing" and "semi-industrialized" groups, are now keenly aware of the advantages of giving an outward orientation to their industry, includ- ing greater emphasis on exports, the obstacles to achieving this are formidable. It frequently requires a reorientation of policies and changes in the structure of the economy which are strongly opposed by vested interests. * Limited Access to Markets In addition, the extent to which any new policies can generate exports will depend importantly on access to markets in the industrial nations. The system of protection in the developed countries tends to discriminate against imports of manufactures from the developing 93 countries, which bear higher tariffs, on the average, than manufac- tures traded among the developed countries themselves. Moreover, tariffs tend to rise with the degree of fabrication, discouraging im- ports of food and raw materials in processed form. Non-tariff barriers to trade have also been introduced by devel- oped countries with increasing frequency in recent years. Restrictions on market access are reflected in a variety of administrative and fiscal measures including quotas, subsidies, valuation techniques and pref- erential buying arrangements under government procurement. Despite these obstacles, the growth in exports of manufactures from developing countries has been rapid in recent years; they in- creased at an annual rate of about 15% during the period 1962-69. Nevertheless, the share of developing countries in the total manufac- tured imports of developed countries is still only about 5%, repre- senting only a third of 1 % of the latter's GNP. If the developing countries are to satisfy their projected import requirements in the 1970s, in line with the targets set for the Second Development Decade, they must strive to maintain the roughly 15% rate of growth in manufactured exports which they achieved in the earlier decade. Application of the general preference scheme tentatively agreed to by the developed countries in October 1970 will be a step toward redressing the tariff discrimination against exports of manufactures from the developing countries. Certain important exports have not been included in the scheme, however, and there is a danger that quantitative restrictions might be imposed when imports in any single category become significant. Consequently the general preference scheme, though representing an advance, does not obviate the need for more comprehensive action on the part of the developed coun- tries to remove discriminatory trade restrictions and other barriers which hinder the entry of manufactures from developing countries into their markets. While it is in the long run interest of the more developed coun- tries to move out of products in which they have ceased to have a comparative advantage, the expansion of imports of manufactures from less developed countries will inevitably have an adverse effect on some sectors of their economies. This is bound to arouse opposition and resistance on the part of the affected interest groups unless ap- propriate adjustment assistance policies and programs are introduced concurrently with reductions in tariff and non-tariff barriers. Few developed countries have adopted such policies or programs. Much greater attention to this problem is required to ensure a smooth tran- sition to new production and trade patterns. QA There is a tendency on the part of developing countries to con- centrate their export efforts on a few labor-intensive products such as simple textiles, footwear or leather goods for which markets in developed countries tend to be limited. There is some danger that the terms of trade of such products could deteriorate relative to capital- intensive goods, just as they turned against tropical commodities in the past. "Non-industrial" and "industrializing" countries' difficulties in competing in export markets, and their fears that they are being pushed into "cheap labor industries" with deteriorating terms of trade, need recognition. The following steps might ease the difficul- ties facing developing countries in their efforts to expand exports of manufactured goods, and should lead to an expansion of interna- tional trade: (a) A gradual shift by "semi-industrialized" countries from sim- ple labor intensive into more capital intensive industries, thus making room for the "industrializing" countries which have a comparative advantage in the former. "Semi-industrialized" countries are emi- nently suited for the production and export of capital goods such as machine tools and technologically advanced labor intensive products such as electronics equipment. At present, however, international markets for such products are dominated by exports from developed countries. These are frequently subsidized, both by marginal pricing from relatively large domestic markets and by special medium and long term credit facilities. (b) Provision of efficient and competitive export credit facilities. To develop capital goods exports, manufacturers in "industrializing" and "semi-industrialized" countries usually require subsidized credits in order to compete with exports from developed countries. The question of providing external refinancing for such export credits has figured for some time on the agenda of the United Nations Confer- ence on Trade and Development (UNCTAD). The Inter-American Development Bank (IDB) has established a facility for this purpose, but it has so far been confined to the financing of export transactions between the Latin American countries. Both the Asian Development Bank (ADB) and the African Development Bank (ADB) have been studying this problem for some time, but no decision to mount such a program has yet been taken. (c) Phasing out of existing practices and franchises by multi- national companies which effectively limit exports from their asso- ciated companies in developing countries. 95 (d) Encouragement of trade in manufactures among developing countries. Regional trade groupings provide one way of benefiting from economies of scale within a multinational framework. The case for such collaboration has often been made on the basis that it would permit the establishment of multinational enterprises designed to operate within the wider regional market and exploit the comple- mentarity of participating economies. While these possibilities are well worth pursuing, such multinational ventures are difficult to or- ganize. Even greater gains in trade are likely to accrue through an increase in the scope for specialization and competition within wider regional markets resulting from the elimination of trade barriers. * Inadequate Infrastructure The lack of an appropriate infrastructure is another important obstacle to industrial development. Conditions have improved greatly in the last 20 years, but manufacturers still lack many infrastructure facilities and pay more for the services they receive than their coun- terparts in developed countries. Some countries have been able to minimize the cost of infrastructure for industry by developing ap- propriately sited and well managed industrial estates. In some cases these have been export-oriented, and have greatly facilitated the growth of industrial exports. Examples are the "free" industrial zones in the Republics of China and Korea. Social Aspects of Industrialization The growing concern with the social aspects of development is focusing on employment, a more equal distribution of income, and problems of industrial location. While the allocation of resources between current consumption and investment is an important deter- minant of the future rate of growth, increased attention to employ- ment and income distribution can help expand internal markets, thereby contributing to the developmental process. A careful selec- tion of investments can ensure that social benefits such as employ- ment and the needs of productive efficiency are mutually reinforcing and conducive to sound industrial growth. Employment Although manufacturing has been the largest single source of new urban employment in most developing countries, it has failed to ful- fill the widely held expectation that it could resolve unemployment problems resulting from the population explosion. Indeed, the frame- work of incentives adopted to promote industrialization has in some cases aggravated the problem of unemployment by penalizing the 96 farmer, redistributing incomes in favor of manufacturing, and encour- aging an exodus from rural to urban areas. Manufacturing output in developing countries grew at a rate in ex- cess of 7% per annum over the decade 1955-65, while labor absorp- tion by this sector grew at the rate of 4%. It is, of course, unrealistic to expect industrialization to resolve by itself the employment prob- lems of developing countries. Nevertheless, more could certainly be done at both policy and project levels to expand the employment im- pact of the industrial sector, possibly along the following lines: * By developing a pattern of industry, at the earliest stages of indus- trialization, which encourages the production of mass consumption articles rather than capital intensive consumer durables. * By phasing out incentives which discourage the substitution of labor for capital, such as credit policies which make borrowing for fixed investment cheaper than for working capital, and preferential foreign exchange treatment and duty exemptions for imported capital goods. * By ending licensing and other quantitative controls which tend to discriminate against small industrialists because of the administra- tive problems involved in dealing with them. This is particularly important in "non-industrial" countries which need small scale indus- tries to develop technical skills and management capabilities. Such industries can also play an important role in the production of con- sumer goods, and as "sub-contractors" for larger industries, both domestic and foreign. Assembly of electronic components in coun- tries with low cost labor is an example of such international "sub- contracting." * By encouraging maximum utilization of existing industrial facili- ties through multi-shift production and other appropriate methods, and by greater attention to the process of technological transfer and adaptation, with a view to devising modern but labor-using equip- ment and production methods. Income Distribution Industrialization can contribute to a more equitable income distri- bution by broadening the middle class group of society and creating a new industrial work force with relatively high incomes. Such groups normally remain small, however, and a growing concentration of economic wealth frequently offsets gains from such trends. The distribution of income is, of course, affected by the manner in which it is divided between capital and labor, as well as among differ- ent sectors and regions within the economy. While there can be an 97 economic argument in favor of high rewards for capital, thereby en- couraging capital accumulation for further industrialization, the resulting inequalities in wealth can endanger the prospects for social and political stability, which are critical preconditions for uninter- rupted economic development. For these reasons, it is desirable to focus on a policy of more equal distribution of income through in- creasing employment and lowering prices of manufactured products, rather than trying to protect industry to ensure high industrial profit- ability and high incomes for factory workers. Location and Environment The advantages of geographic concentration of industry have led to some over-concentration of both industry and population. In some countries this is now being followed by a swing to extreme disper- sion, which also entails high social costs. Indeed, manufacturers' costs may become so high that they require very high levels of protection. The developing countries can avoid some of the excessive social costs that have accompanied industrialization in the developed coun- tries. A balanced and planned urban growth, combining investment in utiiities, industrial estates and social facilities-such as schools, hos- pitals and housing to make for viable "growth poles" as an alternative to both urban sprawl and rural neglect, presents some possibilities of turning urbanization from a problem to a tool of development. The organizational and resource mobilization problems, however, are formidable. Developing countries also have reason to be concerned with the impact of industrialization on the environment, and on public health and welfare. A carefully planned and implemented "preventive" strategy in this area, reflecting the particular needs and stage of de- velopment of each country, is likely to avoid costly investments at a later stage to overcome the consequences of earlier neglect. Links with Other Sectors Where policies affecting industry have been divorced from those for other sectors, industrialization has tended to occur in enclaves, isolated from the rest of the economy. This frequently results in a weakness of linkages between sectors, and in lower overall develop- ment. The intersectoral implications of industrial development need careful attention. In many developing countries, excessive incentives for industrialization have resulted in erosion of the comparative ad- vantage in agriculture. An optimal economic strategy requires the balanced growth of agriculture and industry, and special attention must be given to the linkages between these two sectors. 98 Strategies and Policies Any consideration of strategies and policies to promote industrial- ization must take into account the fact that industry in most countries is owned and operated largely by private enterprise. Therefore, a great deal of importance must be attached to policies and incentives which will have the desired impact on the motivations of private entrepreneurs and the investment climate in general. The predominantly private character of the industrial sector pre- sents certain special problems in the implementation of industrial investment programs. It also has a bearing on the choice of instru- ments to be used. It is obviously within the province of every government to deter- mine the relative importance it wishes to assign to the public and private sectors in industry. It should be borne in mind, however, that public manufacturing enterprises tend to be successful when they are left free from political interference and forced to face the pressures of domestic or foreign competition. There is always a danger that excessive government intervention may inhibit the mobilization of entrepreneurial talent and the decentralization of initiatives and deci- sion making which are necessary for rapid industrial growth. Apart from the predominantly private nature of industry, there are other factors which contribute to the complexity of the industrial sector. The rapid pace of scientific and technological progress which characterizes this field is in itself a complicating factor from the stand- point of industrial policy formulation and investment decisions. Moreover, like agricultural development, industrialization is not merely a matter of economic evolution but depends on a country's social and political organization and policies. Insofar as the economic aspects are concerned, it is evident that industrial strategy must be part of an overall economic strategy to be successful. Basically, outward-looking strategies make sense for all developing countries, even those in the "non-industrial" category, which can thereby avoid some of the difficulties now being experienced by the more industrialized countries. An outward-looking policy is not syn- onymous with export orientation; it means attention to comparative advantage, specialization, and development of the appropriate struc- ture and techniques of industry, so that the sector can develop efficiently and in line with the outside world. A policy of import- substitution based on the utilization of domestic resources could be characterized as "outward-looking" if the efficiency of the industries concerned were comparable to that of similar enterprises elsewhere. Some of the best opportunities for manufactured exports may arise in conjunction with import substitution, but they will be realized only 99 if the import-substitution itself is reasonably efficient and thus outward-oriented. In many developing countries, at least until recently, there was an excessive shift in emphasis from primary production to industry, and from outward-looking to inward-looking policies oriented toward the domestic market. A reaction against colonial patterns of special- ization has frequently been responsible. The resulting combination of policies has often proved inefficient, slowing down economic growth by penalizing investment in primary production. It has also encouraged undue reliance on instruments such as tariffs and quotas which favor production for the domestic market over exports, in contrast to a more balanced system of incen- tives including adjustments in exchange rates which would keep domestic costs and prices in line with the outside world. Import substitution and a certain degree of protection for infant industries is a normal and often indispensable initial phase in the in- dustrialization process. Yet a number of developing countries, particu- larly those in the "industrializing" and "semi-industrialized" group, are recognizing that further industrial expansion based on a policy of continued import substitution faces critical market constraints. By contrast, greater emphasis on manufacturing for exports will permit the establishment of larger industrial units with economies of scale, provide access to scarce foreign exchange resources, and give new momentum to industrial development. Progress in this direction obviously depends upon a country's resource endowments, the exist- ing structure of production and protection policies, the available in- dustrial skills, and the time required to bring about the necessary changes. It is impossible to give a policy prescription for industrialization except in these very general terms, since the appropriateness of a strategy and policy mix varies from country to country and is related in each case to the level of industrial development and the economic and social objectives the country wishes to pursue. Nevertheless, it might be useful to outline briefly some of the principal policies and policy instruments which appear to be successful in fostering effi- cient and sustained industrialization: (a) A policy of moderate protection, avoiding as far as possible out- right import prohibitions and quantitative restrictions. Such a policy is consistent with protection of "infant industries," especially when efficient operations are contingent upon large scale production, but the protection granted should be temporary and manufacturers should be put on notice with respect to a tariff reduction program. 100 Moreover, tariffs should be as uniform as possible, divided at most into a few categories, or "bands." Widely differentiated tariffs, specifically designed to meet the needs of marginal firms, tend to do the most harm by subsidizing inefficiency, misallocating resources and discouraging exports. Appropriate export incentives should be provided where neces- sary to counterbalance any disincentives to export created by pro- tection against imports. A rational approach to export orientation, however, does not point merely to extensive export incentives; it suggests instead a program to reduce protection to reasonable levels. The more moderate and temporary the protection, the easier it will be to achieve export orientation. (b) Minimum use of direct controls, as on the entry of new firms into manufacturing or on pricing of manufacturing output. When there are problems of industrial location or market fragmentation, or when participation by foreign investment is involved, there can be a case for controls over entry of new firms into manufacturing, but such controls should not be extended indiscriminately. Similarly, price controls should be reserved essentially for those special cases where a monopoly is preferable to market fragmenta- tion, and where abuse of the monopolistic situation cannot otherwise be prevented. For most manufacturing activity, the emphasis should be on the use of market forces to ensure efficiency and low prices. (c) A fiscal policy framework designed to strengthen the competiti- tive position of industry and to encourage resource use in accord- ance with comparative advantage. Such an approach would argue generally for moderate direct taxation, "non-cascading" indirect taxation, and the application of excise duties on both imports and local production, instead of placing exclusive reliance on high import duties to control consumption. It would also suggest that accelerated depreciation is a questionable fiscal incentive for industry in most situations because of its tendency to encourage capital intensive projects. (d) Monetary and credit policies aimed at creating a financial en- vironment conducive to a high rate of savings and the efficient mobil- ization and allocation of available resources. Aside from the obvious need to maintain reasonable financial stability, this usually means avoiding artificially low interest rates or concessionary credit policies. It also means the creation of appropriate institutional facilities and financial instruments to help mobilize domestic resources for pro- ductive investments. (e) Labor and manpower policies designed to achieve maximum employment. This implies avoidance of labor laws and trade union I1 M practices which are inconsistent with employment objectives and encourage capital-intensity, for example by discriminating against night work and multiple shifts. It also implies provision of incentives for training and use of labor. Furthermore, as noted above, it requires the avoidance of financial policies which favor the use of capital over labor. The implementation of policy is, of course, as important as the formulation of the appropriate policy mix itself. Yet weaknesses in administration are one of the main conditions of underdevelopment, and they take time to overcome. The approach outlined offers no final solutions for the problems of industrialization. Rather, it recog- nizes the complexities and seeks to identify policies which are likely to contribute to industrialization with least cost and maximum benefit in the 1970s. MANUFACTURING' AND THE WORLD BANK GROUP Table 1: Summary of Actual and Projected Bank Group Financing' for Manufacturing, FY1969-76 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector and Sub-Sector Studies - 4 8 5 6 1 23 35 Commitments ($ millions)' 314 361 347 675 575 786 2270 3115 Percent of Group total 17 15 13 24 20 14 18 20 No. of countries' 16 20 19 31 37 39 50 60 Financing Operations (No.) 33 40 30 45 55 114 205 300 Projects under Supervison, end of fiscal year 111 138 145 160 180 92' 145' 210' 'Not including credits for industrial import programs. 2Does not reflect financing through regional development finance companies for Africa and Latin America. 'Annual average. Main Features of Past Operations The two main instruments on which the Bank Group has relied since its manufacturing sector operations began in developing countries 'The discussion of Bank Group operational activities is in terms of the "manufac- turing" sector since this paper excludes consideration of mining operations. In dis- cussing general problems and issues relating to industrialization, however, the term "industrial" is used. I n9 in fiscal 1951 have been the direct financing of projects, and their indirect financing through lines of credit to financial intermediaries known as development finance companies (DFCs). Through these channels, the Group committed a total of $1,314 million to develop- ing countries during the five fiscal years 1967-71. This represented almost 15% of total Bank Group financing in the period, in all sectors. Table 2 gives comparative figures for the three five-year periods, 1957-71. Table 2: Bank Group Financing for Manufacturing Projects' In Developing Countries, FY1957-61, 1962-66, 1967-71 (US$ millions) 1957-61 1962-66 1967-71 Arnount % Amount " Amount ° Direct Financing Bank and IDA 38 26 62 12 132 10 IFC 45 31 101 20 305 23 Sub-Total 83 57 163 32 437 33 DFC Financing' 63 43 350 68 877 67 Total 146 100 513 100 1,314 100 'Does not include credits tor Industrial import programs. 'Includes all IFC investments through financial intermediaries. Excludes DFC financing in developed L uO ttries (Finland). Changes in the composition of Bank Group manufacturing opera- tions in developing countries have reflected mainly the commence- ment in FY1957 of financing by IFC and a steep rise in commitments to DFCs. Financial operations in the manufacturing sector in the 1950s included lending to such developed countries as Japan, Italy and Australia; these were virtually phased out by the beginning of the 1960s, while those in developing countries showed a steady rise. Dur- ing the period FY1967-71, direct project financing was more than five times higher than a decade earlier, due largely to the growth in IFC's operations. Financing through DFCs rose very sharply during the early 1960s, and over the past decade has represented roughly two thirds of the Group's assistance to manufacturing in developing countries. The net- work of DFCs associated with the Bank Group increased from eight at the end of FY1961 to 40' a decade later. In many cases the Group played a critical role in establishing or strengthening these institu- tions. Although DFCs do not serve the manufacturing sector exclu- sively, most are heavily specialized in lending to or investing in manufacturing concerns. 'Includes two DFCs exclusively for tourism and a capital markets development insti- tution. I ('i The Bank Group share of total industrial investment in developing countries has been quite small. Very rough estimates suggest that the direct financial contribution of the Bank Group to total capital forma- tion in the manufacturing sector in developing countries in the five years 1967-71 was on the order of 1.5% to 2%. Nearly three fourths of total industrial investment was financed from national sources-the reinvestment of profits and reserves, transfers from financial intermediaries, and private non-institutional savings. Nevertheless, the direct financial contribution of the Bank Group represented nearly a third of all foreign official finance avail- able for this purpose, and approximately 10% of the total from external public and private sources combined. Bank Group operations in manufacturing have had a significant "catalytic" impact. The aggregate cost of Bank Group financed proj- ects was about $7,500 million during FY1967-71. On the average, of every dollar invested in these projects, only 18 cents came from the Bank Group. The participation of associated investors was most pro- nounced in projects directly financed by IFC, and by DFCs with Bank Group funds. More than three quarters of IFC projects were joint ventures involving technical as well as financial partnership between foreign investors and local sponsors (see Annex 2). Sectoral and Geographical Distribution The sectoral and geographical distribution of Bank Group opera- tions in the manufacturing sector does not reveal a consistent indus- trial strategy, on either a country or a sector basis. When "retailing" its assistance, i.e., through direct project financing, the Group has tended to support relatively large projects, most of them in activities which fall under the heading of "heavy manufacturing," and largely in the "industrializing" and "semi-industrialized" countries at the level of development necessary to absorb such projects. When the assistance has been "wholesaled" through development finance companies, a shift can be observed towards smaller projects and a more even distribution between "light manufactures" and "heavy manufactures." Most of the development finance companies which have received Bank Group support, however, have also been in the "industrializing" and "semi-industrialized" countries (see Annexes 3 and 5). In general, direct financing of manufacturing industries in the past has been on a project-by-project basis, with no organized program of project identification and promotion to assist member countries in meeting their industrialization objectives. In the case of IFC, most I flA projects have been referred from outside sources, both domestic and foreign. The Bank and IDA have financed only a few manufacturing enter- prises directly. This reflects the assignment to IFC of the primary role in handling private investment proposals, as well as inhibitions felt until recently by the Bank and IDA in financing government owned enterprises. Other deterrents have been the unwillingness of govern- ments to provide the necessary guarantees, and the unwillingness of some project sponsors to conform to Bank standards of international competitive bidding. Despite the lack of an organized program of direct financing, many manufacturing projects assisted by the Bank Group have contributed importantly to strengthening or diversifying the manufacturing sector, often introducing for the first time the production of new intermedi- ate and capital goods. More than 85% of direct project financing in FY1967-71 was directed to "heavy manufacturing," including metal- lurgic and machinery industries, chemicals, non-metallic minerals and pulp and paper (see Annex 3). Manufacturing operations of the Bank Group during the five years were concentrated in the "industrializing" and "semi-industrialized" countries. From a regional standpoint, the strongest emphasis was on Asia, which absorbed 50% of total project commitments in develop- ing countries'. The Latin America and Caribbean area received only-18%,although it included certain countries with rapidly growing manufacturing sec- tors. There were various constraints on Bank and IDA industrial operations in the area, including non-availability of government guar- antees, questions of creditworthiness, and Bank policy precluding loans to government owned DFCs (a policy which was changed in 1968). African countries, most of which fall in the "non-industrial- ized" category, accounted for only 4% of manufacturing project commitments in developing countries. Deeper insight is provided by an analysis of a sample of 22 coun- tries which accounted for 75% of total Bank Group manufacturing commitments in developing countries during FY1967-71 (see Annex 4). This indicates that Brazil, Colombia, India, Iran, Pakistan and Turkey have been the major countries in the Bank Group program. Their 50%') share in overall commitments is similar to their share in population of developing member countries, though more than their roughly 35% share of GNP. Only Indonesia among the "non-indus- trial" countries figures prominently in the past program. 'Regional percentages shown in the Annexes reflect the inclusion of developed coun- tries (Finland), except wheie otherwise specified. 1ns Other Principal Characteristics The channeling of resources for manufacturing through DFCs con- stituted the bulk of the Bank Group program in many countries during this period (see Annex 4). This was particularly the case in "indus- trializing" countries (Colombia, Iran, Republic of Korea, Morocco, Nigeria, Pakistan, Philippines, Turkey), and in India among the "semi- industrialized" countries. Lending and investing in manufacturing projects through IFC has also acquired prominence in a number of country situations (Brazil, India, Philippines and Turkey). By contrast, direct project lending by the Bank and IDA was confined to only four countries - Brazil, Indonesia, Pakistan and Yugoslavia - and altogether accounted for only about 10% of total Bank Group com- mitments. The majority of projects catered exclusively to the domestic mar- kets of member countries in which they were situated. Import substitution was the predominant characteristic of Bank Group opera- tions, just as it was the prevalent strategy of industrialization adopted by member countries. More than half the total of Bank Group projects during the five years, accounting for a similar proportion of overall commitments, did not export any part of their output. At the other end of the spec- trum were roughly 15% of the number of projects, accounting for a similar percentage of overall commitments, which exported more than half their total output. Available data also suggest a somewhat higher degree of export orientation in projects financed through DFCs (see Annex 2). However, this does not appear to have been the result of deliberate policy. The bulk of manufacturing projects financed by the Bank Group was in the private sector. IFC is limited by charter to lend or invest in non-government enterprises, although nearly a fifth of its client firms during FY1967-71 had some government participation in their equity. Similarly, DFCs have financed private enterprise, with very few exceptions. Until recently, the Bank as well did not finance indus- trial units in the public sector. Recent policy decisions have enhanced the Bank's flexibility in dealing with government industrial enter- prises directly or through DFC-type operations. The bulk of Bank Group funds for manufacturing have been di- rected to the financing of fixed industrial investment, and particularly its foreign exchange component. Permanent working capital and local currency expenditures are financed by IFC, but only when the resources cannot be mobilized domestically. Occasionally, DFCs have also financed permanent working capital requirements. While the present mix of instruments allows the Bank Group to 1n' play an effective role in a variety of industrial and country situations, the Group has not provided much assistance, directly or indirectly, to small scale industry, particularly firms employing less than 20 workers (see Annex 2). A few operations have been carried out for the direct benefit of the small scale industrial sector (industrial estates, etc.), but most have benefited medium and large scale firms. Until recently, Bank Group operations appear to have had rela- tively little impact on the broad evolution of industrial policy in developing countries. The excessive inward orientation of industry and distortions in the system of fiscal and other incentives appear to be almost as marked in recipients of Group assistance as in other countries. The shift towards export orientation in countries such as Pakistan seems largely unrelated to Bank Group operations. While the "project" orientation of operations in the manufacturing sector, plus the predominantly private character of most manufactur- ing investments, are undoubtedly important elements in this picture, it also appears that the general approach in the past has been to undertake direct financing of projects, or to replenish the resources of DFCs, because the investment appeared worthwhile itself rather than as part of a carefully considered industrial strategy in the coun- try concerned. It is difficult to make any reliable assessment of the economic con- tribution made by Bank Group operations to industrialization in indi- vidual countries. For example, while there can be little doubt that DFCs have made an important contribution toward widening the possibilities of industrial growth in the developing countries, their performance appears to have been less than satisfactory in develop- ing local capital markets. Similarly, direct project financing by the Bank Group appears in certain cases to have been directed to projects which have required substantial protection. A strengthened follow-up program, particu- larly on the economic side, is a prerequisite for arriving at better knowledge and judgment about the economic contribution of past Bank Group operations in the manufacturing sector. Projected Activities, FY1972-76 The Bank Group does not yet have a coherent industrial sector program designed to meet specific objectives. As explained below, however, the more intensive work initiated in the area of sector studies, industrial policy analysis and investment promotion should assist in the development of a more integrated and goal-oriented program. 107 An attempt is made in this section to provide a brief account of the activities envisaged in the manufacturing sector by the Bank's Indus- trial Projects and Development Finance Companies Departments and by IFC over the next five years, under the following headings: Financ- ing of Manufacturing Operations, Sector Studies, Research, Technical Assistance, and Project Appraisal and Supervision. Information on FY1972-76 manufacturing operations is largely an aggregation of existing programs of the different organizational units concerned. In order to evolve an integrated manufacturing sec- tor program for the period, a great deal of further effort will be required to complete work now under way to: * Re-define Bank Group objectives in the manufacturing field. Cer- tain operational issues and alternatives are presented in the final section beginning on page 118. * Prepare a more definitive program of sector and subsector studies pertaining to the manufacturing field. * Evolve specific programs and proposals for Bank Group manu- facturing operations on a country-by-country basis, taking note of the relative priority of the sector and the financial constraints on Bank Group activity in particular country situations. * Derive the manpower and organizational requirements for im- plementing the manufacturing sector program. Financing of Manufacturing Operations Aggregate new loan and equity commitments for manufacturing are projected at somewhat more than $3,000 million over the five years, almost two and a half times the level of FY1967-71. The share envisaged for manufacturing operations is almost 20% of projected Bank Group operations as a whole in developing countries, compared to 15% in FY1967-71. Table 3: Projected New Commitments for Manufacturing,' FY1972-76 Percentage Increase U5$ million Percent Over FY19167-71 Direct Financing Bank and IDA 1,200 39 809 IFC 480 15 57 Sub-Total 1,680 54 284 DFC Financing' 1,435 46 64 Total 3,115 100 137 'Data represent all projected commitments to the manufacturing sector through IFC and the Bank's Industrial Projects and DFC Departments, with the exclusion from IFC and Industrial Projects data of 'mining" and "tourism" projects. Data oni DFC financing relate to total projected Bank Group crmmitments to industrial DFCs, part of whose sub-projects will be in non-manufacturing ventures (the proportion was 14% in FY1968-70). 'Including IFC investments through financial intermediaries. -1 nA Increase in Bank, IDA financing The outstanding feature of these projections is, of course, the large increase in direct financing of manufacturing projects by the Bank and IDA. This is attributable largely to the change in policy with re- spect to financing government owned enterprises, and the prospect that an increasing number of such entities may be able to present acceptable loan applications in the coming years. In FY1972, for example, it is expected that eight manufacturing projects calling for direct lending by the Bank and IDA, totaling some $334 million, will be presented to the Executive Directors, all involving government owned or controlled enterprises. This is an exceptionally large program, much affected by planned loans to three Brazilian steel companies totaling $186 million during fiscal 1972. The pipeline for future years is smaller, and progressively undefined because of uncertainties regarding specific proposals which may be submitted for financing. Nevertheless, the projected figure of $1,200 million over the five years FY1972-76 appears to be a reasonable rough estimate of the volume of lending that may de- velop in this category. In the early years, at least, the program is concentrated in large projects with heavy capital inputs, such as steel, fertilizer, forest prod- ucts, etc. Starting in FY1974, however, it is expected that, as a conse- quence of more active project identification and promotional work, Bank and IDA financing will be extended into such areas as industrial estates and other projects for assistance to small scale industry, re- habilitation loans for selected industrial branches, and projects oriented especially toward the promotion of export industries. Financing by IFC In the case of IFC, the project pipeline is a particularly poor guide for estimating future operations, except to some extent for the re- mainder of FY1972. The approach adopted has been to estimate the likely volume based on current levels of IFC activity, identifiable prospects for future IFC involvement, an assessment of the country environment for private investmentand likelyareaswhere new private investment would be consistent with country economic priorities. IFC's reliance on the private sector for new investment proposals, the changing climate for foreign private investment, and the incipient stage of IFC's promotional work make programming for IFC a much more tentative exercise than for other Bank Group organizational units concerned with industry. Also the size and range of future IFC operations will depend importantly upon the equity resources avail- able to it through an increase in its capital or in other ways. 109 As part of its future efforts, IFC will adopt a more systematic pro- gram of promotional work, seeking out, in coordination with the Bank, areas of high industrial priority in which IFC may be an effec- tive catalyst for new investment. While an industry breakdown of future IFC activities is not feasible, a number of areas in which the Corporation is likely to be active can be identified. Steel, machinery, chemicals, fertilizers and synthetic fibers are likely to be important in several Asian countries and in Latin America. In Africa, an attempt will be made to encourage domestic resource-based industries. Financing through intermediaries The program for supporting intermediaries in financing industrial projects is less uncertain. Repeat operations, involving 36 develop- ment finance companies actively associated with the Bank Group, account for two thirds of overall commitments envisaged during FY1972-76. In addition to catering to the needs of DFCs already asso- ciated with the Bank Group, it is planned to build new institutional relationships. Forty proposals of this kind are under consideration, but the likelihood is that no more than 23 will mature by the end of FY1976. The expansion visualized is still ambitious. Two points are particu- larly noteworthy. First, financial support is expected to be extended to three additional development banks of regional scope. Second, the program implies a large expansion of national DFCs in Africa. This is a formidable undertaking, given the special problems of building or reorganizing institutions in that part of the world, and will require a relatively high component of technical assistance from the Bank Group. The establishment of new institutional relationships cannot be pro- jected with as much confidence as the "repeat operations." However, if the Bank Group succeeds in consummating the 23 proposals men- tioned above, the DFC network associated with the Bank Group will increase its country coverage by 10. Moreover, the coverage would be intensified through the establishment of new institutions in five countries in which Bank-financed DFCs already exist. Mobilization of savings An important new element in the FY1972-76 program is more di- rect assistance to member countries in mobilizing private savings for investment, particularly through the development of capital markets and other appropriate financial institutions. Development finance companies supported by the Bank Group have already made a contri- 110 bution in this field, but the need is now felt to establish new inter- mediaries which are equipped to provide specialized financial and technical services to business and industry. IFC envisages direct in- vestment in several new money and capital market institutions; its first investment in this field was in the Korea Investment Finance Corporation (KIFC) in FY1971. Shifts in lending patterns In view of the substantial uncertainties regarding the flow of proj- jects to the Bank Group in the industrial field, it is possible to make only very tentative statements regarding the pattern of these projects in FY1972-76. It seems clear that they will be of differing importance in different kinds of countries. In the more industrialized countries, manufacturing projects may constitute 30% or more of total Bank Group financing. On the other hand, there are several "non-industrial" African countries where the share is expected to be less than 10%. The relatively low share in African countries reflects their earlier stage of development and the fact that technical assistance, rather than finance, is their principal requirement at this time in the indus- trialization field. Overall, less than 10% of FY1972-76 financing of manufacturing is expected to go to "non-industrial" countries. "Semi- industrialized" countries may account for close to half, and "indus- trializing" countries are expected to receive the balance. There will probably be substantial shifts in regional shares from the past pattern of Bank Group manufacturing operations. The propor- tion of total FY1972-76 financing projected for Asia is about 35% as compared with half during FY1967-71. This reduction will be taken up chiefly by a projected increase in the share of Latin America and the Caribbean from close to 20% in FY1967-71 to about 30% in FY1972-76. Although Africa's share is expected to remain low, the actual amount in FY1972-76 may be several times that of FY1967-71. Some rough projections have been made of the likely levels of manufacturing investment in developing countries, in order to obtain some quantitative indication of the relative importance of the Bank Group's five-year financing program. They are based mainly on eco- nomic projections prepared in the Bank in connection with other studies. It is estimated that during FY1972-76 Bank Group disbursements will directly finance 2.5% to 3% of total new fixed capital formation in the manufacturing sector in developing countries. As noted earlier, the comparable figure for FY1967-71 is estimated at roughly 1.5% to 2%/. 111 Financial operations do not reflect adequately the promotional or "catalytic" role the Bank Group performs. Assuming that past shares of total project costs financed by the Bank, IDA and IFC remain un- changed, total manufacturing investment likely to be induced by them through direct project financing might be on the order of about 5% of the projected aggregate. Greater uncertainty attaches to DFC sub-projects financed by the Bank. Including these, however, the proportion of total manufacturing investment with which the Bank Group can be considered to be directly or indirectly associated would be about 13%. It has also been estimated, very roughly, that the contribution of Bank Group disbursements to fixed capital formation in manufactur- ing might represent some 10% of the total new capital flows from developed to developing countries for investments in manufacturing, excluding petroleum. These estimates are subject to a large margin of error, and are useful only as broad points of reference. Industrial Sector Studies Since each country's economic circumstances and industrialization problems are unique in many respects, an expanded program of sec- tor studies is a prerequisite for strengthening the Bank Group's indus- trial program. Careful study and understanding in depth of the industrial sector in individual countries is required to identify their investment and technical assistance requirements and ensure that Bank Group activities make the maximum contribution to indus- trial growth. The principal objective of sector studies may be broadly defined as the acquisition of basic intelligence on which a developmental sector lending program can be based. This means the identification of structural and policy problems in industrial growth to enable the Bank Group to plan its lending operations more effectively and in a longer time perspective than has been possible hitherto in this sector. Such an approach includes the stimulation of project preparation, early identification of Bank Group financing possibilities, and the evaluation of projects through pre-investment programs to ensure that there is a broad enough base for project selection. Project prepa- ration and selection must take into account not only priorities within the industrial sector, but also the impact of industrialization policies. In the evaluation and desigrn of such policies, the Bank Group might, where requested, provide technical assistance to ensure that mem- bers' economies obtain the maximum benefit from given levels of industrial investment. Sector studies thus have three principal aspects: (a) the provision 112 of a basic data framework on the structure of industry, its relevant factors of production, its growth and, to the extent possible, its rela- tionship to the rest of the economy; (b) an analysis of policies relevant to the functioning and growth of industry; and (c) the identification and selection of projects for Bank Group lending. An industrial sector study program was initiated in the Bank two years ago. Some half dozen major industry sector reports have so far been completed and another five or six are planned for each of the years FY1972 and FY1973. In addition, the manufacturing sector is treated periodically in country economic reports. These rarely pro- vide the required information base, however, or the analysis in depth necessary for Bank Group technical assistance on policy formulation and project selection. As soon as sufficient experience has been gained, a comprehensive program of sector studies for FY1972-76 should be evolved in accord- ance with the emphasis on in-depth, problem-oriented country eco- nomic work now being planned. Two important considerations need to be kept in mind. First is the need to have a comprehensive and reasonably up-to- date study of the industrial sector for all countries which are likely to be major recipients of Bank Group industrial assistance in the 1970s. The focus and depth of such studies will vary from country to coun- try, depending upon the individual situation. The objective should be reasonably comprehensive coverage by the end of FY1976. The second major consideration is the importance of developing a capability to organize studies in a select group of "non-industrial" countries which request the Bank Group's assistance in the formula- tion of their industrial development programs and policies. Research Bank Group research in the industry field has concentrated in the past on studies of the structure of specific industries. In the last two years, however, there has been a change toward an emphasis on the problems of industrialization and trade, and on industrial lending issues. Much of this newly oriented research is also country focused, so that broad issues may be viewed simultaneously from the point of view of solving a country's particular industrial problems and from a more generalized viewpoint, taking advantage of the Bank Group's unique comparative perspective. The increase in Bank Group lending activities, together with a growing appreciation of industrialization problems, is leading to in- creasing demands for -esearch as operational back-up. Greater selec- tivity is therefore required if the research undertaken is to meet the 111 principal current needs as well as to identify and anticipate problems which are likely to be most important in the future. A recent review of research proposals has been the first step in improving the relevance of research to the Bank Group's operations, and a proposal to integrate research with country industry sector cov- erage, together with closer coordination with lending activities, is intended to ensure that close links are maintained between research and operations. Technical Assistance The most important aspect of the Bank's technical assistance in industry arises from its sectoral study work. It is more frequently in- formal than formal in nature, and arises as much in the course of the work of country or research missions as from their resulting reports. The staffing of such missions is to some extent, therefore, a form of technical assistance. In addition, the Bank Group assists countries with specific problems by helping to find, and in some cases finance, appropriate consultants. The quality and effectiveness of such assistance depends on its con- tinuity and coordination within the Bank Group. For this reason, and to ensure that the manpower devoted to this task is commensurate with both the requirements of technical assistance and the available resources, the sector study program may be seen as central to the technical assistance program in industry. In a narrower but nevertheless important sense, technical assist- ance is also a concomitant of direct lending activities. The Bank Group's role in project identification, selection and follow-up can play a seminal role in the development of a country's industrial sec- tor. The institution building assistance provided in the course of lending to development banks and other capital market financial institutions is equally important. The Bank Group has an important technical assistance arm in the Economic Development Institute. In addition to a general treatment of industrial problems in its development and general courses in Washington and in the various developing countries, it has conducted an industrial projects course for some 25 candidates annually since the mid-1960s', and from FY1972 two such courses will be scheduled each year. Project Appraisal and Supervision It is generally agreed that industrialization should be oriented to a country's comparative advantage in resource allocation. While the 'In FY1971 there were some 150 applicants for this course. series of analyses and judgments reached at a country, sector and project level can contribute to a better understanding of the direc- tions in which such advantage might lie for individual countries, and can thus avoid serious mistakes in investment, there is no operation- ally foolproof method of evaluating comparative advantage. The only meaningful approach to ensure that investment opera- tions will be in line with comparative advantage is to improve the analysis and understanding of a country's industrial sector, to arrive at a more informed set of judgments in regard to its industrial priori- ties, and to initiate project evaluation at a very early stage. All industrial projects directly undertaken by the Bank Group, and those DFC sub-projects which are subject to "prior approval" by the Bank (roughly a fourth of all sub-projects), are currently reviewed in technical, financial and economic terms. Technical feasibility and a reasonable degree of profitability in relation to the risk involved are, and should of course continue to be, important investment criteria for the Bank Group's industrial operations. In this, the Bank Group has accumulated considerable expertise. Recently, however, improved methodologies have been developed for the assessment of likely costs and benefits accruing to the econ- omy from investments in individual projects. The Bank Group is now engaged in introducing these new methodologies in its own project evaluation, and in assisting development finance companies to adopt them as well. The economic rate of return At the project level, the "economic rate of return" on capital- the rate which equalizes the present value of total economic costs and benefits over the life of a project-provides the most meaningful and comprehensive quantitative indication of the net economic benefits to be derived from the investment. This measure can shed light not only on the contribution of the project to domestic product but also on the protection and balance of payments implications of an investment. There are several difficult problems in estimating the economic rate of return, however, which need careful attention and require the exercise of judgment. These center around the identification of ap- propriate long term international prices for project outputs and in- puts, the measurement of external economies and diseconomies, the estimation of shadow prices for the primary factors, and the rate at which project efficiency will improve over time. Since industrial products are generally traded internationally, the availability of reliable international price data is a key element in 11 r evaluating industrial projects. Such data are often difficult to obtain, however, because of the multiplicity of industrial products, changing sources and conditions of supply and the prevalence of temporary "marginal" pricing. Moreover, external economies are generally difficult to quantify and shadow price estimation for wages and capital requires a pro- found knowledge of a country's economy. Probability and sensitivity analysis can assist in evaluation of these aspects but judgment has to be exercised about the degree to which it is worthwhile extending the analysis for any project. The issue of protection An important issue in industrial project evaluation is that of pro- tection. Developments in industrial policies in a number of develop- ing countries make it necessary to take a careful look at this aspect in connection with project evaluation. Measures of "nominal" protection and "effective" protection of value added can throw useful light on the competitiveness and effi- ciency of industrial projects. They need careful interpretation, how- ever, in the context of the overvaluation of a country's currency, the financial return on the project in relation to the opportunity cost of capital, and the international prices used as a yardstick for measuring the degree of protection. In particular, the "effective" protection measure can be very sensi- tive to small changes in assumptions regarding prices and exchange rates. The economic rate of return, by introducing a time dimension, is a more comprehensive measure of a project's net benefits on the basis of a zero level of effective protection. Alternative approaches Since no single economic measure is best suited for all projects and all occasions, alternative approaches to economic evaluation have to be used. Moreover, the results have to be interpreted in the context of the country's economic situation and industrial strategy. No uniform rules can be formulated with respect to the appropri- ate degree of protection for individual projects. This is because, among other things, the stage of industrial development of the coun- try, the average level of protection in the economy, and the "matur- ity" of a particular industry have an important bearing on this matter. This is not to suggest that the problem of economic appraisal of projects is an insoluble one, but merely to underscore the fact that a number of elements at the country, sector and project levels must be taken into consideration in reaching a final judgment on individual 1 1h1 projects. If approached in this way, existing techniques of project appraisal can give perfectly adequate guidance for sensible invest- ment decisions. Non-economic objectives Another important problem in project appraisal is to take account of objectives (e.g., employment, the development of backward re- gions, income redistribution and environment) which are not directly related to growth criteria but which might have particular importance in individual country situations. These objectives have to be treated primarily in the context of preparing an industrialization strategy for the country. They can also be considered in the context of sector surveys, and in project selec- tion and design. Attention might be given to the possibility of new types of projects which have as a major objective the creation of direct employment, or similar social goals. At the project appraisal level, the possibility of furthering such goals is more limited. The best methodological treatment of these non-economic objectives in project appraisal is an important area for further research within the Bank Group. A process of approximations In summary, economic project evaluation is a process of successive approximations, and it varies with the size and importance of the project. For DFC sub-projects particularly, the introduction of even relatively simple approaches to economic project evaluation can only be a step-by-step process entailing a great deal of effort to ensure progressive improvement. In the Bank Group's project appraisal work, primary reliance should continue to be placed, in general, on the assessment of the economic rate of return as a measure of the overall economic bene- fits of industrial projects. In this connection, continuing attention is necessary to test the new methodologies for estimating the economic rate of return and to refine existing techniques, particularly those aimed at assessing non-economic costs and benefits. There is scope for improving follow-up of existing projects, par- ticularly in the area of economic assessment. IFC has established a separate project supervision unit; the Bank's appraisal reports on development finance companies review the performance of Bank financed institutions, and the Operations Evaluation Division has been organized to undertake economic appraisals of past Bank and IDA investments. 117 The Role of the Bank Group The discussion of industrialization in the first section of this paper indicated that industry in the developing countries has changed radically since World War IL. Industrialization has become the key element of development policy in many of the more advanced de- veloping countries, and the Bank Group should bring its own policies and operations into line with the new requirements and potential for industrial development. A more systematic approach to industrial work, based on well- defined goals and on more intensive sector studies and policy work, must form the basis of a more effective, integrated program of action to cope with the new challenges and opportunities in industrial development. The following appear to be the major issues which deserve consid- eration in the formulation of such a program: * Should there be a concentration of financial support for industry in selected "semi-industrialized" and "industrializing" countries? What contribution is feasible in the "non-industrial" countries? * What can be done to secure improvements in the general policy framework for industrial development in developing countries? * What can the Bank Group do to encourage developing countries to achieve a more outward industrial orientation? * What type of operational orientation is suggested by the small share of the Bank Group in total industrial investment in develop- ing countries? * How can the Bank Group achieve more effective collaboration and division of labor with other international and regional bodies in supporting industrialization? These issues are interrelated; their resolution will affect the overall volume and pattern of future Bank Group activity in the manufactur- ing sector. Need for Concentration of Financing Effort In view of the serious constraints in the immediate future on the scope and size of any Bank Group program to promote industrializa- tion, there is a strong case for concentrating financial support in the manufacturing sector on selected "semi-industrialized" and "indus- trializing" countries where the infrastructural base has been suffi- ciently developed and the manufacturing sector is large enough to have created significant external economies. This will permit the Bank Group to deploy its resources with maxi- mum effectiveness. Concentration does not mean simply allocating 118 a large share of financing for manufacturing to these countries; it means an organized effort based on detailed country and sector eco- nomic studies to assist them in their further industrialization. Assistance to "Non-industrial" Countries This does not imply, however, that the Bank Group will neglect the "non-industrial" countries. On the contrary, a much greater effort will be made to meet their special needs in the manufacturing sector. They are concentrated chiefly in Africa, and many have relatively small internal markets. Frequently, the biggest obstacle to their indus- trial development is a critical shortage of relevant skills and entre- preneurship. The Bank Group's approach will be along two main lines. First, assistance will be provided in selected cases to develop and strengthen specialized local institutions concerned with industry, including DFCs. This will frequently involve the establishment of new types of institutions, oriented toward small and medium sized enter- prises and actively engaged in technical assistance, including ex- tension and promotion work. Where the prospective local volume of business falls short of the critical minimum required to sustain such entities, the possibility of providing some support through insti- tutions established on a regional basis will be explored'. Second, the Bank Group expects to strengthen its activities in cer- tain areas, such as small scale industry, support for extension services, and technical advice on sector priorities and project identification and promotion, which are particularly suited to the needs of the "non-industrial" countries. An effort in these areas will first be of a pilot nature and will be undertaken in close collaboration with other international and regional agencies. Improvements in Policy Framework The developmental contribution of Bank Group industrial opera- tions depends critically upon the policy framework in recipient coun- tries. Improvements in industrial policy would facilitate the indus- trialization process and at the same time enhance the effectiveness of Bank Group financial assistance. To secure improvements in industrial policies, the Bank Group might link its operations in the manufacturing sector more closely to desired policy changes and improvements. It could, for example, refuse to finance industrial investments in countries where the indus- 'Projected FY1972-76 operations include a program for expanding local finance com- panies in Africa and channeling financial assistance to regional banks, such as the Central American Bank for Economic Integration, but there is scope for further ex- pansion of these activities. 119 trial sector benefits from undue protection, or decline support of development finance companies that do not make an adequate effort to raise local resources. Stated more positively, it could indicate its willingness to participate in investment programs and projects which are associated with improvements in sector policies and practices. The mix of instruments to be employed for supporting industriali- zation could also be related more closely to the policy frame in dif- ferent countries. In situations where direct financing promises to be a more effective instrument for advancing desirable policy reforms, the Bank GroLIp could rely on this method rather than financing through intermediaries. In some situations, policy reforms could be facilitated through non-project assistance. There is much to be said for linking operations in this sector more closely to policy reforms than in the past, as the Bank Group normally does in other sectors. Such a link in this case, however, involves the giving of advice on difficult and sensitive questions to which un- equivocal answers are seldom possible. While in some respects the direction of desirable changes may seem clear, it is often difficult to arrive at policy prescriptions in the industrial sector with any degree of assurance. Furthermore, such a link between finance and policy reform re- quires a thorough knowledge and understanding of the industrial priorities and requirements of member countries. Such knowledge at present is limited to only a few countries. Advice based on an incom- plete grasp of the facts or on a less than sympathetic appreciation of the real problems can often be counterproductive. As the Bank Group's knowledge of the sector improves, it may become increas- ingly possible to mesh operations in industry with policy goals. The Bank Group also should make every effort to assist developing countries in defining appropriate patterns of industrial development suited to their resource endowments and to other factors determining their comparative advantage in industrial production and trade. This is not an easy task. The success or failure of individual countries, especially those in the "non-industrial" category, to develop particular lines of industrial activity, might depend less upon definable measures of comparative advantage than upon the availability in particular situations of the entrepreneurial, management, and technical skills required to launch a successful industrial venture. These might best be cultivated by the maintenance of a general policy regime which will foster investment activities as well as by specific measures for the promotion of techni- cal and management skills. The Bank Group can, however, carry out a comprehensive program 120 of sector studies and preinvestment work, taking advantage of its ability to view a country's problems in a broad comparative frame- work. Such studies will also facilitate better identification of the areas and projects where, through direct and indirect financing operations, the Bank Group can more effectively support a country's industriali- zation plans. An Outward Orientation for Manufacturing Import substitution has been the key characteristic of Bank Group operations in manufacturing, just as it has been the dominant indus- trial strategy in the developing world. An issue confronting the Bank Group now is whether it can make a contribution towards promoting an outward orientation of industry in the developing countries-in other words, whether it can help them achieve a more efficient and internationally competitive manufacturing sector. A strengthened program of advice and assistance in policy formulation, plus greater insistence on the efficiency of industrial production through more rigorous project appraisal, should help in meeting this objective. The Bank Group can also play a more active role in export promo- tion. As already seen, a critical problem confronting developing coun- tries in the 1970s is how to increase their manufactured exports in order to secure the foreign exchange they need for future develop- ment. In the light of this, the Bank Group will make a special effort in its own operations to provide support for export-oriented projects. This cdoes not mean, of course, that the Bank Group will not con- tinue to finance projects that result in efficient import substitution. Its willingness to support export-oriented industries will be rein- forced, however, by a much more determined and systematic effort. This will be supported through appropriate sector and preinvestment work to identify concrete possibilities for export-oriented industries in individual countries, and to actively promote such investments. It will take time for such efforts to show results, but they should help bring Bank Group operations in industry more closely in line with the industrial priorities of the more advanced developing countries. A Promotional Role in Industry The very small share of Bank Group assistance in industrial invest- merit in developing countries highlights the need for the Group to play a strongly promotional role, so that the developmental impact of its contribution can be maximized. There are at least four major areas in which a much more organized promotional effort by the Bank Group can make a real contribution to the industrial develop- ment of its members. 121 Promotion of "growth poles" The Bank Group will actively promote the creation of "growth poles" for industry, particularly in those developing countries where industrial growth needs new stimulus and direction. Such activity in- volves complex intersectoral organization and may require applica- tion of sophisticated analytical techniques. Large capital outlays may be required in constructing a set of interrelated infrastructure and industrial investments. Projects of this nature are now in an explora- tory phase in Iran and Colombia. There may well be other countries in which similar needs exist. Regional development Another proposal, touched upon in an earlier part of this paper, is for the active promotion of industrial investment and trade on a regional (multinational) basis. The present status of experiments in regional cooperation is disquieting. It is hardly fashionable to speak affirmatively about the potential for multinational cooperation in the developing world. Even so, a clear lead by the Bank Group can serve a useful purpose. The present program includes proposals to finance industrial devel- opment on a regional basis through a regional development bank, as in Central America and East Africa. The Bank Group will put its weight behind economic integration efforts by supporting policies, programs and projects which further regional cooperation. Aside from these regional groupings, the Bank Group will assist in the identification and promotion of investment projects involving more than one country which are likely to contribute to the estab- lishment of viable industries and increase trade among developing countries. A number of such possibilities have been identified by the staff in the course of their normal country work. Promotion of such projects in selected cases could yield great benefits to the countries concerned in terms of increased product specialization, economies of scale and exports of manufactures. Small scale industry Another promising area for promotional effort is in small scale in- dustry. The development of such industry is an important objective, which should occupy a prominent position in the strategy of several countries. Small scale industry also has a special contribution to make in helping achieve important social objectives, such as employment, income distribution and development of backward regions. The need in this area is for indigenous organizations which have the facilities to identify potential entrepreneurs, select suitable activi- 122 ties in which small scale production can be viable, prepare feasibility studies, provide technical and managerial guidance, and secure ac- cess to finance on appropriate terms. The Bank Group has so far provided direct support for small scale industry only in Pakistan. It will be essential to develop a small expert group which concerns itself solely with the problem of small scale industry development, since the requirements are very different from those of large scale industrial projects. The general approach will be to collaborate with specialized local institutions dealing with small scale industry, or to help build such institutions where they do not exist, and to channel the necessary technical and financial assistance primarily through them. A program in the field will be developed in close collaboration with the regional development banks and other specialized institutions. Improving technology There is also a strong case for greater support by the Bank Group for improvement of technology in the developing countries. In the past, the Bank Group has been more concerned with the transfer of existing technology. Its adaptation to local conditions is equally im- portant. The most effective contribution in this area could be made by en- couraging countries to adopt policies which result in the emergence of realistic factor prices in the economy and make "intermediate" technology more profitable for entrepreneurs. Assuming progress on this front, specialized domestic institutions could assist in identifying, developing and disseminating appropriate technology. The Bank Group will explore the desirability of providing assistance towvard the establishment or strengthening of institutes working in the area of industrial technology, productivity, design and standards. There are a number of successful institutions, such as those in the Republics of China and Korea, which demonstrate the extraordinary productivity such institutions can attain if properly organized, staffed and financed. Although the need for assistance in this field is acknowledged in certain country programs, there is no provision in FY1972-76 opera- tions for any specific projects. To prepare an effective program, it will be necessary, as in the case of small scale industries, to develop in- house expertise and to work in the closest cooperation with other international institutions, such as the United Nations Industrial De- velopment Organization (UNIDO) and the International Labor Or- ganization (ILO). Both the Pearson Commission and a Committee of the Economic and Social Council have urged a more active Bank Group role in this area. Cooperation with Other Agencies In strengthening its industrial operations, a major concern of the Bank Group must be to ensure that it does not assume responsibilities and burdens which should appropriately be carried by or shared with others. At present, the Bank Group cooperates informally with a number of U.N. agencies in the industrial field. UNIDO is the princi- pal U.N. agency concerned with industrial development. The Bank's Industrial Projects and DFCs Departments and IFC have all estab- lished informal operating procedures for cooperation with UNIDO. While the framework for cooperation exists, instances of meaning- ful and significant collaboration in industry work between the Bank Group and other agencies concerned have been rare. The difficulties arise partly from the multiplicity of institutions interested in this area, partly from the disinclination of each agency to surrender the lead on projects it has under consideration, and partly from the severe financial and technical constraints under which some of the agencies have to operate. Nevertheless, the range of problems and requirements in the indus- trial field is so great that there is obvious justification for a pooling of resources and efforts to the extent permitted by the technical and operational capabilities of each institution. Each agency in this field, including the Bank Group, has its strengths, which can best be utilized if applied in concert with others in accordance with an agreed divi- sion of labor between the interested institutions. There is a strong case for the Bank Group to take the lead in dis- cussing with UNIDO, ILO and the other agencies concerned the basis for collaboration, particularly in regard to activity in the "non- industrial" countries, the development of small scale industry and the promotion of improvements in technology. Other Issues There are some other unresolved issues which have an important bearing on Bank Group operations in the manufacturing sector. These have not been treated in this paper, partly because they have been discussed in the past at other levels and partly because they raise complex institutional and procedural issues which justify their sepa- rate treatment. Bank and IDA policy regarding preference for domestic suppliers is one such issue. In a situation where a number of developing coun- tries are beginning to produce capital goods, and should be en- couraged to expand such production, the policy on preference for domestic suppliers has important implications from the standpoint of the countries' most appropriate industrial strategy and the effective- ness of the Bank Group's role in industrialization. I')A Conclusion In summary, the 1970s present new challenges and opportunities to the Bank Group to assist its member countries in industrial devel- opment. The Group is just beginning to develop an integrated pro- gram in this field. Industrialization is a complex process, difficult to schedule, but it is a key element in the development strategy of many of the more advanced developing countries. The most effective con- tribution the Bank Group could make would be to assist its member countries in pursuing sound industrialization policies. The Bank Group's role should be selective, concentrating its finan- cial assistance on "semi-industrialized" and "industrializing" coun- tries in the context of a continuing policy-oriented dialogue regarding industrialization in these countries. It should concentrate on infra- structure and agricultural investment in "non-industrial" countries, but should develop the capability, in a cooperative division of labor with other agencies, to make a much greater contribution towards meeting their specialized requirements in the industrial field. The Bank Group's approach should be promotional and innovative, searching for new ways to facilitate the industrial progress of its mem- bers, such as investments in "growth poles" and in regional projects. It is believed that the directions outlined in this paper provide a useful guide for the future, though they will be subject to much test- ing as additional experience is gained. In moving along these lines, consideration will be given to necessary adjustments in existing or- ganizational arrangements, instruments and procedures, as well as manpower budgets, in order to enhance the effectiveness of Bank Group operations. Annex 1 Some Indicators of Industrial Development Page 1 Gross Value Added Average Annual Percentage Share in Manufacturing Growth c %1 of Average Annual of Eoports ot Non- as a Percent of Real Gross Growth %) of Resource-Based Value Added in Value Added Real GNP Manufactures') Commodity Production in Manufacturing per Capita in Commodity 1968 1960-1969 1960-1969 Exports 1968 Eastern Africa Non-industrial Botswana 17.1 10,5, Burundi . 0.0,5, Ethiopia 13.6 9.8 2.3 2.0 Lesotho 0.9121 0 Malagasy - 6.8(4, 0.0 Malawi 18.8 8.7 1.0 2.1 Rwanda 9.2?2) 0.8 1.0 Somalia 1.5(5 3.1 Sudan 11.1 11.1 0.6 0.07 Tanzania 10.8 16.4 1.6 6.0 Uganda 8.4 1.7 1.7 0.7 Zambia 19.0 19.8 5.4 1.0 Industrializing Zaire 38.6 12.7 0.2 4.8(6 Kenya 23.2 6.8 1.5 16.6 Mauritius 35.6 1.9 0.4<5 0.0 Swaziland 20.63°10, - 3.2 Semi-industrialized Rhodesia 42.0 5.5 0.4 31.6(6 South Africa 49.2 715(4) 3.7012 42.3 Western Africa Non-industrial Central African Republic 13.5(2) 0.0 0.4 Chad - 1.3 2.8 Congo, People's Republic of - 2.3 15.2(7, Dahomey 5.2(2. 0.0 0.9 8.312, Gabon 9.0.21 0.6 7.1 Gambia 0.7'5 0.0 Guinea 3.912? 2.6 Ivory Coast 19.1 11.2 4.7 12.3 Liberia 8.1 1.3 0.1 Mali 13.2 10.0 1.2 2.2'31 Mauritania 4.9 - 4.6152 2.502 Niger 8.5(2 -0.9 1 .1'3 Sierra Leone 9.7 1.2 0.02 Togo 15.32 14.4 0.0 3.5(6) Upper Volta 7312) 17.2 0.1 4.1 Industrializing Cameroon 21.7(3) 12.7 2.0 3.112. Ghana 23.5(2) 3.7(4) 0.0 9.7 Nigeria(",o 19.5(21 10.6 0.3 1.3 Senegal 29.0 4.3 -0.1 15.3 East Asia & Pacific Non-Industrial British Solomon Islands 1.9 0.0 Fiji 2.715' 8.8 Indonesia 13.7 1.2 0.8 4.5 Papua & New Guinea 5.0 9.9 2.0 4.2 Industrializing Korea, Republic of 33.8 14.2 6.4 74.3 Malaysia 24.2(81 12.0'4' 3.8 9.7 Annex 1 Page 2 Gross Value Added Average Annual Percentage Share In Manufacturing Growth M%i of Average Anirual of Exports of Non- as a Percent of Real Gross Growth { % of Resource-Based Value Added in Value Added Real GNP Manufactures''' Commodity Production In Manufacturing per Capita in Commodity 1968 1960-1969 1960-1969 Exports 1968 Philippines 32.0 4.3 1.9 13.4 Thailand 27.9 8.1 4.7 8 3 Semi-industrialized China 45.9 13.5 6.3 77.0 New Zealand 55.703' 4.7'4' 2.0 8.9 Industrialized Hong Kong 87.5 14.3 8.7 91.8 Singapore 65.9 19.0 4.5 11.6 South Asia Non-industrial Afghanistan 16.5 4.30' 0.3 0.5(30 Nepal 13.4,2, 1.1 0.4 Industrializing Burma 20.3.3 3.503 1.8 1.1 Ceylon 21.4 14.8'4, 2.1 1.1 Iran 3790'8, 11.014) 4.9 3.2(3, Pakistan,'4 19.2 7.6 2.9 50.5 Semi-I ndustrialized Indial'5' 20.7 2.5'4s 1.1 51.7 Europe, Middle East and North Africa Non-industrial Algeria 17.0'3s 16.3 -2.8 7.8 Iraq 13.9 5.9 3.0 0.8 Industrializing Cyprus 27.7 7.2 5.8 12.7 Jordan 21.1'2" 4.7'5' 13.3(3' Morocco 23.4 2.7 3.4 16.2130 Syria 20.0(2''9' 4.7 13.2 Tunisia 39.4(2) 6.2 2.1 23.0 Turkey 30.9 8.6 3.4 3.9 Semi-industrialized Egypt 41.7'-' 1.2 27.0 Finland 58.4 3.7(4) 3.9 67.0 Greece 40.0 6.2 6.2 16.5 Ireland 40.0k2,09 3.5 41.2 Israel,7 61.5 5.3 73.2 Lebanon 48.7'3'1 ' 2.1 25.2 Malta 52.7 6.4 4.1 76.8 Portugal,'7" 61.6 7.9 4.9 71.7 Spain 56.8 7.4 6.5 57.7 Yugoslavia 58.1 13' 4.6 69.4 Central America and Caribbean Non-I ndustrial Haiti 18.8'2' -1.0 Industrializing Dominican Republic 38.4 3.6 0.4 3.0,3) El Salvador 38.7 9.0 1.9 34.2 Guatemala 35.4 7.9 1.9 20.6'3' Guyana 20.9 5.0 0.7 1.2 Annex 1 Some Indicators of Industrial Development Page 3 Gross Value Added Average Annual Percentage Share in Manufacturing Growth (%) of Average Annual of Exports of Non- as a Percent of Real Gross Growth (%) of Resource-Based Value Added in Value Added Real GNP Manufactures(') Commodity Production in Manufacturing per Capita in Commodity 1968 1960-1969 1960-1969 Exports 1968 Honduras 26.7 7.7(4) 1.1 8.3 Jamaica 35.6 4.4 3.0 10.3 Nicaragua 31.0 12.0 3.5 9.5 Panama 37.2 10.7 4.8 2.0 Trinidad and Tobago 30.90"1 10.1 3.8 12.8(3) Venezuela 29.7 7.0 2.5 1.2 Semi-Industrialized Costa Rica 41.0 2.9 19.0 Mexico1'7' 61.2(0 10.0 3.5 19.4 South America Industrializing Bolivia 29.1 6.4 2.4 0.0 Colombia 32.7 4.5 1.5 9.6 Ecuador 32.3 5.4 1.2 4.6 Paraguay 32.6 4.0 1.0 0.0 Peru 33 9(2) 2.94) 1.4 0.0 Semi-industrialized Argentina,7" 61.8 4.3 2.6 21.8 Brazil 56.3 6.6 1.4 11.5 Chile 56.2 5.6 1.7 3.5 Uruguay 51.2(20 9 -0.8 5.0 Selected Industrialized Countries' 1 Australia 67.0 6.0 2.9 21.0 Canada 65.5(30 43(4, 2.8 53.3 France 78.6 5.9'4 4.8 74.0 Germany, Federal Republic of 83.7 4.374) 3.7 89.0 Japan 73.6 11.6'4) 10.0 95.3 Netherlands 73.4(2) 3.0'4, 3.1 62.3 Norway 69.3 4.0'4' 4.0 58.2 Sweden 76.8 4.64'4 3.4 72.7 United Kingdom 80.2 2.0'(4 1.8 82.7 United States 80.5 4.6's4 3.2 68.9 NOTES: The allocation of countries to areas for purposes of this table corresponds to assignments of the Bank's area departments. 2) Exports of non-resource-based manufactures exclude petroleum and related products, and unworked non-ferrous metals. This column includes goods classified under SITC codes 57 and 8;6 except part of 68:and parts of 0-4. For non-ferrousmetals(code68) and food products(code0), thedefinition of manutactures suggested in UANCTAD. Trade in Manufactures of Developing Countries: 1970 Review has been followed. except for refined sugar (code 061.2) which has been included. ) 1966. )3 1967. (4 1968. (') Estimates of GN P per capita and growth rate are tentative. 06) 1965. (7) Includes industrial diamonds. (8) 1969. 0) Includes mining. 0n Commodity production includes construction. (i) Excludes oil refining. (27) 1961-68. (i1) Includes petroieum but no other mining. ") Exception made due to the large size of domestic market. ) Exception made duets t he large size of domestic market, structure of production, and engineering and managerial skills. 06 Countries in this group with per capita income of $1,300 or niore in 1969, and with gross value added in manu- facturtng as a percent ot value added in commodity production of more than 60% in a recent year, are considered "industrialized". 7) Exception treated as "semi-industrialized" to take into account the effect of protection in overstating value added in manufacturing. Sources: U.N. Yearbook of National Accounts Statistics, 1969. Vol. 1; U.N. Yearbook of lnternational Trade Statistics, 1968: World Bank. Annex 2 Characteristics of Bank Group Manufacturing Projects by Organizational Unit, FY 1967-1971 Organizational Unit Project Characteristics World Bank and IDA I FC Through DFCs(') Direct(') Average of Total Project Cost ($ millions) 29.0 2.0 14.9 Average Commitment ($ millions) 5.2 0.3 5.5 Proportion of Protects Proportion of Commitments Bank-IDA Bank-IDA IFC DFCs0i Directi2) IFC DFCs.1) Directz Total Project Cost Under $200,000 0% 21% 0% 0% 6% 0% Under $1 million 0% 61% 4% 0% 19% 0% Under $15 million 51% 98% 84% 21% 90% 28% Size of Commitmentt3 Under $50,000 0% 15% 0% n.a. n.a. n.a. Under $400,000 2% 67% 13% n.a. n.a. n.a. Under $5 million 63% 100% 79% n.a. n.a. n.a. Proportion of Project Output Exported Zero 73% 54% 21% 74% 50% 66% Less than 30% 86% 77% 59% 90% 75% 90% Less than 50% 89% 85% 76% 93% 83% 94% Less than 100% 93% 97% 100% 94% 94% 99% Number of Employees in Enterprise", Under 20 0% 9% 0% 0% 4% 0% Under 100 2% 39% 0% 3% 21% 0% Under 500 34% 79% 4% 29% 57% 0% Proportion of Government Ownership Zero 81% 97% 92% 77% 93% 75% Under 25% 93% 99% 92% 94% 98% 75% Under 50% 100% 100% 92% 100% 100% 75% Proportion Involving Foreign Equity13 New projects 85%5' 23% 100% n.a. n.a. n.a. Expansion projects 64%'5' 15% 9% n.a. n.a. n.a. All projects 76% 5' 16% 13% n.a. n.a. n.a. .1) Figures relate to a representative 1970 sample of Bank Group financed DFC sub-projects. n2) Excluding import credits and other non-project lending. 13) Symbol n.a. as used opposite this heading means not readily availabie. .° I ncludes employees in the enterprise before the project as well as those added . Involving technical as well as tinancial participation by foreign partners Bank Group Manufacturing Operations by Sector, FY 1967-71 (US$ millions) Indirect I.S.I.C. Industry Category(°' Total Direct Financing Bank-IDA Direct Financing of of DFC Total IFC(2) Financing(') Projects Sub-Projects(-) Bank Group Description Amount % Amount % Amount % Amount % Amount % Food manufacturing 11.1 3.8 - 11.1 2.6 65.4 11.5 76.5 7.7 Textile manufacturing 19.5 6.8 2.0 1.5 21.5 5.1 184.1 32.3 205.6 20.7 Leather and products of leather - - - - - 11.2 2.0 11.2 1.1 Furniture and wood products - - 2.2 1.7 2.2 0.5 19.9 3.5 22.1 2.2 Paper and pulp 36.3 12.6 2.0 1.5 38.3 9.1 19.8 3.5 58.1 5.9 Printing, publishing and allied industry 7.0 2.4 - - 7.0 1.7 6.0 1.1 13.0 1.3 Chemicals manufacturing 74.4 25.8 62.0 47.0 136.4 32.4 46.2 8.1 182.6 18.4 Petroleum refineries 8.0 2.8 - - 8.0 1.9 3.3 0.6 11.3 1.1 Manufacturing of miscellaneous products of petroleum and coal 22.8 7.9 - 22.8 5.4 - 22.8 2.3 Manufacture of rubber products - -.- -_ 26.4 4.6 26.4 2.7 Manufacture of plastic products - - 1.5 1.1 1.5 0.4 4.0 0.7 5.5 0.6 Pottery, china, earthware 2.3 0.8 - 2.3 0.6 3.3 0.6 5.6 0.6 Glass and glass products 11.6 4.0 11.6 2.8 13.9 2.4 25.5 2.6 Other non-metallic mineral products 47.1 16.3 - 47.1 11.2 35.0 6.1 82.1 8.3 Iron and steel basic industries 11.3 3.9 8.2 6.2 19.5 4.6 30.4 5.3 49.9 5.0 Non-ferrous metal basic industries 8.6 3.0 27.0 20.5 35.6 8.5 4.7 0.8 40.3 4.1 Fabricated metal products 11.5 4.0 - - 11.5 2.7 23.1 4.0 34.6 3.5 Machinery, except for electrical -- - - -_ 28.4 5.0 28.4 2.9 Electrical machinery manufacture 0.2 0.0 0.2 0.0 19.1 3.3 19.3 1.9 Transport equipment manufacture 17.0 5.9 20.3 15.4 37.3 8.9 23.1 4.0 60.4 6.1 Other manufacturing industries - -. 6.8 5.1 6.8 1.6 3.3 0.6 10.1 1.0 Totals 288.7 100% 132.0 100% 420.7 100% 570.6 100% 991.3 100% Of which: Heavy manufacturing 87.0% 90.6% 88.1% 43.7% 62.7% Light manufacturing 13.0% 9.4% 11.9% 56.3% 37.3% ' i') U.N. 1968 International Standard Industrial Classification. D I FC data are qualified as in footnote (i, Annex 8, except that investments in service industries are exciuded. w Bank and IDA direct financing data are qualified as in footnote in, Annex 9 41 DFC figures in this table (apart from the total amount of sub-proiect investments) are not actuals, but are based on percentage distributions obtained from a representative sample of sub projects in fiscal years 1968-70. Agriculture, mining, tourism and services are excluded. Annex 4 Pattern of Bank Group Manufacturing Operations in Selected Countries by Organizational Unit, FY 1967-1971 (Percentages) Percentage of Bank Group Financing Percent of Total for Manufacturing Projects Percent Shares of Bank Group Finan- Bank Group Devel- cing of Manufac- Bank-IDA oping Members turing Projects in Developing Through Total Countries(°) Stage(°) Population GNP Countries DFCs(') Direct IFC Bank Group India Si 31.3 16.8 7.3 67.3 - 32.7 100.0 Brazil Si 5.3 7.1 6.6 28.9 25.4 45.7 100.0 Mexico Si 2.8 8.1 0.9 - - 100.0 100.0 Iran 1 1.6 2.7 11.2 94.3 - 5.7 100.0 Pakistan 1 7.4 3.9 10.6 68.3 25.2 6.5 100.0 Turkey 1 2.0 3.3 8.9 77.9 - 22.1 100.0 Colombia 1 1.2 2.0 6.2 96.3 - 3.7 100.0 Philippines 1 2.1 2.0 5.9 64.3 - 35.7 100.0 Morocco 1 0.9 0.8 5.7 100.0 - - 100.0 Korea, Rep. of 1 1.7 1.7 4.8 89.4 - 10.6 100.0 Nigeria 1 3.7 1.4 1.3 95.2 - 4.8 100.0 Kenya 1 0.6 0.4 1.1 - - 100.0 100.0 CACM(4) I 0.8 1.5 0.2 -- - 100.0 100.0 Ghana 1 0.5 0.5 - - - - Indonesia NI 6.7 3.6 4.0 - 57.6 42.4 100.0 Ethiopia NI 1.4 0.6 0.7 - - 100.0 100.0 Tanzania NI 0.7 0.3 - - - _ _ Uganda NI 0.6 0.3 - Sub-total 71.3 57.0 75.5 69.6 8.8 21.6 100.0 Other Developing Countries 28.7 43.0 24.5 57.6 14.0 28.4 100.0 Total Bank Group Developing Members 100.0 100.0 100.0 66.7 10.0 23.2 100.0 (") Countries selected here are the 22 reviewed in detail for the purposes of this paper. (2) NI, non-industrial; Si. semi-industrialized; 1, industrializing. (') Includes IFC investments through financial intermediaries. (') Costa Rica, one of the five members of the Central American Common Market (CACM), is categorized as a semi- industrialized' county n Annex 1, but is here combined with the other CACM countries for purposes of presentation. Sources: Financing data extracted from Annexes 6, 7, 8 and 9, but developed countries (Finland) have been excluded to arrive at percentages in this table. GN1P and population percentages, World Bank. 131 Annex 5 Bank Group Manufacturing Operations() Page 1 by Organizational Unit and Major Recipient, FY 1967-1971 (USS millions Bank-IDA Through Total Stage 2) DFCs()3 Direct IFC Bank Group $ % $ % $ %~ $ % Eastern Africa *East African Community - 15 5 15 1 *Ethiopia NI - 10 3 10 1 Other 6 1 - -- 6 0 Sub-totals 6 1 24 8 31 2 Western Africa *Nigeria I 16 2 1 0 17 1 Other …-- 3 1 3 0 Sub-totals 16 2 - 4 1 20 1 Africa Unallocated 1 0 1 0 East Asia and Pacific China, Republic of Si 33 4 - - 10 3 42 3 *Indonesia NI - - 30 23 22 7 52 4 *Korea, Republic of 1 56 6 -- - 7 2 63 5 *Philippines I 50 5 28 9 78 6 Thailand I - - 22 7 22 2 Other 5 1 - 5 2 10 0 Sub-totals 144 16 30 23 94 31 268 20 South Asia *India Si 65 7 -- - 32 10 97 7 *Iran I 140 16 - - 8 3 148 11 *Pakistan I 95 10 35 27 9 3 139 10 Other 5 1 - - 3 1 8 1 Sub-totals 305 33 35 27 52 17 392 29 Europe, Middle East and North Africa Finland Si 42 5 - - -- - 42 3 Greece Si 33 4 - 9 3 41 3 Israel Si 40 4 - - 40 3 *Morocco I 75 8 - - - 75 6 Tunisia 1 21 2 - - -- 21 2 *Turkey I 91 10 26 8 117 9 Yugoslavia Si 2 0 45 34 17 6 64 5 Other 10 - --- 2 1 12 1 Sub-totals 313 34 45 34 54 18 412 30 132 Annex 5 Page 2 Bank-IDA Through Total Stage(2' DFCs(3) Direct IFC Bank Group $ % $ % $ % $ % Central America and Caribbean nCACM(4) 1 _ _- 2 1 2 0 *Mexico Si - - - 12 4 12 1 Venezuela 1 12 1 - - 4 1 17 1 Sub-totals 12 1 - - 19 6 31 2 South America Argentina Si - - - - 15 5 16 1 *Brazil Si 25 3 22 17 40 13 87 6 *Colombia I 79 9 - - 3 1 81 6 Other 8 1 - - - - 8 1 Sub-totals 112 12 22 17 58 19 192 14 Latin America Unallocated 10 1 - - - - 10 1 Grand totals 918 100 132 100 305 100 1356 100 Of which Industrialized 5 1 - - - - 5 0 Semi-Industrialized 249 27 67 51 137 45 453 33 Industrializing 654 71 35 26 137 45 826 61 Non-Industrial - - 30 23 32 10 62 5 Unallocated 10 1 - - - - 10 1 NOTES: The allocation of countries to areas for purposes of this table corresponds to assignments of the Bank's area departments. Totals may not add due to rounding. Indicates countries included in the sample of 22 countries reviewed in detail for the purposes of this paper. Ghana was also a sample country but received no Bank Group funds for industry in this period. ') IFCand Bank-IDAdirectlendingdataexcludemining and tourism projects and commitments subsequently cancelled. DFC data relate to Bank Group commitments to industrial DFCs though some of their sub-projects are in sectors other than manufacturing. DFCs for tourism ventures are excluded. (2) NI, non-industrial; 1, industrializing; SI, semi-industrialized. 3) Includes IFC investments through financial intermediaries. I4) Central American Common Market. Sources: Annexes 6, 7, 8 and 9. 133 Annex 6 Bank Group Manufacturing Commitments(1) Page 1 by Regions and Countries, FY 1967-1971 (USS millions) Total Stage(2) 1967 1968 1969 1970 1971 1967-71 % Eastern Africa Ethiopia NI - 9.0 - 0.6 - 9.6 1 Kenya I - - - 14.7 - 14.7 1 Zaire I - - - 5.8 - 5.8 0 Sub-totals - 9.0 - 21.1 - 30.1 2 Western Africa Nigeria 1 0.8 - 6.0 -- 10.0 16.8 1 Senegal 1 3.5 - - --3.5 0 Sub-totals 4.3 - 6.0 10.0 20.3 1 Africa Regional - - - 0.5 0.5 0 East Asia and Pacific China, Republic of Si - 14.6 25.2 2.6 42.4 3 Indonesia NI - - 30.0 22.1 52.1 4 Korea, Republic of I - 5.7 21.7 5.0 30.7 63.1 5 Malaysia I - 3.5 0.2 1.5 - 5.2 0 Philippines 1 37.0 - - 32.8 8.0 77.8 6 Singapore Ind. - - -- 5.0 - 5.0 0 Thailand I - - 22.1 - 0.2 22.3 2 Sub-totals 37.0 23.8 44.0 99.5 63.6 267.9 20 South Asia Ceylon I - 2.3 - 6.3 - 8.6 1 India Si 12.7 25.0 15.9 43.0 - 96.6 7 Iran 1 24.6 24.9 43.9 -- 54.5 147.9 11 Pakistan 1 5.2 35.0 75.9 23.0 - 139.1 10 Sub-totals 42.5 87.2 135.7 72.3 54.5 392.2 29 Europe, Middle East and North Africa Finland Si - - 22.0 -- 20.0 42.0 3 Greece Si - 12.5 - 28.6 - 41.1 3 Ireland Si - - -- 10.0 10.0 1 Israel Si - 15.0 - 25.0 - 40.0 3 Lebanon Si - - - 2.1 2.1 0 Morocco I - - 15.0 15.0 45.0 75.0 6 Spain Si 0.2 - - - - 0.2 0 Tunisia I - 10.0 - 10.6 - 20.6 2 Turkey 1 25.3 -- 27.3 14.8 49.1 116.5 9 Yugoslavia Si - 10.5 16.0 28.5 9.0 64.0 5 Sub-totals 25.5 48.0 80.3 122.5 135.2 411.5 30 134 Annex 6 Page 2 Total Stage(2) 1967 1968 1969 1970 1971 1967-71 % Central America and Caribbean CACM13' 1 2.1 -- 2.1 0 Mexico Si 0.4 - 12.0 12.4 1 Venezuela I 7.5 2.1 7.0 16.6 1 Sub-totals - 10.0 2.1 19.0 31.1 2 South America Argentina Si - 10.0 -- 5.5 15.5 1 Brazil Si 10.7 22.1 9.5 33.4 10.9 86.6 6 Colombia 1 0.9 12.7 26.0 1.9 40.0 81.5 6 Ecuador I 0.3 - 8.0 8.3 1 Peru I 0.01 0.1 -- 0.1 0 Sub-totals 11.6 34.9 45.8 35.3 64.4 192.0 14 Latin America Regional - 10.0 - 10.0 1 Grand Totals 120.9 212.9 313.9 360.7 347.2 1,355.6 100 Of which: Industrialized - 5.0 -- 5.0 0 Semi-Industrialized 23.6 100.1 73.4 183.7 72.1 452.9 33 Industrializing 97.3 103.8 240.5 131.4 252.5 825.5 61 Non-Industrial 9.0 - 30.6 22.1 61.7 5 Regional - 10.0 0.5 10.5 1 Of which: Bank-IDA Direct -- 32.5 48.0 51.5 - 132.0 10 Through DFCs'4) 74.9 165.0 194.6 224.5 259.4 918.4 68 IFC 46.0 15.4 71.3 84.7 87.8 305.2 22 NOTES: The allocation of countries to areas for purposes of this table corresponds to assignments of the Bank's area departments. Totals of percentages may not add due to rounding. 0) Data used in compiling this table are qualified as in Annexes 7, 8 and 9. 12) Ni, non-industrial; 1, industrializing; SI, semi-industrialized; Ind., Industrialized. 0) Central American Common Market. (4 Includes IFC investments through financial intermediaries. 135 Annex 7 Bank Group Commitments to Industrial DFCs') by Regions and Countries, FY 1967-1971 (USS millions) Total Stage(') 1967 1968 1969 1970 1971 1967-71 % Eastern Africa Zaire I - - - 5.8 - 5.8 1 Western Africa Nigeria 1 6.0 - 10.0 16.0 2 Africa Regional - - 0.5 0.5 0 East Asia and Pacific China, Republic of Si - 14.6 - 18.0 - 32.6 4 Korea, Republic of I - 5.7 20.0 - 30.7 56.4 6 Philippines 1 25.0 - - 25.0 - 50.0 5 Singapore Ind. - - -- 5.0 - 5.0 1 Thailand I - - - - 0.2 0.2 0 Sub-totals 25.0 20.3 20.0 48.0 30.9 144.2 16 South Asia Ceylon I - 2.3 - 3.0 - 5.3 1 India Si - 25.0 - 40.0 - 65.0 7 Iran 1 24.6 24.9 40.0 - 50.0 139.5 15 Pakistan I - 35.0 40.0 20.0 - 95.0 10 Sub-totals 24.6 87.2 80.0 63.0 50.0 304.8 33 Europe, Middle East and North Africa Finland Si - - 22.0 - 20.0 42.0 5 Greece Si 12.5 - 20.0 - 32.5 4 Ireland Si - - - 10.0 10.0 1 Israel Si 15.0 - 25.0 - 40.0 4 Morocco I - 15.0 15.0 45.0 75.0 8 Tunisia I 10.0 - 10.6 - 20.6 2 Turkey 1 24.3 - 25.4 - 40.0 90.7 10 Yugoslavia Si - - - 2.0 - 2.0 0 Sub-totals 25.3 37.5 62.4 72.6 115.0 312.8 34 Central America and Caribbean Venezuela I 7.5 - 5.0 12.5 1 South America Brazil Si - -- - 25.0 - 25.0 3 Colombia I - 12.5 25.9 0.1 40.0 78.5 9 Ecuador 1 0.3 - 8.0 8.3 1 Sub-totals - 12.5 26.2 25.1 48.0 111.8 12 Latin America Regional - 10.0 - 10.0 1 Grand Totals 74.9 165.0 194.6 224.5 259.4 918.4 100 Of which: Industrialized - - 5.0 - 5.0 1 Semi-Industrialized - 67.1 22.0 130.0 30.0 249.1 27 Industrializing 74.9 97.9 172.6 79.5 228.9 653.8 71 Regional - - - 10.0 0.5 10.5 1 NOTES: The allocation of countries to areas for purposes of this table corresponds to assignments of the Bank's area departments. Totals of percentages may not add due to rounding. (I) Data in this table relate to actual Bank Group commitments to industrial DFCs (net of cancellations),and include funds committed by the latter for sub-projects in sectors other than manufacturing (14% during FY1968-70). Also includes IFC investments through financial intermediaries. (2) 1, industrializing; Si, semi-industrialized; Ind., industrialized. Annex 8 IFC Manufacturing Commitments(') Page 1 by Regions and Countries, FY 1967-1971 (US$ millionsl Total Stage(2) 1967 1968 1969 1970 1971 1967-71 % Eastern Africa Ethiopia NI - 9.0 - 0.6 - 9.6 3 Kenya I -- - 14.7 - 14.7 5 Sub-totals - 9.0 - 15.3 - 24.3 8 Western Africa Nigeria 1 0.8 -- - - 0.8 0 Senegal 1 3.5 - - - 3.5 1 Sub-totals 4.3 - - - - 4.3 1 East Asia and Pacific China, Republic of Si - 7.2 2.6 9.8 3 Indonesia NI - - - 22.1 22.1 7 Korea, Republic of I - 1.7 5.0 - 6.7 2 Malaysia I - 3.5 0.2 1.5 - 5.2 2 Philippines 1 12.0 - - 7.8 8.0 27.8 9 Thailand I 22.1 - - 22.1 7 Sub-totals 12.0 3.5 24.0 21.5 32.7 93.7 31 South Asia Ceylon I - - - 3.3 - 3.3 1 India Si 12.7 -- 15.9 3.0 - 31.6 10 Iran I - - 3.9 - 4.5 8.4 3 Pakistan 1 5.2 - 3.9 - - 9.1 3 Sub-totals 17.9 - 23.7 6.3 4.5 52.4 17 Europe, Middle East and North Africa Greece Si - -- - 8.6 - 8.6 3 Lebanon Si - - - 2.1 2.1 1 Spain Si 0.2 - - 0.2 0 Turkey I - - 1.9 14.8 9.1 25.8 8 Yugoslavia Si -- - - 8.0 9.0 17.0 6 Sub-totals 0.2 1.9 31.4 20.2 53.7 18 Central America and Caribbean CACM'3' I - 2.1 - - 2.1 1 Mexico Si - 0.4 - - 12.0 12.4 4 Venezuela I - - 2.1 - 2.0 4.1 1 Sub-totals - 2.5 2.1 - 14.0 18.6 6 South America Argentina Si - - 10.0 - 5.5 15.5 5 Brazil Si 10.7 0.1 9.5 8.4 10.9 39.6 13 Colombia I 0.9 0.2 0.1 1.8 - 3.0 1 Peru I 0.01 0.1 - - - 0.1 0 Sub-totals 11.6 0.4 19.6 10.2 16.4 58.2 19 Grand Totals 46.0 15.4 71.3 84.7 87.8 305.2 100 Annex 8 IFC Manufacturing Commitments(') Page 2 by Regions and Countries, FY 1967-1971 (US$ millions) Total 1967 1968 1969 1970 1971 1967-71 % Of which: Semi-Industrialized 23.6 0.5 35.4 35.2 42.1 136.8 45 Industrializing 22.4 5.9 35.9 48.9 23.6 136.7 45 Non-Industrialized - 9.0 - 0.6 22.1 31.7 10 NOTES: The allocation of countries to areas for purposes of this table corresponds to assignments of the Bank's area departments. Totals of percentages may not add due to rounding. (') IFC data relate to original projectcommitments whether for new or expansion projects. Excluded are: tourism pro- jects and COFITOUR, a tourism financing institution in Tunisia; IFC investments in DFCs and a capital market venture in Korea ; mining and ore treatment projects (exclusively ore treatment and/or metals processing projects are included); underwriting commitments, except to the extent that an IFC investment has actually been required; investment commitments subsequently cancelled; and commitments for pre-feasibility studies only. Included are: agro-industry projects involving production, processing and markefing operations (sugar production and refining), and services (utilities). (2) NI, non-industrial; 1, industrializing; SI, semi-industrialized; Ind., industrialized. (') Central American Common Market. 138 Annex 9 Direct Bank-IDA Manufacturing Commitments(') by Regions and Countries, FY 1967-1971 (USS millions) Total Stage(2) 1967 1968 1969 1970 1971 1967-71 % Eastern Africa Western Africa East Asia and Pacific Indonesia NI - -- 30.0 - 30.0 23 South Asia Pakistan I - 32.0 3.0 - 35.0 27 Europe, Middle East and North Africa Yugoslavia Si - 10.5 16.0 18.5 - 45.0 34 Central America and Caribbean - - - - - South America Brazil Si - 22.0 - - - 22.0 17 Grand Total - 32.5 48.0 51.5 - 132.0 100 Of which: Semi-industrialized - 32.5 16.0 18.5 - 67.0 51 Industrializing - - 32.0 3.0 - 35.0 26 Non-industrial - - - 30.0 - 30.0 23 NOTES: The allocation of countries to areas for purposes of this table corresponds to assignments of the Bank's area departments. Totals of percentages may not add due to rounding. (') Excludes a $30 million loan to India for iron and steel, subsequently cancelled after disbursement of $1.7 million. Excludes also loans for mining projects. M) NI, non-industrial; 1, industrializing; SI, semi-industrialized. 139 TRANSPORTATION Characteristics of the Sector ................ ................... 143 Role of Transport in Economic Activity ........ .. ............ 143 Some Characteristics of Transport Systems ..... .............. 146 Trends in Transportation ............... ................... 149 Motorization ................... ....................... 149 Urban Congestion ................. .................... 149 Environmental Pollution .............. .................. 149 Railway Reform and Recovery .......... .. ............... 149 Technological Change ................ .................. 149 Transportation Planning ................ ................... 150 Past Activities in the Sector ................. ................... 151 Lending ................................................ 151 Physical Aspects of Transportation Projects, 1967-1971 ... ...... 155 Preinvestment Activities ............... ................... 155 Other Sources of Transport Finance ......... .. .............. 156 General Lessons Drawn from Past Operations ...... .......... 157 Project Re-evaluation and Monitoring ....... .. ............ 157 Improved Appraisal Tools ............ .. ................. 157 Analysis of Developmental Impacts ........ .. ............. 158 Institution Building and Policy Reform ......... ........... 159 Training ..................... ......................... 160 Projection of Future Operations ....... ......................... 161 Lending Program .................. ...................... 161 Sector Work .................... ........................ 161 Transport Policy Questions ............... ................. 162 Pricing and Financial Targets .......... .. ................ 162 Regulatory Systems ................ .................... 164 Directions in Future Lending ........... .. ................. 166 (a) Road Transport ......... ............................ 166 (b) Railways ........... ............................... 170 (c) Ports and Shipping ........ .......................... 172 (d) Aviation ........... ............................... 174 (e) Urban Transportation ....... ........................ 176 Research ...................... ......................... 178 141 Annexes 1. Basic Transport Data .......... ......................... 180 2. Bank and IDA Lending for Transport through FY1971 .... .... 184 3. Selected Statistics on Bank Group Projects FY1967-1971 Table 1-Highway .......... ......................... 190 Table 2-Railway .......... ........................... 191 Table 3-Port ............ ........................... 192 4. Summary of World Bank Transport Activities with Projection through FY1976 ...... ................... 193 5. Graph: Relation Between Motor Vehicle Density and GNP Per Capita, 1963-1967 ....... ................... 194 142 TRANSPORTATION SECTOR WORKING PAPER * This paper outlines the World Bank's lending, technical assistance and research in the transport sector against the background of pre- vious experience. It discusses the role of transport in overall eco- nomic activity, reviews what the Bank has done in support of the sector, and summarizes the general lessons derived from that ex- perience. It should be borne in mind that the Bank's approach to lending in this sector, as in all others, begins with an assessment of the country's economic position and prospects and of the relative priorities of both sectors and projects. References to the Bank include the International Development Association (IDA) except where the context requires otherwise. Where the term "Bank Group" is employed, it refers to the Bank and IDA but does not include the Bank's other affiliate, the Interna- tional Finance Corporation (IFC). Money amounts are expressed in U.S. dollar equivalents. The Bank's fiscal year (FY) ends June 30. CHARACTERISTICS OF THE SECTOR Role of Transport in Economic Activity Transportation is a necessary concomitant of the exchange econ- omy and is indispensable to economic growth. Where there is no tfansportation, economic activity is restricted to hand-to-mouth subsistence levels. Specialization and the generation of surpluses for exchange on the basis of comparative advantage are not possible without the capability to move resources and goods from one place to another. The demand for transport services increases with the extension of the input-output relationships of the economy, and the provision of transportation services can be an important deter- minant of the pace and locational pattern of development. In a developed economy the volume of goods and people trans- ported from one place to another is enormous. The contribution of transport to the domestic production effort is usually understated by standard national accounting methods; many transport activities are not included under transport headings. However, in developed economies, transport would typically account for 10% to 15% of Gross Domestic Product (GDP). In developing economies the con- tribution of transport to GDP varies considerably from one country to another, since the variables which determine the cost and vol- 143 ume of transport are fewer and individually more important in the simpler economy. However, where production is of relatively lower value per ton and transport systems are relatively inefficient or high cost, it is generally also the case that the contribution of transport to market GDP will be higher. For landlocked Chad, the proportion of transport in market GDP is estimated at about 17%. On the other hand, for refined copper-exporting Zambia (before the political situation in that part of the world distorted the transport picture), the comparable figure was only 7%. Demand for transport is a derived demand depending upon the demand for the commodities carried or for the benefits of personal travel. Likewise, the elasticity of demand for transport is determined mainly by the elasticity of demand for the things being transported, and by the proportion of transport costs in the value of the deliv- ered product. Only to a small extent is demand for transport af- fected by substitutability of other services (e.g. telecommunications and storage) and only in the longer-run is total transport demand given additional elasticity by locational decisions. Investigations in developed economies suggest that transportation costs are not a significant proportion of the final price of most manufactured and some mining products (the average is less than 10%), and for that reason the demand for transport is fairly insensi- tive to the price (or cost) of transport. This inelasticity is reinforced by the highly set locational patterns of production that typically exist in the more developed economies, and by the relatively low cost of transport due to more efficient transport systems. The situation is radically different in less developed economies where a larger proportion of the economically important move- ments are likely to be high-bulk, low-value agricultural and mineral products. The thresholds of viable production or export of these primary products are often determined by externally defined prices. Transport costs on the feeder roads, on the trunk road or railway to the coast, through the port, and on the ship to foreign markets often account for as much as 50% of the receipts from these com- modities. Maize in Kenyanand iron ore in Brazil are cases in point. Where transportation is particularly costly due to a lack of adequate facilities or inefficient operation, the impact on profitability and output can be decisive. Also, in developing economies where inter-industry linkages are less numerous and less strong, transportation costs are likely to have a much greater role in determining locational patterns of development than in developed economies. In short, price elasticity of demand for transport is likely to be relatively higher in less 144 developed countries, and reductions in transport costs are likely to have a larger impact in terms of bringing into production the mar- ginal piece of land, the marginal worker, or the marginal material resou rce. Commensurate with its role in production, transportation absorbs a significant part of total investment, although here again there is great variation from one country to the next. Table 1 gives the results of an attempt to examine the data for 45 countries; it shows the percentage shares of transport and communication in gross fixed investment in the post-1945 years. Table 1: Percentage of Gross Fixed Investment Allocated to Transport and Communication in 45 Countries, 1945-1965' No of 0,. Range of Gros, In,ome Levels Countries Fixed Investment Aserage % Over $500 per capita 15 13-36 19 $200-500 per capita 8 14-22 19 $100-200 per capita 7 4-26 15 Below $100 per capita 5 16-50 27 Socialist Countries 10 7-16 12 D. Bejakovic, The Place of Transport and Communications Investment in Total and Productive Investments in the Course of Economic Development, July 1965, financed by the Federal Bureau of Economic Planning, Belgrade. The figures in Table 1 understate the true significance of transport in investment, since they exclude automotive equipment invest- ments which, in many economies, can amount to as much as infra- structure investment. A very rough indicator of the norm of total transport sector investment would be about 5% of Gross Domestic Product (GDP) per annum. However, one is forced to agree with the conclusion of the study mentioned above and of many others, that there are no general guidelines as to the share of national product or total investment that should go into transport invest- ment at any stage of development. The evidence is also unclear as to whether the proportion of resources devoted to transport in- creases or declines as GNP or per capita income rises. The claim of the transport sector on a nation's resources is clearly very great, however, and the volume of transport investment seems never to taper off. The aim of investment in the transportation sector is to reduce the costs of production by lowering those of transport, thus facilitat- ing higher levels of production and consumption. When a capacity constraint is eased or a new area is "opened up" by a transporta- 1Ar tion investment, it is the effects of this upon the costs of production that counts, and the extent to which these savings are translated through consumer and producer surpluses into new output. Thus the real effects of most transportation investment in terms of em- ployment and income growth and distribution are indirect and, to a large extent, depend upon the market and upon policies and conditions in the transport sector and the economy as a whole which affect it. The safest strategy of investment in transportation is to wait until increases in production or captive productive schemes signal clearly the infrastructure requirements; that is, to wait until bottlenecks appear or solid traffic volumes and clear trends exist, and then to take appropriate action to increase capacity or reduce costs for an assured clientele. Given scarce investment funds, this is a quite common investment strategy and most Bank-financed transporta- tion sector investments follow this approach. The safest approach, however, is not necessarily the most devel- opmental, especially in the very poor country where the develop- ment problem takes on a chicken-or-egg complexity. Furthermore, it tends to reinforce existing patterns of income distribution and geographical patterns of development. Experience shows, and in a rather dramatic fashion, that promotional transportation investments can be the agent of important economic developments. Having said this, it is important to note that improved transport cannot create development by itself; the area and population served must have the necessary additional resources, and the other invest- ments required to exploit them must be forthcoming. To adopt suc- cessfully a more "entrepreneurial" approach to transport invest- ment requires detailed knowledge of the economic environment and of the kind of investment and institutional mix that will stimu- late production. There are enough instances where the invitation to produce extended by transportation investment has not been taken up-the uneconomic railway branch line and the overdesigned and underutilized road into the desert or bush-to show that the re- quired degree of knowledge is not common. Some Characteristics of Transport Systems Transportation needs can be met by any one or combination of the following modes: aviation, inland waterway, coastal and ocean shipping, pipeline, port, rail or road services.' The extent and modal 'Although transportation becomes increasingly capital-intensive as economies grow, human effort and animal power are still important in providing transport services over large parts of the world. 146 structure of a transport system reflects many influences, including the country's geography, terrain, the structure of its economy, and the history of its growth and development. (See Annex I for a basic inventory of transport facilities in 95 countries). Factors such as density of population and the disposition of resources, i.e., the basic supply and demand for goods and travel, are by far the most significant in determining the density and mix of transport modes in the network that has been or will be developed. Within any system, the modes included both complement and compete with each other. The complementarity aspects may in- volve a number of necessary steps in the movement of people and goods from points of origin to final destination: e.g., car and bus journeys to railway stations or airports, truck or rail movements from farms, ports, or pipeline terminals to distribution points or market outlets. Complementarity arises inevitably out of the scat- tered distribution of production, marketing, and residential points as well as technological and economic conditions. The different modes also compete with each other. Since each mode has different mixes of infrastructure and equipment, different economic sizes of vehicle unit, different speeds and other service characteristics, both transport and total distribution costs incurred in the use of any particular mode can vary widely over different lengths of haul, shipment size, terrain conditions, and volume flows. But shippers' choice and the design of public policy are not exclusively a matter of cost comparisons, in the narrow sense. Im- portant differences exist in the marketing capabilities of the differ- ent forms of transportation. For example, one reason for the success of trucking in competition with the railways has been a more decentralized and flexible organization, better able to meet par- ticular shippers' needs. The situation, therefore, is not one of perfect competition or substitutability or one mode of transport for another. Each mode has certain comparative advantages. The decision to move a specific commodity on a specific route by a specific mode is unique to the circumstances. Nevertheless, some broad generalizations about the advantages of each mode are possible. The technical and economic advantages of railroads appear first in handling large volume over long hauls. Equipped as mass move- ment carriers, they are normally the most economic and efficient for bulk traffic on intermediate and long hauls (over 150 miles). Transportation by rail, other than for dense passenger flows in urban areas, is generally uneconomic on short hauls due to relatively high terminal and overhead costs and the need to transship for door-to- I A7 door distribution. Most of the railways borrowing from the Bank are saddled with extensive systems built before the advent of the road transport industry, and with public service, employment, and pricing obligations which threaten their financial viability. This has led to patterns of serious underinvestment which in turn have threatened the railways' ability to fulfill their economic roles. This default on the part of railways has tended to inflate the demand for road transport that might otherwise be uneconomic. Motor transport is generally the most efficient mode for short distance and intermediate hauls. Trucking is equipped to handle commodities of medium and high unit value, and in lots relatively smaller than are advantageous for railway movements. Trucks have a great advantage because they give door-to-door service from shipper to receiver. Motor carriers are generally uneconomic on very long hauls or for high-volume, low-value commodities. The development of the goods transport market in many countries shows, however, that the motor carriers are conquering an increas- ing proportion of long distance traffic, often, apparently, due to the default of rail transport. Water carriers are able to move goods-especially bulky com- modities-at a relatively low cost per ton-mile over long distances. Generally water carriers have capacity to handle large tonnages, but they are normally much slower than other transport means. Extensive terminal facilities are needed, and port handling costs are an important determinant of the overall costs and competitiveness of this mode. Water carriers are often limited to the hauling of goods of low unit value, except in areas where other means of transportation are less developed. The primary advantage of air transport is speed. In order to pay for this quality, shipments and passenger movements generally should possess a high unit value. Air transport is primarily aimed at the movement of people where time is especially valuable. The competitiveness of this mode is vitally affected by the capacity and efficiency of terminal operations. Goods having medium or low unit value are not well suited to movement by air, except in special circumstances. Pipelines are able to move large volumes of liquids and natural gas over long distances and difficult terrain, reliably at a relatively low cost. In the case of land transport of natural gas, there is no economic alternative. The disadvantage of pipelines is that they must operate near capacity to ensure low unit costs. The movement of bulk com- modities such as iron ore in water suspension is a recent develop- ment in pipeline technology. I AQ Trends in Transportation Motorization. One of the most important trends in transportation is the accelerating pace of motorized transport. The growth of vehicle fleets on a worldwide basis is of the order of 15% per an- num; almost every country in the world is experiencing such growth. (See graph, Annex 5.) Urban Congestion. The rapid trend toward urbanization through- out the world, coupled with the growth of motorization, is creating problems of urban transport on a wholly new scale for both devel- oped and less developed countries. The future should see a far- reaching shift from the present emphasis on congestion-producing individualized conveyances toward public mass transport-bus and rapid rail transit systems. Environmental Pollution. The transport sector presents its share of environmental problems, which are just now becoming better understood. Pollution of urban atmospheres by the internal com- bustion engine is presently the most important. Bringing such pol- lution under control is likely to create new problems of vehicle regulation. A similar problem-compounded by noise-exists in air transport, although control may be easier to enforce. A special, but important and worldwide, environmental problem exists in petro- leum transport; oil spills are becoming so large and so common that special measures seem required to prevent excessive social and economic costs. The growing movement of chemicals, acids, flam- mable materials, and the like, by road and rail is also posing grow- ing potential hazards. Railway Reform and Recovery. Almost everywhere, railways are in or threatened with decline as a result of poor financial perform- ance and a long period of serious underinvestment or misinvest- ment. The emerging problems of indirect economic and environ- mental costs connected with individualized motorization and, to some extent, of air transport are forcing a review of the role of rail transport and the pricing, regulatory, and investment policies which affect its potential. Technological Change. The ability of transport investments to continue their cost-reducing contribution to development is being greatly affected by a steady stream of major technological changes. These changes continually alter the investment merit of existing and future transport projects and the competitive position of the various modes. While cost-reducing technological advance has made sub- stantial contributions to developing economies, most of the tech- nological change is consciously or unconsciously in the labor- saving direction, e.g., larger-scale technologies in ocean shipping 149 and air transport, containerization, pipelines, diesel-electrification, computerization, roll-on, roll-off techniques, huge earth moving equipment, etc. This reflects the fact that technological advance tends to originate in higher income, more fully employed eco- nomies. In many instances developing countries will be faced with important economic decisions on whether and how to adopt or adapt these technologies to their own situations. In some cases the problems are especially severe because the technologies may be imposed by outside influences regardless of the economic situation of the receiving country; for instance, airports and ports may have to be equipped to handle the planes and ships developed to meet the needs of developed countries, if trade and communications with these countries are not to suffer. Transportation Planning Because of the important interaction between transportation and the location, volume and growth of economic activity, and be- cause of the high benefits and costs that can be involved in a trans- portation policy or investment decision, comprehensive and long- range transportation planning is especially desirable. It is also diffi- cult. First, the overall national priorities on which transport has a direct bearing-economic, social, strategic-are rarely made ex- plicit, which means that from the outset the goals of planning are often vague. Second, since transport demand is derived, good transport planning implies the accurate assessment of activities and targets in other sectors and translation of these forecasts into traffic volumes along specific routes over time, and of the impact of the investment on these flows. Ideally, this requires a systems approach that captures the basic workings of the economy. The necessary tools for such an approach are unfortunately in a rudimentary stage of development. Third, the transport sector involves lumpy infra- structure investments, with long economic lives, which means that forecasting has to be acceptably reliable over long periods. With respect to the transportation infrastructure, each mode is often the responsibility of a distinct government agency, and even for a single mode there is often a multiplicity of government, busi- ness, and labor interests involved, each with its own political con- stituency. In one Latin American country, for example, seven minis- tries, six government-sponsored agencies, three international bodies, one foreign government, numerous trade unions, tourist in- terests and foreign airlines are vitally concerned about civil aviation investment and operations. In other modes and other countries a similar multitude of interests is often found. 150 Lack of comprehensive transportation policy and planning is prevalent in highly developed countries as well as in the less devel- oped. The situation is aggravated in the latter by the lack of data and trained personnel necessary to proper planning, although in some developing countries the impact of these deficiencies may be lessened by the fact that the problems they face are less complex. The consequences of this difficult and deficient planning situation are often misdirected investment plans and proposals, and transport pricing, investment and regulatory policies that are outmoded or ill- suited to the economic situation. The Bank, therefore, does not start with a clean slate or easy situations in its efforts to obtain an efficient use of resources in the transport sector. The Bank's impact in this sector is in large measure dependent upon its effectiveness in assisting improvement in transportation planning and policy. PAST ACTIVITIES IN THE SECTOR Lending By the end of FY 1971, total Bank and IDA lending for transporta- tion amounted to $6,160 million, divided into $5,220 million (85%) in Bank loans and $940 million (15%) in IDA credits. Table 2 sum- marizes the history of the expansion and changing modal mix of Bank Group lending for transport. Transport loans and credits have been the largest single com- ponent of Bank Group finance, accounting for more than 30% in terms of both lending volume and number of operations. The share of transport in total Group lending has fluctuated over the years, however, rising from 18% in the pre-1955 period to about 40% in the early 1960s and declining to some 30% in the last five years. While the Bank Group is one of the major sources of external finance for transportation projects, its lending represents only a small fraction of developing countries' aggregate investment in the sector. If it can be assumed that about 5% of GDP is allocated to public and private investment in transport, the 85 countries cur- rently borrowing from the Group for this purpose are investing more than $16,000 million per year in the sector. At current levels of lending, therefore, the Group's contribution to transportation in these countries would amount to about 4% of combined public and private investment in the sector, and less than 8% of the public component. In financing the general increases observed in borrowing countries in road mileage, paved surfaces, vehicle-kilometers, railway ton-kilometers, berthing space, etc., by far the most important source of investment funds has been domestic savings, both public and private. 151 Table 2: Bank Group Lending for Transportation (US$ millions) Through FY 1956 FY 1957-61 No. Amount % No. Amount % Highways 16 $162.0 24 16 $ 405.0 32 Railways 17 352.0 52 22 644.5 50 Ports 13 112.2 16 9 165.0 13 Pipelines 1 14.0 2 1 50.0 4 Aviation 2 41.7 6 2 14.8 1 Engineering (2 0 0.0 0 0 0.0 0 Totals 49 $681.9 100 50 $1,279.3 100 (1) The figures understate the full volume of transport lending, since they exclude an unidentified transport equipment element in some of the large post-war reconstruction loans in Europe, as well as the indirect component in the general industrial imports credits to India. In the case of multipur- pose transport projects, the full amount has been allocated to the mode A country-by-country breakdown of past Bank Group transport lending is presented in Annex 2. In about half of the 90 countries and territories in which transport projects have been financed since 1947, the Bank Group's involvement has amounted to less than $25 million. Lending to this large group of countries at such relatively low levels accounts for less than 10% of total Bank Group transport financing. In 15 countries, or a sixth of the total, transport lending exceeds $100 million, and these countries account for about two- thirds of the Group's total financing in the sector.t The East African Community and its constituent countries, together with India, Mexico, Pakistan, Spain and Yugoslavia, account for more than a third of Bank Group transport lending to date. Some important changes have taken place over the years in both the geographic and modal distribution of transport lending. In June 1960, there were 68 member countries in the Bank. A decade later the number had increased to 113, including 33 of the newly inde- pendent African countries. The creation of IDA in 1960 gave many of these new nations access to Bank Group financing that otherwise might not have been possible. Until 1960, the Bank had made loans for transport projects in 33 countries. In the next decade, some 50 additional countries and territories received Bank Group assistance, and half of these were in Africa. The trend has continued in the 1970s, with 13 new coun- ' Amounts in millions: India ($780); Japan ($510); Pakistan ($358); Yugoslavia ($305); Spain ($279); E. Africa ($263)- Kenya, Tanzania and Uganda for roads, and the East African Railways and Harbours Corporations; Mexico ($240); Argentina ($225); Colombia ($205); Thailand ($183); Iran ($170); Brazil ($153); South Africa ($150); Nigeria ($136); Peru ($133). 152 Cumulative FY 1962-66 FY 1967-71 through FY 1971 No. Amount ° No. Amount % No. Amount % 53 $1,069.8 56 78 $1,354.7 60 163 $2,991.5 49 19 674.8 35 20 668.2 29 78 2,339.5 38 14 157.2 8 16 191.2 8 52 625.6 10 1 15.0 1 4 58.0 3 7 137.0 2 0 0.0 0 0 0.0 0 4 56.5 1 1 1.7 nil 10 10.4 nil 11 12.1 nil 88 $1,918.5 100 128 $2,282.5 100 315 $6,162.2 100 accounting tor the highest share in the loan o,r credit. (2) These are loans and credits exclusively for transport engineering and do not include engineering financed under transport project loans or credits. tries being added to the list of Bank Group borrowers for transport, of which six are in Africa. Table 3 summarizes the regional dis- tribution of lending over the past five years. In these five years, about 48% of all transport projects were in Africa, accounting for about 27% of the money volume of transport lending. This relatively large volume is in part a reflection of re- cently reduced project activity by agencies such as the Fonds d'Aide et de Cooperation (FAC) and the United States Agency for Interna- tional Development (USAID). The creation of the African Develop- ment Bank has not yet had an impact on Bank Group transport operations in Africa comparable to that of the Inter-American De- velopment Bank in Latin America, where only half as many Bank projects have been financed in the five year period. In fact, for two years (FY 1967 and 1968) there were no transport operations in South America, though important preparatory work was being done which led to 14 operations totaling nearly $500 million in the three subsequent years. The low number of operations in Latin America Table 3: Transport Lending to Developing Countries, FY 1967-71 % of Borrowing Number % of Amount Dollar Countries of Loans Loans (Millions) Volume Africa 31 61 48 $609 27 Asia (Including Australasia) 20 34 26 727 32 Europe 5 10 8 334 14 Western Hemisphere 13 23 18 613 27 Total 69 128 100 2,283 100 reflects in part the increased activity of the Inter-American Devel- opment Bank, especially in financing smaller projects. Lending in South Asia, particularly for India and Pakistan, has reflected to a large extent the availability of IDA funds. The Asian Development Bank has financed a few transport projects in the South Asia, East Asia and Pacific Areas, but so far its impact on Bank Group trans- port activities has been small. Concurrently with the changes in geographic distribution, there has been a steady trend away from railways and in favor of high- ways in the modal distribution of transport lending. Up to 1960, railway lending accounted for more than half of all Bank transport loans and amounted to twice that for highways. Much of the pre- 1960 railway and port lending was for equipment to make up for the replacement lag caused by World War II and prolonged post- war shortages. There was also a substantial amount of maintenance equipment in some of the highway projects. This emphasis on equipment financing meant relatively quick disbursement of loan funds and limited project supervision problems. With many of the new members being dependent on highways and the general worldwide growth of motorization, highway financ- ing became the dominant feature of the Group's transport activities in the 1960s. In the ten years ending June 30, 1971, for example, highway lending totaled $2.4 billion compared with $1.4 billion for railways. The growing emphasis on civil works-often in new coun- tries which not only were unfamiliar with Bank procedures but also had less effective civil services or transport agencies-inevitably led to slower disbursements than for equipment financing and required more frequent and continuous staff supervision in project execu- tion. Lending for ports has been modest in contrast to that for roads and railways. Most of the small amount of lending for pipelines has been in the last two or three years. Bank involvement in aviation and shipping, for a variety of reasons, has been minimal (see pages 173-175). The Group has made no loans for the direct financing of bus or trucking companies. There has been some indirect in- volvement in vehicle manufacture and assembly, however, via the general industrial import credits to India and in truck and bus company operations through financing of development finance companies. In addition, there was a specific IDA credit to Pakistan for truck assembly and import of buses and spares, and the Inter- national Finance Corporation (IFC), an affiliate of the Bank, has made loans to and equity investments in vehicle manufacturing firms in Brazil and Yugoslavia. Physical Aspects of Transportation Projects, 1967-71 By far the largest element in lending for highways during this period represented road building, primarily of relatively high stand- ard paved roads. Bank Group operations helped to finance nearly 25,000 km of such works. Building of new roads accounted for about a third of the route kilometers while various sorts of improve- ments, including repairing, accounted for the remainder. Though in quantitative terms maintenance has represented a small fraction of lending for highways, equipment for the purpose was purchased under most loans and maintenance was the primary objective of several. Feeder roads represented a small part (estimated at less than 5%) of total kilometers financed by the Bank Group; how- ever, lending for this purpose has been increasing, not only in transport projects as such but as identifiable components of lending in other sectors, such as agriculture. Details of highway construc- tion projects are given in Annex 3, Table 1. In the past five years the largest single component (more than a third) of railway financing by the Bank Group was for rolling stock -approximately 200 locomotives, 600 passenger cars, and the procurement or fabrication of about 40,000 freight cars. The re- mainder of the financing for railways included a wide variety of equipment and materials, a large part of which was for track main- tenance and renewal. Very little new line has been constructed; projects have concentrated on renewal and rehabilitation. Details of railway projects are given in Annex 3, Table 2. Port projects have been devoted primarily to the provision of new berthing capacity, which has accounted for 60% of all such lending in the last five years. It is estimated that approximately 22,000 additional meters of berth were constructed and equipped under Bank loans. Other important objects of lending were the provision of cargo handling equipment such as cranes and forklifts and of harbor service craft such as tugs and lighters. Details of port projects are given in Annex 3, Table 3. Bank Group lending for pipelines in the five years comprised four operations which assisted the construction of nearly 1,200 km of new line and increased the capacity of an additional 300 km. Preinvestment Activities The Bank has acted as the Executing Agency for the United Na- tions Development Programme (UNDP) in most of the transporta- tion-oriented preinvestment work financed by that organization. Over the past five years, the Bank has supervised 57 major studies in 33 countries, plus four regional studies costing in total about 155 $40 million and involving nearly 700 man-years of work in the field by consulting firms and indlividual experts. These studies ranged fronm broad surveys of entire transport systems, with a view to devising investment plans and transport policies, to specific proj- ect feasibility work. Technical assistance funded by the Bank or IDA through their lending operations is a common feature of the Group's work. There is seldom a transport loan or credit which does not include impor- tant studies or technical assistance needed to help the borrower improve some aspect of its work or to prepare a subsequent project for possible lending. The latter "piggy-back" technique has been very successful in maintaining the planning and project preparation momentum developed by the Group's highway lending. Technical assistance has been an especially important feature of the Group's serial lending approach for railway recovery programs. The Bank Group has recognized the importance of good sector work for many years. However, while it has promoted and super- vised sector surveys in many countries (usually with UNDP finance) and made major transport investigations part of general country economic work in others, it has only recently organized a sys- tematic program of comprehensive in-depth transport sector mis- sions to major existing and prospective borrowers, to be carried out over a period of years. This will give the Bank an opportunity to develop specific preinvestment study programs for some of these countries, identifying and providing guidance to the borrower and the UNDP on the appropriate timing and scope of needed studies in the sector. Equally important, it will give the Bank the basic information and understanding necessary to continue and expand its transport lending, especially in areas where its involvement is reaching important dimensions. Other Sources of Transport Finance The main alternative sources of external finance for transportation investment are generally the same as for other types of large capital projects. One of the major sources is private contractor and sup- plier finance, often without government guarantees. Noteworthy multinational organizations are the regional development banks, i.e., the African Development Bank, the Asian Development Bank and the Inter-American Development Bank. The UNDP is the major source of funds for preinvestment work. In Europe, there is the European Development Fund (FED), the European Investment Bank (BEI) and the European Company for the Financing of Rail Rolling Stock (EUROFIMA), all of which are multinational. The major bilateral lenders in transport are Canada, France, Ger- many, Italy, Japan, the United Kingdom and the United States. Most bilateral lenders operate through both official aid and export- promotion agencies. Many agencies specialize in particular types of projects or equipment, or specific areas. For example, the U.S. Ex- port-Import Bank has lent more than 80% of its $1,000 million devoted to transport in the last five years for aircraft financing. EUROFIMA lends for railway equipment purchased in Europe. FED makes grants mainly for technical assistance, and to a lesser extent for capital projects. On the basis of very fragmentary data, it ap- pears that the Bank's lending in the transport field over the past five years is about equal to that of all other multilateral and official bilateral lenders combined. General Lessons Drawn from Past Operations Project Re-evaluation and Monitoring. The Bank is conscious of the need for better evaluation of the impact of lending in the trans- portation sector. In the past, it has been possible to make a full- scale reappraisal of a project's economic results in only a few cases. Furthermore, while the Bank has a system for supervising project implementation, it does not have a wholly satisfactory procedure for monitoring the key economic variables in the post-construction period. It will do so, for example, when it considers further lending for development of the same road, port or other facility, which happens in a significant number of cases. It also monitors the finan- cial results achieved by its revenue-earning borrowers; but these give only a partial indication of economic impact. More frequent project reviews are required, together with a continuous monitor- ing system designed on the basis of the Bank's appraisal, to illumi- nate the actual economic impact of each project. Improved Appraisal Tools. The Bank's research activities in the transport field have expanded greatly in recent years. They have been addressed to the needs of both the Bank and its borrowers for new and better tools of analysis, and for better understanding of certain aspects of transportation policy. In the past five years the transportation research activities of the Bank Group have in- cluded some 27 studies, of which five have been published.* The * World Bank Staff Occasional Papers: Jan de Weille, Quantification of Road User Savings, 1966; Hans A. Adler, Sector and Project Planning in Transportation, 1967; Alan A. Walters, The Economics of Road User Charges, 1968; Herman G. van der Tak and Jan de Weille, Reappraisal of a Road Project in Iran, 1969; Herman G. van der Tak and Anandarup Ray, Economic Benefits of Road Transport Projects, 1971. A "Central Economic Staff Catalog of Studies," which lists studies completed and in progress and explains how copies may be obtained, is available on request. external research budget increased from a few thousand dollars for one study in FY 1968 to about $300,000 in FY 1972. The research output includes computer models for project evalu- ation, risk and sensitivity analysis techniques, studies on road user costs and pricing, queuing theory applied to port operations, and many other topics. While this has been an important contribution, there is considerable room for further study and research. The Bank still has not solved the inherent problems of evaluating the micro- economics of most railway investment packages, since it is unable to account adequately for the system-wide effects and interdepend- ent nature of particular investments.'Highway design options involv- ing large cost differences are not as yet amenable to anything but partial microeconomic analysis, since the Bank is unable to quantify satisfactorily the incremental benefits to highway users or, in terms of savings in maintenance costs, the benefits of such design features as greater road width, stronger pavement and easier curves and gradients. Long-standing questions such as the economically and technically appropriate mix of capital and labor in construction and operation of transport facilities remain unresolved, due to the lack of much pertinent data and problems concerning the use of shadow pricing for labor and foreign exchange where this is necessary to correct for distortions in market prices (see page 169). Analysis of Developmental Impacts. More important than better microeconomic techniques, however, is the need to refine the Bank's analysis of the developmental effects of transportation in- vestment-the connection between investment and economic out- put. The major and undoubtedly justified effort to improve the Bank's ability to quantify the costs and benefits of transport invest- ments has had the effect of directing the approach toward one of micro, or engineering, economic and financial analysis, where quan- tification is more feasible. It has probably also helped to account for the fact that such projects as low standard roads to open up areas of potential production, and similar promotional investments, have not figured large in the Bank's lending program. In short, while the Bank can be fairly satisfied with its assessment of individ- ual projects, the analysis of intra-sectoral and, perhaps more im- portant, inter-sectoral priorities within a development strategy re- mains weak. If the Bank is to be able to assess priorities in the transport sector rather than merely say "yes" or "no" to an invest- ment proposal, it must go beyond the quantification of user savings and be able to define each project's dynamic effects on the econ- omy as a whole. This is all the more important if transport lending is to go beyond serving established demand and reinforcing exist- ing patterns of economic location and income, and to become more useful in helping to solve the serious problems of unemployment and geographic or sectoral income inequalities by generating new output. Better analytical techniques are necessary if the Bank is to make much headway toward quantification of intra-sectoral and inter-sectoral effects. Even more important is a much better knowl- edge of the transport sector and of the workings of the economy in particular borrowing countries, to which the sector study program described on pages 161 - 162 is directed. It should also be noted, however, that the main work to be done in improving this situation must be the responsibility of the govern- ments of member countries. Through periodic missions and special surveys, the Bank should be able to help draw together the results of sector analyses. But this at best can be no more than a review of the situation at a particular point in time, and it can certainly be no better than available statistics permit. The task of gathering and analyzing adequate data on a continuing basis must be performed by governments. The Bank's role should be one of encouraging the creation of suitable sector planning institutions and giving such advice as it can on these continuing operations. Institution Building and Policy Reform. Another important lesson to be drawn from past lending operations is the importance of policy, management, and institutional development and reform in determining the success of Bank-financed projects, and indeed in bringing projects and institutions to the stage where sound invest- ment is possible. In connection with Bank Group transport lending, railway, harbor and airport corporations or authorities have been set up as autonomous entities in an attempt to improve their efficiency. Systems to promote financial integrity and major ac- counting, costing and budgetary reforms have been undertaken in connection with Bank lending. Transport planning units have been created and staffed, and equipped with new investment and policy criteria. Highway authorities have been reorganized. Vehicle licens- ing and road transport regulation laws and practices have been re- formed. Pricing policies and specific rates, fares and taxes have been changed, as have procurement and contracting policies and proce- dures. Policies with respect to hiring and firing of personnel and wages paid have been affected. Limitations have been placed on certain kinds of investments and debt. Individual managements have been appraised and agreements have been reached that bor- rowers would consult with the Bank on certain appointments. In some cases, the transport sector has been approached as a whole and the Bank has sought to set its lending for particular projects in I1In the context of agreements on major policy decisions and perform- ance criteria set out in memoranda of understanding related to the entire sector. These types of project-associated sectoral reforms are a major part of the Bank's business. Most of them are backed by implement- ing studies or technical assistance if necessary. They undoubtedly have a longer lasting and wider impact on the economy and the sector than most of the investments being financed. Work on these institutional and policy matters certainly constitutes the hardest part of the Bank's field investigations. Such questions also raise the most serious difficulties in loan negotiations and project supervision. The aim of all of these attempts at reform is to create the policy and institutional conditions that will both permit and encourage the bor- rower and the borrowing country to make the most efficient use of resources in the transport sector. While the Bank Group has certainly had a large and positive impact on transportation institutions and policies in borrowing countries, the full potential for change in this respect is seldom quickly realized. Usually the most important reforms-those vital to sector efficiency-involve difficult political decisions on the part of the borrowing country, and far-reaching changes which are not easy to set in motion. Such reforms usually result only from close and continuing collaboration, at the staff level, between the Bank and its borrowers. The implications of this for Bank lending where institutional re- form is vital to the success of transport investment are: (a) the Bank should not get involved unless it is prepared to make a sustained commitment through a series of loans to see the process through; and (b) wherever possible, the Bank should adopt a sector, rather than a modal or project-by-project, approach. The key to a success- ful sector approach and the ability to assess policies and institutions tailored to local problems and conditions is, again, a sound knowl- edge of the sector. Training. The Bank's past experience has also shown that for long term transport and general economic development it is not enough to finance improved transport facilities; the training of peo- ple to operate them, and to plan and implement investments, is at least equally important. This has long been recognized, and most transport projects include training of one sort or another. Whatever the source of financing, all technical and project preparation assist- ance supervised by the Bank includes arrangements for training local staff. Most programs of this nature include overseas fellow- ships as well as participation in sector or feasibility studies. These efforts, however, have been only partially successful. It is often difficult to find suitable domestic personnel who can be assigned to such training programs for sufficient time. It is often equally difficult to find foreign consultants who are not only tech- nically capable but also adept at training and prepared to give ade- quate attention to the transfer of their skills. For these and other reasons, few transport institutions in borrowing countries have re- ceived the assistance they need in the propagation of new attitudes and practices at the staff level. The Bank Group is searching for more effective ways of providing it. PROJECTION OF FUTURE OPERATIONS Lending Program The Bank Group's projected transport program for the five year period 1972-1976 contemplates total lending of about $4,300 mil- lion for some 240 projects, not including the transportation elements of projects which are primarily concerned with urban development, tourism, agriculture and industry. This implies a level of lending for transport about 90%N, above the 1967-1971 period and about 135%° above the average for 1964-1968. Sector Work The enlarged program of lending will require more complete knowledge of the sector in many countries, and of its relationship to other sectors and development as a whole. Especially where Bank-assisted transport investments play a large role in the develop- ment strategies of member countries, a project-by-project assess- ment is not sufficient. Whenever possible in these cases, an effort will be made to base the transport lending program on comprehen- sive sector analysis, and to link support for a series of projects to agreed sectoral development strategies and policies. To raise the Bank's sector knowledge to the level appropriate to operational plans, and then to maintain it, will require a major effort. In addition to the necessary work in countries with the larg- est programs, it is highly desirable that in-depth sector knowledge be built up in a number of others, where the absolute amounts to be lent may not be so large, but where the Bank Group will be financing a large proportion of total transport investment. These include a number of relatively poor African and Asian countries where the challenge of using an appropriate mix of trans- port and other investments to open up new areas or stimulate new agricultural procluction and rural employment cannot now be fully met for lack of sectoral and general economic knowledge. 1 £1 The Bank's five year work program to increase its transport knowledge in support of projected lending calls for more than doubling the number of special sector missions. Including ex- panded transport work in connection with general economic mis- sions, some 45 such studies will be undertaken in about 40 coun- tries. These staff reviews will be in addition to sector surveys to be financed by the UNDP and carried out by consultants in a number of countries where they are most needed. As already pointed out, the Bank's own missions cannot do the long term spade work that only government departments, and to a lesser extent consultants working on transport surveys, can hope to accomplish. Instead, they will concentrate on the fundamental issues of sectoral objectives and strategy, institutional soundness, and pricing, regulatory and investment policies. They will also at- tempt to reassess previous Bank Group and other projects to under- stand better the forces at work in the economy and how and to what extent certain projects make an impact on development under these conditions. The Bank expects these missions to establish close and, in some circumstances, continuing relationships with overall transport sec- tor planning authorities, in the same way as project preparation and appraisal missions establish close communications with modal plan- ners and managers. Hopefully, this will help the Bank to achieve at the sector level the kind of institution-building success it has had in working with some highway departments and railway adminis- trations. Transport Policy Questions There is very little that can be held to be categorically true with respect to transportation pricing and regulation. However, the Bank intends to continue the use of certain general guidelines which have been employed in recent years. These are set out below. Pricing and Financial Targets. Pricing is one of the most powerful tools of transport policy for achieving efficient intermodal choice and economic utilization of existing and proposed systems, and therefore for contributing to efficient locational and production decisions throughout the economy. As a general point, the Bank believes that this resource allocation or economic function is the central function of pricing. At the same time, however, pricing must often be used for a num- ber of other objectives, which may or may not be in accord with its primary function but which may, nevertheless, be legitimate and important. These include the achievement of financial viability for 162 transport enterprises in order to ensure adequate resource mobiliza- tion, the generation of surpluses for expansion of facilities or for general taxation purposes, and the subsidization or preferential treatment of certain users or regions for economic or social reasons. Often these objectives conflict with that of ensuring the most effi- cient use of transport resources. Financial analysts and economists working in the transport sector are well aware of these potential conflicts and of the practical barriers to achieving satisfactory com- promises. The Bank's general guidelines are as follows: * For any particular transportation service the price charged to the user should not fall below the relevant economic cost of pro- viding the service. This "relevant economic cost," the short-run mar- ginal cost, may be low, as in a situation of chronically underutilized capacity-a back-haul situation, for example, or a seasonal slack period, or where a major investment mistake has been made. On the other hand, this short-run marginal cost threshold may be a very high one where congestion has set in and the economic costs of accommodating the incremental unit of traffic have begun to soar. Prices set below the relevant threshold would constitute sub- sidies. As a general proposition the Bank is not persuaded that in- direct subsidies through transport pricing are as efficient as more direct ones. * Since the country's economy must cover the total financial costs in any case, there remains the major question of the financial targets to be achieved by transport pricing. In short, the question is whether to cover the "overhead costs" by user charges or through general or other forms of taxation if the pricing rule of the paragraph above does not yield enough financial resources. The basic problem stems from the fact that to raise prices above the economic cost threshold, depending upon the elasticities of demand, may dampen justified output in the economy or shift traf- fic from a low cost to a higher cost service, on the one hand, while to charge marginal cost-based prices may lead to substantial finan- cial deficits and fiscal and cost discipline problems, on the other. Where financial discipline is not at issue, as is generally the case for highway infrastructure, short-run marginal cost coverage from user charges is the only mandatory economic rule (except where subsidies may be justified; see below). In covering the financial costs beyond that level, ideally one should evaluate the economic impact of any additional user charges vis-a-vis that of other forms of revenue-such as land taxes, or income taxes- taking into account the income distribution, production, locational ancl administrative or ease-of-collection aspects. The appropriate 1 61 level of pricing in a case such as highway infrastructure is also affected by the financial targets set for competing modes because the economically desirable allocation of traffic between modes may be unacceptably distorted by different financial targets for one mode as opposed to another. For revenue-earning entities the Bank believes, in the absence of a convincing economic case to the contrary, that financial viability is a valid target. This means that the entity should be able, from internally generated resources (i.e., user charges) to meet all of its financial obligations and make a substantial contribution to its justified renewal and expansion investments-substantial enough to maintain investment at appropriate levels without a dangerous deterioration of its debt position. This predisposition toward finan- cial viability rests largely on institutional grounds which obviously have a direct bearing on long-term economic efficiency: the need to maintain cost and investment discipline and to ensure adequate self financing for needed renewals and modernization. Where such financial targets have serious adverse economic im- plications in terms of dampening or diverting traffic, the first remedy after reaching acceptable levels of operating efficiency is to see what can be done about adjusting the financial burden, such as writing off unemployable assets and discontinuing their deprecia- tion, converting debt to equity, etc., so that all of the unnecessary "fat" in the "financial cost" is removed. If conflicts still exist be- tween financial and economic goals, the question of the extent to which each goal should be pursued has to be faced objectively on a case-by-case basis. It is important to note that financial viability as defined above says nothing about the level of profitability. A railway with a declin- ing role in the transport economy, with many assets which should not be renewed, and the likelihood of few if any justified expan- sions in the future, may be financially viable at a very low average return on net fixed assets. To try to achieve a higher return might mean unnecessarily pricing the railway out of markets where it is the low cost mover in terms of true economic costs, or unneces- sarily dampening the demand for transportation (output) in gen- eral. On the other hand, where the demand for railway, port, or airport services is expanding rapidly, relatively high rates of return on net assets may be required for the entities to be financially viable in the sense outlined above. Regulatory Systems. In a number of countries a variety of restric- tive measures and institutions exist that are ostensibly designed to ensure proper resource allocation in the sector by intervening in 164 the transport market-e.g., by setting "reasonable" prices, avoiding "excessive" competition by restricting entry into the industry, en- suring "rational" coordination, etc. The Bank has reservations about this type of economic regulation of the transport sector, and believes that caution should be exer- cised in using it in preference to sound pricing as a tool for allocat- ing resources. The economic case for regulation has been contested by economists and, most recently, in an important Bank research study.' The Bank's experience indicates that it tends to become rigid in response to vested interests, is often ineffective, and is prone to many abuses. To be efficient, regulation requires a degree of administrative knowledge and flexibility which may not be easily found. Furthermore, these regulatory systems are administra- tively expensive, and tend to cramp the development of entrepre- neurial talent, technological innovation, and general efficiency by inhibiting competition in the sector. Thus the Bank will continue to review regulatory services with care, and to press for their aban- donment where a sound case for them cannot be made. These general views on regulation and pricing do not exclude recognition that economic cases can be made for important excep- tions. While subsidies to certain sectors may be more efficient if made directly instead of through transport sector pricing, there may be cases where this is not true for practical administrative, institu- tional or political reasons. In some cases, sound policy within the transport sector itself may call for subsidies. For instance, where it is impossible for administrative or other reasons to charge conges- tion prices on automobile users causing high cost congestion, it may be desirable to subsidize rail or bus services in order to attract commuters away from private cars. The resulting savings to the economy may be large. In some cases, where the market mechanism is plainly not work- ing and other prices in the economy are also distorted, isolated ra- tionalization of transport pricing and regulatory systems may in fact produce undesirable economic effects beyond the sector. In other cases where massive unemployment and chronic shortages of fiscal and foreign exchange resources exist, it may be rational, at least for a time, to restrict importation of road vehicles, thus protecting a railway from road competition so that its full capacity can be utilized and it can simultaneously operate with redundant labor and earn financial surpluses for support of the government budget. ' Oort, Conrad J., "The Economic Regulation of the Road Transport Industry," World Bank Report No. EC-177, October 1970. 1ICE Directions in Future Lending Areas of particular concern or emphasis with respect to the major subsectors in transportation for which the Bank Group provides finance are discussed in this section. (a) Road Transport The road transport lending program for the next five years will include the same kinds of projects the Bank Group has helped to finance in the past, ranging from low-cost earth and gravel roads to multi-lane limited access highways. However, within that program greater emphasis will be given to "feeder" road development and improvement of highway maintenance, and more effort will be devoted to solving the many training and institutional problems in- volved in projects of this kind. More must be done also to en- courage the development of local civil works contracting capacity, to ensure that design standards fit local conditions, and to substitute labor for capital wherever appropriate. Feeder Roads. As used here, "feeder road" means a road which depends for its justification upon generated traffic, i.e., new output which would not take place if the road were not built or improved. These are almost always low-standard access roads built at very low cost. They can be justified only within the economic framework of the particular region in which they are located and in terms of the cost-benefit calculus for producing the new output. Therefore, the problem of assessing feeder roads is essentially one of measur- ing the response of output to changes in transport costs along with and relative to all other inputs, such as land clearance, settlement, storage and marketing facilities, etc. In practice, this often means an agricultural project with a transport element. One problem is that the measurement of output response is quite complex even in very simple agricultural situations, while the in- vestment costs per mile of road are for the most part so low that expensive preparatory feasibility and design work in each case sim- ply cannot be justified. It would be more economic to make a per- centage of investment mistakes than to conduct sufficient localized studies to remove the uncertainties. The Bank proposes to develop a series of selected feeder road projects based on in-depth analysis, which will be monitored carefully in the post-investment period. From this experience, the Bank hopes to develop sound "rule of thumb" criteria for investment and design decisions and guidelines as to the type of local institutions required to plan and implement these projects effectively. I ;E Another problem emerges in building these roads. Most will be of such low standard and low cost that detailed surveys and precise location prior to actual construction are not required and are too expensive. Thus contracting in the manner followed for larger civil works will not be practical. Typically, this is a force-account opera- tion, carried out by a labor-machinery unit not unlike a mainte- nance unit, with a fixed budget and an area for which to provide roads, and to maintain and upgrade them as required by traffic growth. Force-account construction work has been rare in Bank- financed projects because of the many problems involved in work performance, budgetary and cost control. These problems will have to be overcome if the Bank is to make a worthwhile contribution to this type of road development. This is true even if assistance is restricted to the financing of necessary equipment and technical and management personnel. Road Maintenance. Maintenance is the "cinderella" of the road business. It is customary for Bank Group loans for highways, whether for construction or maintenance, to include a requirement that the borrower maintain its road system adequately. Much effort has already been devoted to this end. Experience includes cases where it has been possible to bring about improvements. Unfortu- nately there are also cases where little has been achieved, and still others where, after a period of improvement, a change of govern- ment has led to a sliding back to unsatisfactory practices. The dependence of maintenance on current budget expenditure for which foreign financing is generally not available, and its rela- tive lack of glamor compared with new construction, militate against it when scarce capital and human resources are allocated among competing needs. It is quite obvious, however, that the highest returns on resources can be earned in maintaining existing assets, and that without solid maintenance new investment is a wasteful venture. Establishing maintenance targets, building up road maintenance organizations, training personnel and equipping them properly will receive even more emphasis in the future, and it will be accompanied by efforts to ensure that adequate current budg- etary resources are devoted to this purpose. Local Contracting Capacity. The civil works contracting industry is a fairly basic one, and a logical one for development at relatively early stages of economic growth. In the Bank's view, it is important both for the road sub-sector and for the economy as a whole that this development be encouraged and supported as much as pos- sible. To this end the Bank has for many years tried the technique of "packages" of contracts. Wherever it has seemed appropriate the I CZ7 Bank has broken road construction projects into the smallest con- tracts consistent with efficient execution and permitted contractors to bid for single contracts or any "package" of them. The objective has been to achieve a contract within what is often the fairly low capacity of local contractors, while also attracting the larger foreign contractor through packaging, and thus gaining for the borrower the benefits of widespread competition. The Bank has also en- couraged, but not required, joint ventures between foreign and local firms and sub-contracting to small local firms. More recently it has tried in one country to assist local contractors more directly by financing consulting services to advise them in bid preparation, costing and the like. In another country it has stimulated an inves- tigation into ways of developing the local contracting industry. The Bank believes that more could be done. It will explore fur- ther what technical assistance might be given to local firms through consultants or in other ways. Recognizing that in some countries shortage of equipment and of the foreign exchange to purchase it have hindered the growth of local firms, the Bank might consider financing equipment for the use of local contractors. This might be possible on a rental or hire-purchase basis through a government department or development bank. It is clear, however, that member governments must also take complementary measures; indeed, these are more important to de- velopment of the local industry than any steps the Bank can take. Local contractors are unlikely to develop unless there is a clear policy of putting road construction and other work out to contract. Similarly there will be a need in some countries to review existing taxation and contracting procedures which are unfavorable to con- tractors. The Bank has recently started a research study to explore these matters further. Design Standards. Because of the large changes in motor vehicle dimensions and operating characteristics that are taking place, espe- cially in developing countries where the vehicle fleet is expanding very rapidly, and with the wide differences among countries in the relative prices of factor inputs, there is need for a thorough review of the road design problem. The objective is to develop criteria capable of yielding economic design answers in response to large variations in the relative costs of inputs. Intensive, long-range re- search into this problem has been started in Kenya by the Road Research Laboratory of the United Kingdom, in cooperation with the Bank. The Bank expects to be able to arrange similar work in Western Africa with the Ponts et Chaussees of France. Controlled experiments are being carried out to develop data on road con- I C Q struction, maintenance and vehicle operating costs, and the trade- offs between them. With the help of consultants connected with the Massachusetts Institute of Technology, the Bank is contributing to the effort by developing computerized models capable of assess- ing these trade-offs in specific cases and giving guidance toward optimum designs. In the next five years the Bank hopes to dissemi- nate the results of this research and encourage their use in projects it helps to finance. Labor-Intensive Versus Capital-Intensive Technology. While there are substantial civil works elements in lending for other modes of transport, and the choice of technology obviously goes beyond civil works design and execution, the Bank has chosen road construc- tion as a convenient area for coming to grips with the problem of labor-intensive versus capital-intensive technology. This is because part of the problem is to define the technical substitutability of labor, capital, and other factor inputs within the context of design options with measurable economic characteristics. The Bank will be better able to do this in the highway field because the design studies mentioned above will help to provide this basic context. The substitutability of labor for capital in highway construction is being investigated in a parallel study recently started by the Bank. In addition to the questions of technical-economic trade-offs, there is the question of what relative prices for factor inputs are valid for use in design, tendering and execution decisions. Where there is a chronic shortage of capital and foreign exchange and a surplus of unskilled labor-and therefore market prices for these inputs do not reflect their true economic cost-the need arises for shadow pricing in designing pro4ects and for methods to ensure that contractors will use the proportions of labor and capital that are most economic on the basis of these notional prices. The Bank has no illusion that appropriate shadow pricing can be easily determined or that results can be much better than rough approximations. Perhaps until better techniques are available, the best that can be done is to recognize situations where there is a substantial disparity between market prices and economic costs and attempt to allow for them in design preparation and appraisal work, by giving full consideration to local resources and to the possibility of substituting labor for capital. In construction, it may be possible to develop reasonable arrangements whereby governments provide financial incentives to contractors (without discrimination between local and foreign firms) to use labor more intensively. It is pri- marily for governments to establish the necessary financial and ad- ministrative systems, but the Bank is prepared to help in formulat- 169 ing practical arrangements. Road Transport Industry. Most of the Bank's highway lending operations have been designed to reduce transport costs through improved or more direct roads or better maintenance, but little at- tention has been given to potential savings from improvements in the vehicle fleet and in the road transport industry generally. In some cases the full potential in transport savings attributable to road improvement is dissipated in the general deterioration of the vehicle fleet. In many countries with balance of payments difficul- ties, severe import restrictions often result in buses and trucks being kept in service much beyond the point of economic replace- ment or remaining idle for long periods because of lack of spare parts, thus substantially increasing transport costs. In some cases also, vehicles are not adapted to modern traffic needs and the high costs of operation may dampen economic growth. The Bank Group's appropriate role in lending for road vehicles is still a moot question. If the operating units are small, such lending might raise difficult problems in terms of international com- petitive bidding and project supervision. Long term Bank Group loans may also not be well suited to the relatively short economic life of the equipment involved. A more suitable form of Bank in- volvement might be to support banking institutions designed to finance the purchase of buses and other commercial vehicles. Such institutions could also provide guidance to operators and thus as- sist the healthy growth of the road transport industry. In addition to these problems of the mechanics of lending, pos- sible difficulties also arise from the existence in some countries of high cost and heavily protected local manufacturing industries. The Bank does not propose to become active in motor vehicle or road transport industry financing before a great deal more thought is given to the problems entailed. As a part of this examination, as well as for other reasons, the research program includes a series of studies of the trucking industry. The Bank also proposes to give more attention during its sector analyses to this aspect of road transport. (b) Railways Most of the railways seeking Bank Group assistance can be placed somewhere in the following highly generalized model of decline. Typically, the railway was built in large part prior to the advent of road transport, and was intended to serve all kinds of transport needs. Since the railway began with a virtual transport monopoly, it was easy to assign to it developmental tasks. These I 7X tasks were often undertaken through pricing structures that did not reflect railway costs. Traffic was charged according to what it would bear; overall financial targets were met by cross-subsidiza- tion, or at least by very heavy differentials in the contribution of particular traffics to overhead costs. Because the railway had no competition, it was also easy to use it as a means of providing em- ployment, with the costs of excess labor borne by the public in the form of higher tariffs. Also, dubious investments could be, and were, undertaken and absorbed without financial disaster. When competition from road transport became a reality, financial targets were immediately threatened. Rather than release the railway from its developmental and employment tasks and its by now traditional pricing system, attempts were made to protect it from the new competition. Many of the Bank's railway borrowers are to be found in this brittle monopoly stage, propped up by restrictions on road trans- port operations and road investment. The length of this stage de- pends on the administrative strength of the restrictions, the degree of identification between government and railway, and the entre- preneurial talent in the local road haulage business. Usually it cannot last long. In most of the countries concerned the situation is rapidly changing and the monopoly stage is about to end. Where the monopoly is perpetuated, the eventual problem will be even larger and more difficult to solve. Saddled with excess labor and prices in which there are heavy vested interests, the railways begin to lose money once their monopoly is broken and the competition takes away the high-rated traffics which are priced far above railway costs, leaving those that make little or no contribution to overheads. In most cases, result- ing deficits must be borne by the state. Because these deficits rep- resent subsidies to industry, agriculture, passengers and labor, end- ing them poses a major political problem. Nevertheless, the state is usually loathe to bear these deficits, and the railway cannot con- vincingly defend itself against the inevitable charges of inefficiency. Self-defeating economies are undertaken: first by slashing depre- ciation, then by cutting maintenance expenditure, and finally by curtailing renewal and modernization investment. Thus there is a vicious cycle of financial deficits, underinvestments, declining serv- ice, loss of traffic and still larger deficits. The situation is often aggravated by weak railway management which has been protected and demoralized by the effects of public policy and deficit operations. Poor public policy often accom- panied by weak management eventually leads to the point where 171 the railway is in grave danger of losing its role in the economy. By this default, major and often excess investment in competing road transport capacity is encouraged, at high economic cost to the country. This stage has been reached in many developed countries, as well as in a number of the Bank's developing member countries which are current and prospective railway borrowers. The strategy of Bank lending in the railway sector has been to help break or avoid this cycle of railway decline, where the rail- ways have a proper economic role to play in the future develop- ment of member countries. While there may be a few projects in- volving new construction to serve captive demand, by far the majority of railway projects will require a continuation of that strategy. They will focus on rehabilitation, extensive reform, and re- treat from over-extended systems with the aim of enabling the rail- way to fulfill its economically justified role in a competitive envi- ronment. Major emphasis will be placed on transport sector pricing, regu- latory and investment policy reforms, the achievement of appropri- ate financial targets (see pages 162-165), the strengthening of management and the development of better management tools, especially in the marketing and cost-pricing fields. Where neces- sary, efforts will be made to reform relationships between the rail- way enterprise and the state. Disinvestment and the elimination of uneconomic services will be important complements to the invest- ment programs. Comprehensive agreement on sector policies and long term commitments through a series of Bank loans are neces- sary features of this kind of undertaking. Research activities related to the railway field and the Bank lend- ing program over the next five years will include: (a) a study of the methodology to be employed in making line and service elimi- nation decisions; (b) development and adaptation of existing com- puter models into flexible tools for use in Bank Group appraisals, with the aim of allowing better quantification of the system effects of investments; and (c) development of railway costing-pricing guidelines which can be adapted by borrowers to their individual needs. (c) Ports and Shipping The future lending program for ports and shipping calls for some- what more emphasis on coastal and ocean shipping than in the past. It will be affected significantly by the rapid pace of change in technology and by the sheer size of ships and loadings now enter- ing into trade. 172 The main arguments for development of ocean shipping for some countries are that shipping under national flag saves foreign ex- change and that it can be an important element in promoting ex- port trade and cutting the economic cost of transporting imports and exports. In some cases these arguments are persuasive; how- ever, there are many prima facie counter arguments which have molded the cautious approach of the Bank Group to the finance of ocean shipping. First, fuel costs, ship capital costs, crew expenditures, repair and foreign port charges are often foreign exchange costs which cut significantly into expected foreign exchange savings or earnings. Second, the profitability of national flag shipping from an economic viewpoint often depends fundamentally upon rates being charged in the market; these can fluctuate widely from one period to the next, making investment decisions based on them especially uncer- tain. It is questionable whether capital-poor countries can afford to run the risks of the shipping trade, unless the potential gains are commensurately large. Coastal and inland shipping are different matters. Here the ques- tion is whether to meet demand by water or by land-air transport systems. The investment decision problem is no different from de- ciding whether to build a gravel or a paved road on the basis of cost-benefit analysis. Shipping is potentially an important mode of transport for many borrowers with long coastlines, navigable rivers and great lakes. For the most part the potential for this mode seems to be underdeveloped in view of the minimum of infrastructure that is usually involved, and the typically large capacity for bulk movement that can be made available at relatively low investment costs. An important problem appears to be in the management and organizational fields, and in ensuring that port operations and procedures do not lead to undue delays and thus to high costs. As to ports, in addition to building berths and equipping them to handle increasing traffic, future projects will also involve modifica- tions to existing infrastructure and new designs to provide for the recent changes in ship technology. Deep sea, single point moorings with pipeline connections are needed in many parts of the world to handle the behemoth oil tankers now coming into service and re- lieve congested port areas. The trend toward large ships will proba- bly affect dry bulk operations as well, requiring substantial dredging to deepen harbors, or pipeline arrangements to permit offshore operations. Probably the most talked-about advance in shipping technology is containerized loading, using specialized vessels and dockside 173 equipment. Containerization is developing rapidly due to the eco- nomic advantages entailed, particularly in the North Atlantic, Eu- rope-Australia, and possibly North Pacific trade lanes. Shipping companies are investing heavily in container ships of ever-increas- ing size. First generation smaller ships are being released to second- ary routes. A major problem of technological adaptation is likely to emerge for many developing countries where containerization economics (labor saving) do not apply, but where containerization is likely to be imposed by the specialized equipment of the developed trading and shipping nations. When investment in new equipment and capacity speed turn- around time for foreign owned ships and perhaps reduce domestic employment, important questions about the distribution of eco- nomic benefits arising from port improvement investment must arise. How these benefits accrue to the country making the invest- ment is a question which has in the past been largely answered by hypothesizing plausible shipping price movements. In the future the characteristics of shipping and port pricing in each case should be analyzed in more depth, to see what can be done with port pric- ing to retain benefits or to encourage behavior on the part of shippers and shipping companies which will yield maximum bene- fits to the country. To this end, the Bank is starting an investigation into the pricing practices of ports in relation to shipping and inland customers, and into the potential efficacy of port pricing as an eco- nomic tool. (d) Aviation Civil aviation, despite its importance in any modern transporta- tion system, is a new field for the Bank Group l and one which it is entering cautiously. The Bank's program over the next five years is quite modest, consisting predominantly of airport projects. As in the shipping discussion above, a broad distinction can be drawn between international and domestic aviation and most of the same arguments apply. International Aviation. Bank involvement in international aviation is expected to be limited to the financing of airports. The Bank is not planning to become active in financing aircraft for international airlines. This is in part because finance is readily available on rea- sonable terms from bilateral sources, and because there are ques- ' During the 1950s the Bank made six loans wholly or in part for aircraft procure- ment to the Netherlands, Australia and India, but made no further loans for aviation until 1971, when it agreed to assist an airport modernization project in Panama. 174 tions as to the priority of such investment. International airlines in many developing countries are operated as a matter of national prestige; with some notable exceptions, they tend to lose substan- tial sums of money and often involve net foreign exchange losses as well. The growth of international air travel and technological change in aircraft confront many developing countries with the prospect of mounting congestion or unsafe operations unless relatively large airport investments are made. The main question the Bank faces in appraising these projects for possible financing has to do with their economic priority-establishing the links between aviation operations, airport congestion, and economic activity. This is not easy to do, especially since relieving or avoiding congestion basi- cally means saving time. Time for capital equipment (aircraft) is relatively easy to evaluate, but often before congestion occurs on the runways and in the airspace around an airport, the situation within the terminal in terms of delays to passengers has become acute. Time savings for individuals, however, are difficult to assess in terms of their economic benefit to the nation. They may in part amount to a form of non-monetary income (consumption) for the individuals, many of whom are foreigners, that is not translated in any way into further output. Thus while these time savings and increased passenger comfort may be of economic value, it is par- ticularly difficult to trace how and to what extent they benefit the borrowing country until congestion begins to retard traffic develop- ment with adverse consequences for the level of business activity or tourist expenditure. To assess adequately these congestion-relieving projects, good market information is required on elasticities of de- mand to time losses and other costs. In regard to airport projects the Bank also faces questions of an institutional nature, such as whether or in what circumstances it is important for airport authorities to be set up as semi-autonomous agencies with separate viable financial structures, rather than as government departments as in the case of most highway authori- ties. There are also questions as to pricing policy and financial targets in this field, which lacks real competition, serves foreigners to a large extent, and where demand, at least for the use of the airport, is quite inelastic in any reasonable price range. The Bank intends to strive for direct recovery of costs from the users. Where the con- sumer surpluses accruing to passengers are large, important fiscal surpluses may be available and justified because they help to keep 175 benefits within the country or contribute to public savings and a better income distribution. Domestic Aviation. Many borrowing countries have geographical or topographical features as well as population distribution that make domestic air transport economically and commercially attrac- tive. In almost every economy there is a distinct and growing role for aviation--from opening up new areas before investment in ex- tensive fixed transport facilities can be justified, to generally en- hancing the mobility of scarce entrepreneurial and administrative talent within the economy. In planning their investments in international airports, the devel- oping countries have largely to react to aircraft types determined by the needs of other countries. But in planning development of their domestic aviation they have some opportunity to select the optimum combination of aircraft, airports, navigation aids and the like. The Bank will try to help in developing some projects with this aim in view. Any Bank finance involved would probably be for air- ports and navigational aids rather than for aircraft, for which other sources of finance are readily available. At the same time, it should be noted that other sources bring little in the way of critical institu- tion-building, management training, or concern for pricing policies that would accompany Bank lending in this field. Therefore, the Bank intends to keep an open mind toward aircraft lending for domestic aviation, and may be open to persuasion for high priority proposals. (e) Urban Transportation Transportation has significant effects on where and how cities grow and how they function. The convergence of transport routes or the transshipment points from one mode to another frequently provide the focus and impetus for urban development. Mobility and accessibility to goods, labor, and markets permit and encourage the urban concentration. Development patterns within cities are, to a large extent, shaped by transportation investment decisions which have a large impact on land values and land uses, just as the latter have an important bearing on transport requirements. Among other important reasons why transportation consistently ranks among the chief concerns of cities is that it is a prerequisite for jobs and busi- ness; it accounts for a substantial part of personal and public ex- penditure; and it typically occupies from 20% to 40% of the city's physical space. The rise of motorization in cities where land use patterns and intensity of habitation had a prior development, as is the case in 176 most of the big cities of the developing world, creates even more severe problems of congestion than in more modern cities. Whereas the typical Western European or North American city devotes about 25% of urban land area to city streets and highways, the percentage is considerably lower in Asia; for example, it is 14% in Bangkok and 5% in Calcutta. Shortage of road space is often aggravated by competing claims on its use from animal-drawn and other slow non- motorized vehicles, and by the lack of or ineffective use of traffic control equipment. Those worried about congestion reaching intolerable levels can derive little comfort from the fact that most people in low-income countries cannot yet afford automobiles. Total automobile registra- tion in many of these countries is increasing by 10% to 20% per year, and nearly all of these vehicles are concentrated in a few big cities. Bangkok has 730% of all the automobiles in Thailand and is adding about 15,000 new cars every year. In Sao Paulo, population increased between 1960 and 1967 by 42%; car registration in- creased 324%. For many of these cities, the amount of resources that would be needed to transform them to accommodate continued expansion of the private vehicle fleet is enormous. It is clear that for economic and environmental reasons more reliance must be placed on mass transit services to meet the transport needs of cities. In planning urban systems along most economic lines, attention will have to be paid as much to the demand side and how it can be affected by pricing, regulation and levels of service as to the supply of the facilities to meet demand. The problems of urban transportation are so complex and varied, and so inextricably linked with urban development problems, that there is no single solution for them. The first requirement is to analyze the overall investment needs for the urban area or the region under study and to assess the relative priority of urban transport as compared to other crucial needs such as sanitation, employment or housing. When urban transport ranks among the priorities, there is no prima facie reason to exclude any aspect of urban transportation from consideration by the Bank. The most suitable types of projects, however, appear to be mass transit sys- tems, urban roads, traffic control systems, terminal facilities, and the development of joint uses of transportation rights-of-way to conserve and employ high-value land in city centers. The figures for future lending mentioned on page-161 do not in- clude finance for projects of a strictly urban nature. The Bank's urban transport lending program and the problems to be faced in 177 its development are to be addressed later in this series in an Urban Sector Working Paper. Research It is probably already clear that research is an important part of the Bank's activity in the transport sector. This is true for a number of reasons. First, there has been a general lack of pertinent research work by international bodies in this field, aside from simple anal- ysis and the publication of data. Most have been concerned more with improving operations, furthering technological advances, or exchanging views on practical experiences. Some research is under- taken in the field of ports and shipping by the United Nations Con- ference on Trade and Development (UNCTAD), by government agencies such as the U.S. Department of Transportation, the U.K. Road Research Laboratory, and Ponts et Chaussees in France, as well as by universities and by some consulting firms. More important is the fact that most research in transport is geared to the needs of developed countries. The results have to be reexamined and adapted to the needs of the Bank's borrowers. It is here that the Bank can play a constructive role by commissioning selected studies. These may be research into new directions or the testing and development of existing knowledge and techniques, to help the Bank improve the quality of its analysis and strengthen the foundation of its advice. A good example of the lead which can properly be taken by the Bank, supported by appropriate funding, is the cooperation already mentioned with the road research agen- cies of the British, and probably the French, Government on high- way design standards in Africa. In its study of the use of labor and equipment in road construction, the Bank hopes to have the asso- ciation of the International Labour Organisation (ILO) and some borrowing member countries for data on projects being executed. In addition to focusing on problems affecting developing coun- tries, it can be noted from the kind of studies described in this paper that the Bank's research activities in transport have turned increasingly to practical subjects. Theoretical issues have been fairly well covered in previous studies, but there are wide gaps between theory and practice, and the Bank must learn more about the actual effects of deviations from "ideal" policies (e.g., economic user pricing) and of distortions in the markets of transport services (e.g., restrictive regulatory systems). Another important gap is that of data; the paucity of reliable facts often calls into question the use of sophisticated models and the resulting conclusions. Recognizing that these problems may not 17R lend themselves easily to generalization, the Bank feels that the best way of achieving its goals would be through actual case studies in conjunction with its sector work. These require both time and man- power. While theoretical studies can be done relatively quickly and with little manpower, data collection and case studies are expensive. In the execution of an expanded research program in the trans- port field, the Bank hopes increasingly to involve the developing countries themselves, especially through grants to universities and other research institutions. A start has been made in this direction. For example, the Bank has contracted with the Asian Institute of Technology to do some analysis of berth operations at the Port of Bangkok. Similarly, it is attempting to bring the University of East Africa into one aspect of the highway design studies and to moni- tor the results of a Tanzanian feeder road project. 17Q Basic Transport Data Mid-1969 GNP Population Area per cap. Road Network Length (Kin) (million) (Km2) (US$) Paved Gravel & Earth Tntal Eastern Africa Botswana 0.6 712,195 95 16 4,976 4,992 Burundi 3.5 27,833 48 66 5.000 5,066 Ethiopia 24.8 1,031,647 67 1,957 5,347 7,304 Kenya 10.9 582,601 133 2,489 39,171 41,660 Lesotho 0.9 30,342 83 160 1,680 1,840 Malagasy Republic 6.7 590,520 111 2,900 35,100 38,000 Malawi 4.4 93,492 69 315 10,173 10,488 Mauritius 0 8 1,865 232 1,040 240 1,280 Rwanda 3.5 26,328 52 10 5,990 6,000 Somalia 2.8 637,611 66 900 15,100 16,000 Sudan 15.2 2,505,632 105 281 986 1,267 Swaziland 0.4 17,362 180 205 2,369 2,574 Tanzania 12.8 941,931 97 2,234 31,647 33,881 Uganda 9.5 236,019 113 6,382 17,791 24,173 Zaire 17.9 2,345,227 78 1,900 138,100 140,000 Zambia 4.1 752,560 289 1,579 31,759 33,338 Western Africa Cameroon 5.7 475,438 151 1,270 20,230 21,500 Central African Rep. 1.5 611,952 130 75 21,325 21,400 Chad 3.5 1,283,901 66 74 30,651 30,725 Congo, People's Rep. of 0.9 341,973 220 250 8,050 8,300 Dahomey 2.6 112,612 83 765 5,785 6,550 Equatorial Guinea 0.4 28,000 285 396 885 1,281 Gabon 0.5 264,390 324 111 5,444 5,555 Gambia 0.4 10,373 110 260 380 640 Ghana 8.3 237,855 189 3,975 28,200 32,175 Guinea 3.9 245,837 95 862 12,138 13,000 Ivory Coast 4.8 330,251 242 1,300 32,700 34,000 Liberia 1.5 111,361 203 483 2,300 2,783 Mali a.9 1.201,667 92 1,300 11,700 13,000 Mauritania 1.1 1,085,724 142 575 8,550 9,125 Niger 3.9 1,266,902 79 488 5,875 6,363 Nigeria 64.6 923,701 80 15,300 72,000 87,300 Senegal 3.8 196,177 200 2,046 13,376 15,422 Sierra Leone 2.5 71,735 168 530 7,600 8,130 Togo 1.8 52,832 103 481 6,692 7,173 Upper Volta 5.3 274,180 56 419 16,481 16,900 East Asia & Pacific China, Republic of 13.8 35,962 301 5,832 9,449 15,281 Fiji 0.5 18,271 391 181 2,304 2,485 Indonesia 116.6 1,905,743 100 15,650 66,350 82,000 Korea, Republic of 31.1 98,423 213 3,120 34,050 37,170 Malaysia 10.6 332,608 342 13,845 4,607 18,452 New Zealand 2.8 268,676 2230 39,170 54,762 93,932 Papua New Guinea 2.3 475,332 212 100 11,900 12,000 Philippines 35.9 299,658 208 12,371 51,222 63,593 Thailand 35.1 518,343 162 5,400 11,900 17,300 South Asia Afghanistan 14.0 657,449 85 2,000 15,000 17,000 Ceylon 12.2 65,605 187 15,297 5,658 20,955 India 526.0 3,267,284 107 324,940 647,390 972,330 Iran 28.5 1,647,872 349 10,200 25,600 35,800 Nepal 10.8 140,787 80 661 2,019 2,680 Pakistan 126.7 946,647 105 27,568 77,093 104,661 180 Annex I Page 1 Railway Route Road Railway Vehicles Length Vehicles Density Density per 1,000 (Kim) Light Heavy Total (Km . Km2) (Km-. Km') Population - 2,255 3,426 5,681 .007 - 9.5 - 3,528 596 4,124 .182 - 1.2 781 39,000 12,200 51,200 .007 .001 2.1 2,160 115,000 10,300 125,300 .072 .004 11.9 - 2,111 2,723 4,834 .061 - 5.4 864 38,000 30,500 68,500 .064 .002 10.2 467 8,893 6,654 15,547 .112 .005 3.5 - 12,281 5,324 17,605 .686 - 22.0 - 4,365 742 5,107 .228 - 1.5 - 10,250 5,050 15,300 .025 - 5.5 4,752 27,400 25,000 52,400 .001 .002 3.4 220 4,111 3,475 7,586 .148 .013 19.0 2,400 41,989 20,485 62,474 .036 .003 4.9 1,150 29,163 10,460 39,623 .102 .005 4.8 4,185 79,639 14,700 94,339 .060 .002 5.5 1,120 68,424 15,809 84,233 .044 .002 20.5 949 38,500 16,500 55,000 .045 .002 9.6 - 6,100 2,400 8,500 .035 - 5.7 - 3,299 5,527 8,826 .024 - 2.5 800 6,000 1,500 7,500 .024 .002 8.3 579 6,600 4,400 11,000 .058 .005 4.2 - 4,000 3,000 7,000 .056 - 17.5 - 5,230 4,624 9,854 .021 - 19.7 - 1,800 2,200 4,000 .062 - 10.0 948 29,700 24,300 54,000 .135 .004 6.3 822 7,600 11,300 18,900 .053 .003 4.8 1,173 42,721 7,543 50,264 .103 .004 10.5 344 13,700 7,900 21,600 .025 .003 19.6 640 10,400 3,140 13,540 .011 .001 2.8 650 2,300 3,700 6,000 .008 .001 5.5 - 7,301 2,441 9,742 .005 - 2.5 3,505 63,000 30,000 93,000 .095 .004 1.5 1,032 33,063 20,030 53,093 .079 .005 14.0 500 10,000 6,000 16,000 .113 .007 6.4 442 5,050 750 5,800 .136 .008 3.2 - 5,013 5,879 10,892 .062 - 2.1 976 55,959 26,401 82,360 .425 .027 6.0 676 9,982 4,555 14,537 .136 .037 29.1 7,703 212,124 116,157 328,281 .043 .004 2.8 3,062 56,254 64,070 120,324 .378 .031 3.9 1,933 245,000 78,000 323,000 .055 .006 30.5 4,981 829,672 166,428 996,100 .350 .019 355.8 - 12,935 12,524 25,459 .025 - 11.1 1,190 269,610 177,161 446,771 .212 .004 12.0 3,765 195,200 153,400 348,600 .033 .007 10.0 - 12,040 21,228 33,268 .026 - 2.0 1,989 84,678 39,539 124,217 .319 .030 10.2 59,761 963,605 371,800 1,335,405 .298 .018 2.5 3,510 200,000 73,000 273,000 .022 .002 9.9 103 5,142 3,095 8,237 .019 .001 0.8 11,393 137,406 38,991 176,397 .111 .012 1.4 181 Basic Transport Data Mid -1969 GNP Population Area per cap. Road Network Length (Km) (million) (Km2) (US$) Paved Gravel & Earth Total Europe, Middle East & North Africa Algeria 13.3 2,381,741 259 51,700 23,900 75,600 Cyprus 0.6 9,251 965 3,397 4,268 7,665 Egypt, Arab Republic of 32.5 999,992 159 9,334 12,232 21,566 Finland 4.7 336,982 1977 21,828 50,041 71,869 Greece 8.8 131,935 842 12,000 23,000 35,000 Iceland 0.2 102,991 1853 51 11,862 11,913 Iraq 9.4 448,706 306 4,700 4,700 9,400 Israel 2.8 20,700 1573 4,000 5,100 9,100 Jordan 2.2 90,177 284 2,448 348 2,796 Lebanon 2.6 10,398 580 5,800 1,300 7,100 Morocco 15.1 447,605 190 20,735 30,745 51,530 Portugal 9.6 91,964 506 27,688 11,280 38,968 Spain 33.0 504,708 820 82,600 50,400 133,000 Syrian Arab Republic 5.9 187,072 255 7,000 11,000 18,000 Tunisia 4.9 150,208 227 8,300 8,992 17,292 Turkey 34.5 767,876 346 38,000 13,000 51,000 Yemen Arab Republic 5.6 194,983 80 430 1,195 1,625 Yemen, People's Dem. Republic of 1.2 290,058 118 416 4,005 4,421 Yugoslavia 20.4 255,784 575 20,623 69,072 89,695 Central America & Caribbean Costa Rica 1.7 50,695 506 1,300 17,300 18,600 Dominican Republic 4.0 48,730 276 5,100 4,900 10,000 El Salvador 3.4 21,392 294 1,200 7,300 8,500 Guatemala 5.0 108,880 353 1,851 8,533 10,384 Guyana 0.7 214,953 340 592 1,932 2,524 Honduras 2.5 112,079 258 900 3,026 3,926 Jamaica 2.0 11,424 548 3,050 1,300 4,350 Mexico 48.9 1,963,739 586 40,333 29,094 69,427 Nicaragua 1.9 139,689 380 1,200 8,800 10,000 Panama 1.4 75,643 662 1,801 4,981 6,782 Trinidad & Tobago 1.0 5,125 885 4,025 2,737 6,762 Venezuela 10.0 911,993 1003 18,000 21,600 39,600 South America Argentina 24.0 2,795,741 1061 29,000 107,500 136,500 Bolivia 4.8 1,098,497 163 724 17,485 18,209 Brazil 92.3 8,511,308 269 42,378 897,235 939,613 Chile 9.6 741,708 509 7,000 54,000 61,000 Colombia 20.6 1,179,227 289 5,300 38,900 44,200 Ecuador 5.9 301,116 238 2,450 12,350 19,800 Paraguay 2.3 406,720 237 810 5,499 6,309 Peru 13.1 1,285,116 331 4,900 40,400 45,300 Uruguay 2.9 185.911 562 2,300 7,000 9,300 Selected Capital Exporting Countries France 50.3 551,458 2460 629,593 853,146 1,482,739 Germany 60.8 248,477 2190 290,271 124,402 414,673 Japan 102.3 369,660 1430 127,188 878,243 1,005,431 United States 203.2 9,374,826 4241 2,533,374 3,395,420 5,928,794 NOTES: All data are for 1969. They are based on various standard publications, revised according to Information available In the Bank, However, many of these figures are inaccurate and should be considered only as orders of magnitude. The weakest information concerns the length of gravel and earth road networks where the distribution between an earth road and a track is 182 Annex 1 Page 2 Railway Route Read Railway Vehicles Length Vehicles Density Density per t,OO0 (Km) Light Heavy Total (Km . Km2)(Km . Km2) Population 4,241 117,000 88,500 205,500 .032 .102 15.5 - 64,000 12,000 76,000 .829 - 126.7 5,110 123,061 25,414 IP8,475 .022 .005 4.6 5,636 695,000 57,800 752.800 .213 .017 160.2 2,568 169,985 97,709 267,694 .265 .020 30.4 - 37,304 6,272 43,576 .116 - 217.9 1,635 61,789 40,963 102,752 .021 .004 10.9 692 155,000 64,000 219,000 .440 .033 78.2 366 15,250 6,589 21,839 .031 .004 9.9 417 129,674 16,239 145.913 .683 .040 56.1 2,051 150,000 55,000 205,000 .115 .005 13.6 3,619 458,100 73,600 531.700 .424 .040 55.4 17,510 1,999,000 683,000 2,682,000 .264 .035 81.3 859 29,466 16,081 45,547 .096 .005 7.7 2,461 59,017 33,130 92,147 .115 .016 18.8 8,030 171,000 135,000 306,000 .066 .015 8.9 ' - 6,900 1,100 8,000 .088 - 1.6 - 24,074 6,389 30,463 .015 - 21.8 16,437 562,500 144,500 707,000 .351 .064 34.7 548 36,000 20,000 56,000 .367 .011 32.9 600 41,734 23,452 65,186 .205 .012 15.5 620 31,300 15,900 47,200 .397 .029 13.9 1,019 33,000 20,500 53,500 .095 .009 10.7 205 14,196 7,924 22,120 .012 .001 31.6 649 14,200 16,600 30,800 .035 .006 12.3 394 60,000 20,000 80,000 .381 .035 40.0 20,207 112,100 524,600 1,636,700 .035 .010 33.5 4,981 14,000 22,000 36,000 .072 .036 18.9 700 41,335 13,785 55,120 .090 .009 39.4 - 67,600 18,800 86,400 1.319 - 86.4 471 498,000 200,000 698,000 .043 .001 69.8 41,148 1,200,000 701,000 1,901,000 .049 .015 79.2 3,626 14,855 19,197 34,052 .017 .003 7.1 35,410 1,537,000 953,900 2,490,900 .110 .004 27.4 9,860 187,483 67,077 254,560 .082 .013 26.5 3,342 141,100 123,200 264,300 .037 .003 12.8 1,121 25,500 35,300 60,800 .066 .004 10.3 522 10,060 13,670 23,730 .016 .001 10.3 2,604 160,000 103,000 263,000 .035 .002 20.1 3,046 142,000 90,300 232,300 .050 .016 80.1 39,660 11,155,000 2,065,000 13,220,000 2.689 .072 262.8 33,660 11,682,556 1,045,297 12,727,853 1.669 .136 209.3 24,140 5,209,319 7,027,538 12,236,857 2.720 .065 119.6 356,619 82,821,000 17,137,000 99,958,000 .632 .038 491.9 an extremely difficult one to make and may vary from country to country. GNP per capita figures are derived from data employed for the World Bank Atlas;'the fact that they are not rounded, as in the Atlas, should not be taken as an indication of greater precision. 183 Bank and IDA Lending for Transport Through 1971 (Fiscal Years, US$ million) Throulh I96h 1964 _ 1965 1966 EASTERN AFRICA Botswana - - 3.6 - Burundi 4.8 - - - East African Community 24.0 - - 38.0 Ethiopia 33.5 - - Kenya - - 7.5 - Lesotho - - - 4.1 Malagasy Republic - - - - Malawi - - - - Mauritius - - - - Rwanda - - - - Somalia - - 6.2 - Sudan 39.0 - - 31.0 Swaziland 2.8 - - - Tanzania - 14.0 - Uganda - - - Zaire 73.0 - - Zambia 23.5 Total Bank 184.3 - - 69.0 IDA 16.3 14.0 17.3 4.1 Total 200.6 14.0 17.3 73.1 No. of Operations 11 1 4 3 WESTERN AFRICA Cameroon - - - - Central African Republic - - - - Chad - - - - Congo, Peoples Republic of - - - - Dahomey - - - - Equatorial Guinea - - - - Gabon - - 12.0 - Gambia - - - - Ghana - - - - Guinea - - - 1.7 Ivory Coast 1.9 - - - Liberia - 3.25 - 1.0 Mali 1.9 - - - Mauritania - - 6.7 Niger - 1.5 - - Nigeria 41.5 - 15.5 32.0 Senegal 1.9 - - - Sierra Leone - - Togo - - - - Upper Volta 1.9 - - - Total Bank 49.1 3.25 12.0 34.7 IDA - 1.50 22.2 - Total 49.1 4.75 34.2 34.7 No. of Operations 3 2 3 4 184 Annex 2 Page 1 Total Total 1967 1968 1969 1970 1971 1964-68 1967-71 - - - - - 3.6 - - - - 0.38 - - 0.38 - - - 77.4 - 38.0 77.4 - 21.2 - - - 21.2 21.2 5.3 10.7 - 23.5 12.6 23.5 52.1 - - - - - 4.1 - 10.0 - 8.0 9.6 - 10.0 27.6 0.49 11.5 - - - 11.99 11.99 - - - 9.3 - - 9.3 - 2.3 0.55 - - 8.5 2.85 _ - - - - 31.0 - - 3.0 15.0 7.5 - 17.0 25.5 - 5.0 - 11.6 - 5.0 16.6 - - 6.0 - 7.0 - 13.0 17.5 - 10.7 - - 17.5 28.2 17.50 13.5 21.20 100.90 - 100.00 153.10 15.79 40.2 19.05 38.38 19.6 91.39 133.02 33.29 53.7 40.25 139.28 19.6 191.39 286.12 4 6 5 8 2 18 25 - 0.55 - 24.2 1.5 0.55 26.25 - - 4.2 4.3 - - 8.5 - - 4.1 - - - 4.1 - - 0.63 1.5 - - 2.13 - - - - 3.5 - 3.5 - - 6.0 - - 12.0 6.0 - - - 2.1 - - 2.1 - - - 1.5 - - 1.5 - - 64.5 - 9.0 1.7 73.5 - 5.8 - - 20.5 5.8 26.3 - - 3.6 - - 4.25 3.6 9.1 - - 7.7 - 9.1 16.8 - - 3.0 - - 6.7 3.0 - - 6.12 - 5.7 1.5 11.82 - - - 35.6 - 47.5 35.6 13.0 - - 2.1 - 13.0 15.1 - - - - 7.2 - 7.2 - - 3.7 - - - 3.7 4.0 5.80 74.10 52.8 33.20 59.75 169.9 18.1 0.55 21.75 26.2 14.20 42.35 80.8 22.1 6.35 95.85 79.0 47.40 102.10 250.7 3 2 9 10 6 14 30 Bank and IDA Lending for Transport Through 1971 (Fiscal Years, US$ million) Through 1963 1964 1965 1966 EAST ASIA & PACIFIC China, Republic of 2.2 - 20.0 - Fiji - - - - Indonesia - - Korea, Republic of 14.0 - Malaysia - - - New Zealand - 7.8 - 42.0 Papua New Guinea - - - Philippines 8.5 - - Singapore - - - - Thailand 79.8 - - 36.0 Total Bank 88.3 7.8 20.0 78.0 IDA 16.2 - - - Total 104.5 7.8 20.0 78.0 No. of Operations 9 1 1 2 SOUTH ASIA Afghanistan - - - - Burma 33.35 - Ceylon - - - - India 594.11 - 62.0 68.0 Iran 72.0 18.5 40.5 - Nepal - - - Pakistan 124.5 106.5 5.25 Total Bank 676.46 50.5 40.50 IDA 147.50 74.5 67.25 68.0 Total 823.96 125.0 107.75 68.0 No. of Operations 28 7 4 1 EUROPE, MIDDLE EAST AND NORTH AFRICA Algeria 50.0 - -- - Cyprus - - Egypt, Arab Republic of 56.5 - Greece - - - - Finland - - 28.5 20.0 Iceland - - - - Iraq - - Israel 49.5 - Jordan - - Lebanon - - _ _ Morocco - - - - Spain - 33.0 65.0 40.0 Syrian Arab Republic - 8.5 - - Tunisia - 7.0 Turkey 16.3 - Yemen Arab Republic - - Yemen, People's Dem. Rep. of - - - Yugoslavia 35.0 35.0 70.0 Total Bank 207.3 75.0 163.5 60.0 IDA - 8.5 - - Total 207.3 83.5 163.5 60.0 No. of Operations 7 4 3 2 Annex 2 Page 2 Total Total 1967 1968 1$9 1970 1971 1964-68 1967-71 - 17.5 31.2 - 15.0 37.5 63.7 - - - - 11.8 - 11.8 - - 28.0 - 34.0 - 62.0 - 11.0 3.5 55.0 54.5 11.0 124.0 - - - - 16.1 49.8 16.1 - - - ~~ ~~~~ ~ ~~- 16.0 498 16.0 - - - 9.0 - - 9.0 - - - - 8.0 - 8.0 15.0 - - - - 15.0 15.0 - 29.0 23.0 - 12.5 65.0 64.5 15.0 46.5 54.2 44.5 133.9 167.3 294.1 - 11.0 31.5 19.5 34.0 11.0 96.0 15.0 57.5 85.7 64.0 167.9 178.3 390.1 1 3 4 2 8 8 18 - - 5.0 - - - 5.0 - - 9.8 - - - 9.8 _ _ - 55.0 - 130.0 55.0 - - - 42.0 - 59.0 42.0 - - - - 2.5 - 2.5 14.5 - 57.5 _20.2 - 126.25 92.2 13.5 - 62.4 61.2 - 104.50 137.1 1.0 - 9.9 56.0 2.5 210.75 69.4 14 - 72.3 117.2 2.5 315.25 206.5 2 -5 4 1 14 1 2 - - 11.5 - - - 11.5 - - - - 13.0 48.5 13.0 - - - - 4.1 - 4.1 23.0 - - - - 23.0 23.0 - - - - 6.0 - 6.0 - - - 14.6 - - 14.6 - 50.0 - - 90.0 188.0 140.0 - - - - - 8.5 - 26.35 - 31.5 7.0 57.85 _ _ - - 1.6 - 1.6 10.0 50.0 30.0 40.0 35.0 165.0 165.0 33.0 100.0 59.35 47.3 173.6 431.5 413.25 - - 8.50 7.3 7.6 8.5 23.4 33.0 100.0 67.85 _54.6 181.2 440.0 436.65 2 2 5 2 8 13 19 1 87 Bank and IDA Lending for Transport Through 1971 (Fiscal Years, US$ million) Through 1963 1964 1965 1966 CENTRAL AMERICA AND CARIBBEAN Costa Rica 11.0 - - - Dominican Republic - - - - El Salvador 24.1 - - - Guatemala 18.2 - - - Guyana - - - - Haiti 2.95 - - - Honduras 18.7 - 9.5 - Jamaica - - 5.5 - Mexico 116.5 40.0 32.0 - Nicaragua 10.2 - - - Panama 13.1 - Trinidad & Tobago - - - - Venezuela 45.0 - 30.0 - Total Bank 236.90 40.0 73.5 - IDA 22.85 - 3.5 - Total 259.75 40.0 77.0 - No. of Operations 19 1 4 SOUTH AMERICA Argentina 48.5 - - Bolivia - - - - Brazil 28.0 - - - Chile 25.0 - - - Colombia 162.65 - - Ecuador 36.6 17.0 - - Paraguay 6.0 - - 4.85 Peru 57.83 3.1 - 42.1 Uruguay 18.5 - 2.2 - Total Bank 338.58 12.1 2.2 46.95 IDA 44.50 8.0 - - Total 383.08 20.1 2.2 46.95 No. of Operations 26 2 1 4 PAST BORROWERS (Pre-1963) Rhodesia 9.5 - - - South Africa 147.8 - Australia 132.33(" -' Japan 160.0 125.0 100.0 25.0 Belgium 30.0 - - - Netherlands 19.0 - - - Norway 25.0 -- Total Bank 523.63(1) 125.0 100.0 25.0 IDA - - - - Total 523.63"'' 125.0 100.0 25.0 No. of Operations 25'2' 2 2 1 GRAND TOTAL Bank 2,304.57('l 313.65 411.70 313.65 IDA 247.35 106.50 110.25 72.10 Total 2,551.92(") 420.15 521.95 385.75 No. of Operations 128(2) 20 22 17 (1) Including $123.1 million for the transportation portion of five general development loans to Australia. (2) Including the Australian loans, referred to in footnote (1), as five distinct operations. 188 Annex 2 Page 3 Total Total 1967 1968 1969 1970 1971 1964-68 1967-71 - - - 15.7 - - 15.7 - 2.8 - - - 2.8 2.8 13.4 - - - 6.0 22.9 19.4 - - - - - 5.5 - - 27.5 - 21.8 - 99.5 49.3 8.6 - - - - 8.6 8.6 - - 20.0 - - 30.0 20.0 22.0 30.3 20.0 37.5 6.0 165.8 115.8 - - - - - 3.5 - 22.0 30.3 20.0 37.5 6.0 169.3 115.8 3 2 1 2 1 10 9 - - 25.0 - 151.5 - 176.5 - - 260- 23.25 - - 23.25 - - 26.0 100.0 45.0 - 171.0 - - 11.6 10.8 - - 22.4 _ - 35.5 32.0 - - 67.5 _~~ ~~~~~~ _ _ 17.0 - - - - 6.0 - 4.85 6.0 - - - - 30.0 45.2 30.0 _ _ _ - - 2.2 - = - 98.1 172.05 226.5 61.25 496.65 _ - - - - 8.00 - - - 98.1 172.05 226.5 69.25 496.65 - - 5 5 4 7 14 100.0 - - - - 350.0 100.0 100.0 -- - - 350.0 100.0 1 - = - - 6 1 205.00 196.10 389.35 516.25 573.20 1,440.10 1,879.9 34.89 51.75 90.70 147.38 77.90 375.49 402.62 239.89 247.85 480.05 663.63 651.10 1,5.59 2,282.52 16 15 34 33 30 90 128 1 an Selected Statistics" on Bank Group Highway Construction Projects, FY 1967-71 Gravel TwooLore Paved Roads H i,gh-Standard Grand Roads Improvement Construct,on Total Paved Roads- , Total Total Length of Roads (km) 2,768 21,480 5,423 26,903 650 30,321 Number of Road Sections 25 120 73 193 19 237 Average Length of Each Section (km) 111 179 74 139 34 128 Total Construction Cost ($ millions) 47 930 788 1,718 628 2,393 Average Unit Cost ($000 per km) 17 43 145 64 966 9 Minimum Unit Cost ($000 per km) for any Road Section 4 5 43 5 140 4 Maximum Unit Cost ($000 per km) for any Road Section 84 883 593 883 2,943 2,943 Total Foreign Exchange Component ($ millions) 33 454 343 797 153 983 Average Foreign Exchange Component Ivy) 70 49 44 46 24 41 Minimum Foreign Exchange Component for any Road Section (%) 65 23 20 20 12 12 Maximum Foreign Exchange Component for any Road Section (%) 75 75 85 85 65 85 Regional Distribution 1. Africa Total Length (km) 2,628 7,133 1,668 8,801 50 11,479 Total Cost ($ millions) 37 239 140 379 8 424 Unit Cost ($000 per km) 14 34 84 43 160 37 2. Asia, Middle East and Oceania Total Length (km) 141 4,404 1,420 5,824 179 6,144 Total Cost ($ millions) 10 185 265 450 356 816 Unit Cost ($000 per km) 71 42 187 77 1,989 133 3. Europe Total Length (km) - 957 546 1.503 135 1,638 Total Cost ($ millions) - 18 167 185 119 304 Unit Cost ($000 per km) - 19 306 123 881 186 4. Western Hemisphere Total Length (km) - 8,986 1,789 10,775 285 11,060 Total Cost ($ millions) - 488 217 705 145 850 Unit Cost ($000 per km) - 54 121 65 509 77 Note: Totals and averages may not check due to rounding. (c Costs given are those estrmated at time of project appra,sal. ito no- Roads w,th more than two lanes and I,m,ted access. A I Selected Statistics on Bank Group Railway Projects, FY 1967-71 (Amounts in US$ million) Track. Marshal- Components Freight Cars Locomot ves Passenger Cars Signali ng and Other Railway Consulting Interest, ng Yards, Parts and I (All Types) (Including (Incl. Ra Icars Communications Equpment, Services Contingencies Country Loan or Credit Structures and Materials Spares) & Traylers) Equipment and Buildings and and and Number Date of Amount 3ic .,;.i for Roll ng -Materials Materials Training Unallocated Agreement E ir . Stock Number Amount Number Amount Number Amount Amount Amount Amount Amount Mali 95 929, 66 9.10 1.66 0.90 107 1.26 9 1.30 22 1.41 0.53 1.32 0.72 Senegal 96 9,/29, 66 9.00 4.19 0.97 16 0.12 7 0.98 19 1.30 0.03 0.67 0.74 - Pakistan 496 5 /26,/ 66 13.50 7.20 4.50 … … … … … …- 1.30 - 0.50 Spain 507 8/4/ 67 50.00 6.82 - 1,150 14.00 40 13.00 115 9.47 2.77 1.67 - 2.27 Korea, Republic of 110 12/18 67 11.00 - - 1,050 10.22 - - - - - - 0.50 0.28 China, Republic of 524 1, 18 68 17.50 1.60 0.40 470 3.60 26 5.10 89 4.60 1.00 - - 1.20 Yugoslavia 531 3'22,68 50.00 31.30 - - - - - - - 2.60 6.00 0.20 9.90 Colombia 551 7/25 '68 18.30 10.50 1.80 - - - - - - 0.40 2.60 1.70 1.30 Guinea 557 9/18/68 32.80,21 27.61 26 0.84 3 0.84 4 0.62 0.47 0.45 1.97 - China, Republic of 603 5 29/69 31.20 2.70 0.50 600 5.80 18 4.00 283 13.20 1.00 1.60 0.30 2.10 Tunisia 606, 150 6/4/69 17.00 7.96 2.33 - - 7 1.77 15 2.08 - 0.76 - 2.10 Pakistan 621 6/26'69 14.50 12.00 - - - - - - - - 1.30 0.70 0.50 India 162 9/24 '69 55.00 2.40 37.80 - - - - - - 4.20 10.60 - - Korea, Republic of 669 183 5 14 070 55.00 2.40 - 2,740 28.40 50 19.50 - - 2.60 - 0.50 1.60 East African Community 674 5 25'70 42.40 23.00 0.60 600'' 4.80 - - - - - 5.90 0.40 7.70 Cameroon 687 6'9 70 5.20 1.96 - 185 2.18 - - - - - - 0.05 1.01 Nigeria 694 6'26,/70 11.200 2.80 - 800 7.00 6 1.40 - - - - - - New Zealand 725 3'1'71 16.00 0.27 2.60 1,200 11.27 - - - - 0.08 0.26 0.50 1.02 Argentina 733 4 28 71 84.00 13.90 54.80 - - - - - - 3.30 - - 12.00 China, Republic of 750 6 11,71 15.00 8.10 - 160 2.34 - - - - 0.75 1.53 0.90 1.38 Guinea 766 6 25/71 1.80(4) 1.40 - - - - - - - - 0.84 - - Spain 772 6/30 '71 90.00 36.20 - 565 12.10 34 16.60 60 10.00 1.50 4.50 0.50 8.60 TOTALS: 649.50 205.97 107.20 9,669 103.92 200 64.49 607 42.68 21.23 40.46 9.68 53.46 E rstimate. '2, Total loan amount s $64.5 million of which railway portion is $32.8 million and balance amount of $31.7 million is for port and townste. Total loan amount iS $25 m I on of ywhich railway portion is $11 2 million and balance amount of $13.8 million is for highways ($12.4 m lion) and ports ($1.4 mill,on). 0 I Total loan amount is $9 million of which railway portion is $1.8 million; balance amount of $7.2 IS for port. Source. appraisal reports prepared between 1966 and 1971. DX :s i 4 Selected Statistics on Bank Group Port Projects, 1967-71 (Amounts in US$ million) Berth(''' Transit Sheds Floating Loan or Credit Construction Cargo Handling Silos Dredging (Except Contingencies(6) Consultant Interest Date of & Ancillary Equipment and Warehouses & Dredging Dredging) and Services & Other During Country Number Agreement Amount Civil Works Quay Cranes Misc. Buildings Equipment Equipment Breakwater Unallocated Training SchemesConstruction Singapore 462 8/11/66 15.01 8.10 2.50 - - 3.01 - 1.00 0.40 - Honduras 463 8/25/66 4.80 3.19 0.50 0.22 - - - 0.03 0.38 0.48 Senegal 493 5/1/67 4.00 1.35 - - 1.85 - - 0.53 0.27 - Guinea 557 9/18/68 31.70`' 7.15 - 9.95 2.10 1.00 - 5.60 3.10 2.80 Tunisia 573 11/29/68 8.50 0.32 0.87 2.50 2.44 - 0.37 1.15 0.85 - Liberia 617 6/20/69 3.60 0.20 - - 1.60 1.15 - 0.43 0.22 Cyprus 628 6/30/69 11.50 2.51 1.74 0.92 1.58 - 2.47 1.45 0.83 - East African Community 638 8/25/69 35.00'3' 15.50 1.55 6.75 0.55 0.75 - 4.00 2.60 3.30 Gambia 187 5/26/70 2.10 0.96 0.24 0.16 0.34 0.07 - 0.25 0.08 - Pakistan S-9 6/10/70 1.00 - - - - - - - 1.00 Malagasy Republic 200 6/19/70 9.60 3.05 0.90 0.74 - - 2.17 1.44 1.30 Nigeria 694 6/26/70 1.40'4' - 1.40 - - - - - - Thailand 702 8/6/70 12.50 7.68 - 2.04 0.37 - - 1.15 1.26 - Cameroon 229 1/14/71 1.50 0.96 - - 0.20 - - 0.22 0.12 - Zaire 255 6/21/71 7.00 - - - 1.00 3.00 - 1.08 1.92 - Brazil 756 6/21/71 45.00 16.67 8.72 3.85 5.88 - - 9.18 0.70 - Guinea 766 6/25/71 7.20'3) 2.00 - 2.40 - - - 1.00 1.30 0.50 Honduras 767 6/25/71 6.00 3.21 0.63 0.16 0.07 0.42 - 0.94 0.57 - Malaysia 774 6/30/71 16.10 11.45 0.57 - - - - 2.95 1.13 Totals 223.51 84.30 19.62 29.69 17.98 9.40 5.01 32.40 18.03 7.08 (1) Includes offshore tanker moorings and also stacking areas, road and railway inside port. (2) Total loan amount $64.5 million, of which $31.7 million is for port and townsite and balance of $32.8 million for railways. (3) Loan amount is $35.6 million. Carryover from Loan 428-EA of $3.33 million brings total to $38.33 million. i4i Total loan amount is $25 million, of which $1.4 million is for port; balance of $23.6 million is for railways ($11.2 million) and highways ($12.4 million). s'5 Total loan amount is $9 million, of which $7.2 million is port portion and townsite; balance of $1.8 million is for railways. (6e Mostly for civil works, berth and shed construction. Source: appraisal reports prepared between 1966 and 1971. d a Summary of World Bank Transport Activities, With Projection through FY1 976 Actual Program Actual'2' Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Bank Sector Missions(' 1 8 6 10 9 13 34 45 UNDP-Financed Sector Surveys and Preinvestment Studies initiated 13 7 14 18 20 30 72 98 Commitments (US$ millions) 480 664 651 750 785 1,816 3,330 4,300 % of Total Bank and IDA 27 30 26 27 28 32 28 28 Number of Countries 29 29 28 31 38 75 81 85 Lending Operations (No.) 34 33 30 37 44 90 178 240 % of Total Bank and IDA 31 28 23 24 24 32 25 24 Projects under Supervision'3' 122 131 141 150 164 110'4' 142'4' 181'4' A1 (1' As of July 1968, the Bank had satisfactory sectoral knowledge of 15 countries. D (2) Including scheduled for FY1972-73 '(3) End of Fiscal Year. (4) Annual average over period. x .0.~~~ GNP PER Relation Between Motor Vehicle Density and GNP Per Capita, 1963-1967 CAPITA $10,000 - - -_ _ _ _ _ $5000 _ _ _ _ - __ _ United States $4000- - - Swed!en $3000- -- _ _ _ _ Norway France $2000 X _ _- _ N thel d nited Kingdom $1000 - 7 - - t f Greece poa,n by $500 Africa Prtugal Philippinles China Jordru Iran, yrian Arab Republic ,Morocco sioo 8~~~~~~~~01,V Oa 01 1 2 3 4 5 10 20 30 40 50 100 200 300 400 500 1000 : MOTOR VEHICLE DENSITY (MOTOR VEHICLES PER THOUSAND INHABITANTS) x Source: Un,ted Nat,ons, Slatistical Year Book 1968 TELECOMMUNICATION Introduction ................................................ 197 International Perspective ..... 198 Special Characteristics .. . 200 Nature of Expansion and Investment Decision Making ..... ..... 201 Impact of Technology .203 Standardization and Local Manufacture ................. 203 Pricing and Demand ..................................... 205 Main Features of Bank Group Operations .206 Finance ............... .. 206 Sector Studies and Research .207 Technical Assistance ..................................... 208 Procurement ............................................ 209 Annexes 1. Telecommunication Loans and Credits FY1962-1971 .... .... 211 2. International Sector Data-Telecommunication ..... ....... 212 3. Summary of World Bank Telecommunication Activities with Projection through FY1976 ........ ...................... 214 I Cr TELECOMMUNICATION SECTOR WORKING PAPER INTRODUCTION Telecommunication networks may be described as the central nervous systems of complex societies, transmitting information and commands between their various parts. Without these facilities, many dimensions of life in interrelated communities, especially govern- ment and economic activity, could not function efficiently, if at all. Wherever public telecommunication services of adequate quality and reliability do not exist, there is widespread recourse to com- munication systems built by business or government units for their own use. These private links are not only much more costly than public systems; they are also only a partial substitute for they do not provide access to the wider community that would normally be con- nected to a well-functioning public system. Limitations of access to such private systems imply serious constraints on efficiency and growth in the unconnected parts of the economy. Telecommunication now encompasses a number of services and facilities for transmission of written messages, voice communication and a variety of other communication modes such as TV and radio programs, data and facsimile transmission, telemetering and tele- command applications. Telecommunication entities provide tele- phone, telegraph, telex, and data facilities to the public, giving cus- tomer service for voice and written communications; in addition, these entities lease telecommunication facilities for private use to meet a wide variety of needs. All these facilities have one feature in common: they use the same methods of transmission from one point to another. Of these, the public telephone service is the most significant in terms of size and coverage and constitutes the basic telecommunication facility. All others are derived by relatively minor additions of equipment to transform the telephone network to other kinds of communication needs. The choice of telecommunication as against the other two basic forms of communication-personal meetings and postal service or other means of transporting written messages-depends on the rela- tive cost, speed, convenience and reliability of each form. Transpor- tation is critical to the speed and cost of personal meetings and also of the postal service. Yet improved transportation, rather than reduc- ing the demand for telecommunication, can create a greater demand due to the widening community of interest generated by increased mobility. The three forms of communication are much more com- plementary than competitive in meeting the total communication needs of a society. The value of telecommunication services to the users is attested to by unsatisfied demand throughout the developing world, even at prices which generate very substantial money surpluses for reinvest- ment and thus probably capture a much larger share of user benefits than in any other public service. It is not surprising to find that the economic return on investment in expansion of telecommunication services, when measured by willingness of users to pay for installa- tion and for use, can approximate 50% in some cases. Such high returns cannot be read as an indication of national expenditure priorities because they cannot be compared with the benefits of expenditures geared to satisfying social needs, which may not be measurable by willingness to pay. We have as yet no adequate technique for comparing intersectoral priorities at the margin: the merits of additional investment in telecommunication vs., say, addi- tional investment in sewerage, education or power. These high re- turns are consistent, however, with what one would expect to be the expanding communication needs of a growing economy: for efficient administration of government and industry, for effective competition in business, for orderly marketing and the rational move- ment of goods, for reduction of industrial and business inventories, for tourism, etc. Also, where literacy is low, telephone conversations are more suitable than mail for settling transactions which require checking and bargaining. Telephones provide amenities which some would regard as of questionable priority in poor countries; these amenities, however, are a by-product of the important economic service provided by a good telecommunication system. Telephone facilities in developing countries are useful mainly for government, business and professional purposes. INTERNATIONAL PERSPECTIVE At the beginning of 1970, the total number of telephones in the world was estimated at about 255 million. The developing countries of Central and South America, Africa and Asia which have 60.7% of the world population and about 16.1% of world GNP have some 17 million telephones or about 6.8% of the total. Except for the United States and Canada, virtually all telephone networks are owned and operated by governments. Table 1: World Distribution of Telephones Percentage Percentage of Telephones Millions ot of World Total World per 100 Telephones Population Telephones Population Developing countries Latin America, 17.5 60.7 6.8 1.0 Africa and Asia North America 124.0 7.9 48.7 55.0 Western Europe 66.5 13.1 26.1 18.0 Japan 23.0 3.6 9.0 22.0 Oceania 5.4 0.7 2.1 27.0 Among the developed countries, the United States and Sweden have more than 50 telephones per 100 of population, Switzerland, Canada and New Zealand over 40, followed by Denmark, Australia and the United Kingdom with densities above 25. Japan, Germany, France and Italy, for historical reasons, have lower telephone densi- ties, but theirs are increasing at annual rates of about 10% or more. Variations between the three developing regions of South and Cen- tral America, Asia and Africa are much wider, and there are great disparities in telephone densities between countries within each of these regions. Some examples are given in Annex 2. Within a country, telephones ordinarily are concentrated in the larger cities. In the industrialized countries, the range of variation in density as between cities and other regions is small. In less devel- oped countries, however, a large proportion of telephones is usually concentrated in a few cities, with little or no service for large popu- lated areas. In developing countries telephone expansion has generally lagged behind development in other sectors, but recently there has been a trend towards faster growth. This trend commenced in the late 1950s in Asia and has now begun to spread to other areas. As a result, the rate of growth of telephones in service is now higher in many de- veloping countries than in the more developed, as shown by Table 2. Table 2: Telephone Growth Rate (Annual percentages) 1959-1969 1969-1970 Central America 4.9 12.9 South America 5.1 7.8 Africa less South Africa 6.0 5.8 Asia less Japan 11.2 11.0 World 6.6 7.3 There is a large gap, however, between supply and demand for telephone services in the developing countries. The waiting list often approaches the number of telephones in service, and the average waiting time for a new connection may be several years. Likewise, traffic generated by subscribers usually exceeds the network's capac- ity, giving rise to poor service, particularly on long distance, where calls are frequently delayed many hours. Typically, investment in telecommunication facilities in develop- ing countries is in the neighborhood of 3% of total public investment. Table 3: Telecommunication and Total Public Investment in Four Countries (Millions ot U.S. dollars) 1965-69 1970-74 Total Tele- Total Tele- Public communi- Bank Public communi- Bank Invest- cations Financed Invest- cations Financed ment Sector % Projects ment Sector % Projects Brazil"' 5,301 133 2.5 -- 7,172 167 2.3 - Mexico 7,939 191"' 2.4 - 11,984 296 2.3 - Colombia 1,450 57 3.9 28 3,190 133 4.2 32 Malaysia"' 1,082I0i 44.3 4.1 49''' 1,567 108 6.8 82 "'Annual averages because historic data not available prior to 1968. "'Estimate for 1965 and 1966. "'The figures are estimates, since the planning periods are different in Malaysia. "'Excludes defense expenditure; figure for 1969 is estimated. "'The project covers the period 1968-1`70 SPECIAL CHARACTERISTICS The most significant feature of this sector is its inherent dynamism. There is a continuous and high demand for facilities, and rapid and continuing technological advances are opening up possibilities for new types of service and for reducing investment and operational costs. Telecommunication systems are unique in that the value of a connection to a single subscriber increases with the number of other subscribers connected to the system. Nevertheless, the prevailing situation in the developing countries is one of gross inadequacy of telecommunication facilities, with obsolete plant and a lack of appli- cation of new technology. These deficiencies are beginning to be corrected in some coun- tries, but greater effort will be necessary to meet the communication needs of economic growth. In many countries this will require wider recognition of the sector's essential role in development, more will- ingness to free telecommunication management from the day-to-day operating constraints of bureaucracy and a more positive approach by management, as well as steps to provide more adequate funds. Nature of Expansion and Investment Decision Making Setting aside complex problems associated with large and sophis- ticated networks, the criteria for making investment and engineering decisions for routine development of telecommunications are fairly well established; but they require a good management with strong engineering and financial groups to ensure that the necessary dis- ciplines are followed and that decisions are taken correctly and in time. The principal objectives of these decisions are to ensure that expenditures are in line with network development priorities and that engineering designs provide the least-cost solutions to meet given requirements and standards of service. The fundamental decisions in planning are those which relate to the basic network structure and which remain valid for 15 to 20 years ahead. These decisions involve the location of the principal ex- changes, the numbering and routing plans, the principles of charging for local and long distance calls, the choice of a switching system and of a signaling system, and the pattern of broadband transmission system for long distance calls. Decisions on each of these different points must take into account the desired rate of implementation of new facilities and the associated economies of scale, and must be reviewed when new technological alternatives become available. Within this long term structural development perspective, system managers must decide on the rate at which services should be mecha- nized. The capital cost of manual exchanges is generally less than for automatic exchanges. However, this differential decreases with the increasing size of the exchange. Also, the continuing operating ex- penses for telephonists are such that, even with low wages, automa- tion is the cheaper solution for exchanges of more than a few hundred lines. Moreover, in developing countries, literacy and other skills required of a telephone operator are a scarce commodity and prob- ably of greater economic benefit in other occupations. For these reasons, automation is now the accepted objective in nearly all countries, for both local and long distance (trunk) service. Subscriber trunk dialing also brings economies through lower costs for setting up connections and for collection of revenues, higher utilization of trunk circuits and improved service. These lead to cheaper calls, and trunk charges can be directly related to the conversation time rather than charging the subscriber in multiples of three minutes, as is usual with manual trunk service. The expansion of a telecommunication network is a continuous process. It consists of thousands of installations annually. The design, dimensioning and timing of these installations are done by "in house" planning engineers who follow established plant practices based on economic analysis designed to minimize the initial and subsequent costs of each class of plant or equipment. Many of the optimizing techniques used widely today in other sectors were developed by the telecommunication industry. In deciding on the components of any program, the first priority is to provide a satisfactory quality of local and long distance service, taking into account the increasing use by existing subscribers and any necessary improvements in the service they receive. The avail- able remaining physical and financial resources are used for new sub- scriber connections, the objective being the balanced development of all network facilities needed for efficient service. As in other utilities, telephone users are connected to a network and can use the service at their discretion. The capacity of the system must therefore be adequate to meet the peak traffic. Inadequate capacity results in congestion, rapid deterioration of service and failure of calls. Delay in meeting requests for telephone connections causes the usage of many telephones in the system to become higher than the system was designed to accommodate. Persons trying to reach these telephones get busy signals, try again, tie up other avail- able circuits and thus add to congestion and failure of calls. Conges- tion also increases costs through increased wear of equipment. A common result of such unsatisfactory public telecommunication service is the establishment of separate private telecommunication networks by governmental, commercial, industrial, transportation and mining groups to meet their individual needs. The total costs to the national economy of such developments are very high, since these requirements can be served at considerably lower cost, and better, by a well-run public telecommunication network. The rate at which a system can be expanded is limited by the ability of the enterprise to provide trained staff to plan, install and operate it. Historically, no country has been able to sustain, over long periods, a growth in new subscribers of substantially more than 15% per year, which requires doubling its facilities every five years, even if demand is growing at a faster rate. A higher rate of increase may be achieved for a short period of a year or so, but at the cost of lowering service standards, with subsequent penalties of network congestion and lower quality service. Impact of Technology Telecommunication has benefited greatly from recent advances in scientific knowledge and technology. The result has been an ability to provide very large blocks of circuits of extreme reliability and high quality over any distance, at unit costs which have been steadily and significantly decreasing. There are further developments in prospect which promise continuing reductions. The new coaxial, microwave and other long distance systems per- mit large increases in the number of circuits on a route with marginal increases in costs; typically a hundredfold increase in the capacity of a route will require about three times the initial investment. Circuits can be brought into commission by the addition of equip- ment at a relatively small fixed cost per circuit. Similarly, satellite communication offers dramatic possibilities for international service with relatively small expenditure on ground stations and rental of satellite facilities. Without being as dramatic or spectacular as the advances in co- axial, microwave and satellite systems, steady improvements have been made in the design and manufacture of telephone switching systems. They are now more reliable, require reduced maintenance, permit savings in the costs of distribution networks and provide con- siderable increase in functional capabilities without increase in costs, so that nationwide automatic subscriber dialing is now not only the better but also the cheaper way of providng service. Standardization and Local Manufacture Traditionally, telephone utilities have shown a marked tendency to adopt standard designs and brands and to adhere to these over long periods of time. This is particularly so in the case of switching equipment (local and long distance), which makes up about 30% of the cost of system expansion. Standardization has the obvious advantages of simplifying techni- cal training and enlarging the flexibility of staff and the interchange- ability of supplies and spares. For switching equipment in telephone exchanges, there are additional special and compelling arguments in favor of standardization. Economies are possible by designing the exchange installation for its ultimate capacity and making the initial installation only to meet the immediate needs; future needs can be met by small increments as and when required. If these economies are to be realized, however, further additions must be duplicates of the initial installation. There are similar, though less significant, advantages in adding to the same types of long distance equipment. A particularly strong reason for standardization of switching equip- ment arises from the problems of interworking with other equipment. In most other sectors, new additions share the function with exist- ing units, e.g., a new electrical generator produces electrical energy exactly the same as existing units, which will pass through any switch or transformer on the system. In the case of telecommunication equipment, however, the interrelationship is different. The language by which one exchange "talks" to another must be consistent, but only recently have the manufacturers of the world progressed to- wards adopting common languages. If different brands or "systems" are mixed, an intermediary "translating" station must be installed so that one manufacturer's switch can "understand" someone else's switch. The planning of interworking equipment, drawing up of bid documents for the new additions, procurement of necessary addi- tions to the existing installations, and commissioning of the equip- ment makes the process complex. As new types are added, the subsequent additions become even more complicated. The penalty for non-standardization is difficult to quantify, since it increases rapidly with the number of different systems in use and also depends on the proportions of each. However, the advantages of standardization are such that most of the developed countries negotiate bulk supply agreements extending over a number of years, and it is the concerted view that a change in the basic switching system cannot feasibly be undertaken more often than every 10 to 15 years. This is perhaps the main reason for the delay in widespread introduction of electronic exchanges, the imminent use of which was forecast more than a decade ago. Standardization and bulk supply agreements have also encouraged the establishment of local telecommunication manufacturing indus- tries. At present, the large developing countries such as Argentina, Brazil and India produce a wide range of telecommunication prod- ucts; and other countries-among them Pakistan, Iran, Israel, Egypt, t)MA Malaysia, Thailand, Singapore, Colombia, Venezuela-produce a part of their needs with foreign collaboration arrangements or in foreign owned local factories. With a few exceptions, these factories arework- ing satisfactorily. Costs of production, net of widely varying import duties and other taxes charged, are generally in line with costs in the industrialized countries. There have been abuses in the name of stand- ardization, but an alert management can protect itself against them. While a judgment on the setting up of any of these industries in a particular country has to be made on its merits, current indications are that more and more countries will find it advantageous to have domestic manufacturing facilities. These industries are not particu- larly difficult to set up, have high value added, and thus can lead to sizable savings in foreign costs. They also serve as a training ground or a base for development of other light engineering industries. The establishment of a switching factory contributes to the growth of manufacture of other light metal engineering products; a cable factory, producing a variety of household electric and other wires as well as power cables (telephone cable, power cable and other wires are often made in one factory), and the manufacture of multiplex equipment helps promote the manufacture of other professional and consumer electronic products. The reasons outlined above for standardization in switching sys- tems suggest that international competitive procurement of this type of equipment should not be carried out in ways which would force borrowers into premature changes in switching equipment. The im- plications for future Bank operations of standardization are discussed further on pages 209-210. Pricing and Demand The tariff structure for telecommunication services varies among countries. In some there is a "flat rate" tariff whereby rental and local calls are covered by a single charge, but in most countries a separate unit fee is charged for each local call. Long distance calls are related to one of a number of mileage categories to which different charges apply; concessional night rate tariffs are traditionally applied to diminish peak loading and increase off-peak revenues. There is considerable flexibility in structuring tariffs to meet par- ticular objectives, such as altering demand by time of day or sub- sidizing one group of users by another; for example, by having different charges for rural and urban subscribers. However, tele- communication demand overall is highly inelastic to price: price increases normally cause only a temporary (about one year) depres- 9nr; sion of demand. In these circumstances, and given the monopoly supply position for this service, most systems set average charges at levels which both fully cover costs and generate substantial surpluses for reinvestment. Because demand so far exceeds capacity in many countries, various forms of rationing have developed, ranging from explicit priorities for certain categories of users on waiting lists (doctors, businesses, etc.) to unplanned black market pricing for telephone connections, which gives a scarcity windfall to individuals already connected to the system. Some entities have resorted to high installation charges to discourage demand for new connections (and collect the scarcity windfall for the entity) and high usage charges to discourage use of existing connections. Since the value of a telecommunication system derives largely from the access it gives to connected customers, these pricing policies frequently conflict with the objective of maximizing system use by the community and ensuring that increases in capacity, where limited, are allocated to best serve community needs. MAIN FEATURES OF BANK GROUP OPERATIONS Finance The World Bank Group is the principal multilateral source of finance for telecommunication development. Through fiscal 1971, the Bank and IDA provided nearly $600 million for this purpose in 34 loans and credits to 21 countries, or more than 90% of the total from all multilateral institutions. The bulk of external finance, how- ever, has probably been provided through bilateral lending agencies and by suppliers in the industrialized countries. Well over 70% of the Bank Group's lending in this sector has been concentrated in the last five years, as summarized in Table 4. Table 4: Summary of World Bank and IDA Lending for Telecommunication, 1949-71 Number of Amount Fiscal Years Loans (US$ millions) 1949-60 4 $ 24.4 1961-66 7 139.1 1967-71 23 428.0 T6tals 34 $591.5 A full list of telecommunication loans beginning with fiscal 1962, together with the total cost of projects the Bank and IDA have ,)nr assisted, is given in Annex 1. As will be seen, the volume increased sharply from $40.2 million in fiscal 1967 to $195.5 million in 1971, while the cost of related projects rose from $72.6 million to $889.6 million. The average size of telecommunication loans in this period was $18.6 million; the largest was to India for $78 million and the smallest to Upper Volta for $0.8 million. Table 5 shows the regional distribution of lending during the same five years. Table 5: Regional Distribution of Bank and IDA Lending for Telecommunication, 1967-71 No. of No. of Amount Total Cost Region Countries Loans of Loans of Projects (US$ millions) Africa 5 4 $ 28.7.' $ 81.6"' Asia 9 13 275.6 1,159.5 Europe 1 1 40.0 95.0 Latin America 5 5 83.7"2' 389.0"'' Totals 20 23 $428.0 $1,725.1 "Two of these loans were to East African Posts and Telegraph, serving Kenya, Uganda and Tanzania. 'One loan of $11.2 million to Jamaica was later cancelled at the request of the borrower. The five-year level of Bank Group lending for telecommunication attained through fiscal 1971 is expected to increase by perhaps 50% in the period 1972-76. (See Annex 3.) Sector Studies and Research In the earlier years, telecommunication projects were considered from time to time only as they were presented for financing, instead of as an integral part of country development programs. As a result, the Bank had developed adequate knowledge of the sector only in those countries where projects had been supported. To overcome this deficiency an enlarged program of sector studies is to be carried out as staff resources permit; of necessity, such work will be limited for several years to countries where lending operations are in prospect, and to the evaluation of sector programs in those with the most acute communication problems. Sector missions for this purpose are not inherently complex, however, since in most coun- tries public telecommunication services are provided by one operat- ing entity and the sector usually comprises a single integrated network. The five-year Bank Group program through fiscal 1973 calls for a total of 31 sector surveys compared with 18 in the previous five years; and for 34 lending operations involving commitments of about $580 million, compared with nine loans totaling $153 million. nn71 A number of important questions require further research, and a research program is in the process of being formulated. One of the major topics under consideration is the relationships between struc- ture of costs and prices in expanding networks. Other work is con- templated regarding the nature of demand for telecommunication services. Technical Assistance In countries where the Bank has been involved in telecommunica- tion development, its participation has often helped operating enti- ties obtain an adequate measure of autonomy in accordance with sound public utility practice, to set tariffs at levels which allow a reasonable percentage of capital expansion to be financed from in- ternal sources, to establish proper accounting systems, and to develop stronger management, planning and engineering capabilities. Good investment planning, designed to produce both balanced and eco- nomical expansion plans, is essential in this sector where network development continuously involves large numbers of small additions to the system. While external finance has been available, much of it has been unrelated to the institutional needs of the operating entities or to the development priorities of the countries concerned. Financing provided by suppliers is oriented, of course, towards the goods financed rather than projects as a whole or the efficient growth of communication systems and the fulfillment of their role in develop- ment. In fact, even within the narrow focus of goods, such financing has been further concentrated on switching equipment and micro- wave systems, largely to the exclusion of cables, which account for about half the cost of most expansion programs. This did not present serious difficulties in some countries where telecommunication administrations had adequate engineering and planning staffs and strong management, with no inhibitions about resorting to international competitive bidding. These, however, were -and still are-in the minority. In many others, expansion programs were unbalanced as well as unnecessarily expensive, with such results as completed exchanges but no cables to connect subscribers. The Bank has also been able to help in the introduction of better procurement practices. International competitive bidding has led to considerable reduction in costs-about 30% on average and much more in some cases. Although standardization of switching equip- ment can have major advantages (see Page2O4),there is no case for retaining an outmoded system or paying excessive prices, as many 9ng telecommunication entities have done. Especially for cables, radio and transmission equipment, the advantages of standardization are minimal. What appeared to be lacking in most countries was a means of relating available supplier finance to the more comprehensive ap- proach which characterizes lending by the World Bank, including conditions which both require and facilitate the necessary technical and institutional improvements. Together with the wider use of inter- national competitive bidding, this is the most important contribution the Bank can make in the telecommunication sector. Ethiopia is a good example of Bank assistance in institution-build- ing. Bank participation commenced in the early 1950's, when the telephone system was run by a government department and con- sisted of a small obsolete system in bad repair. Four loans and 15 years later, the telephone entity is now an autonomous corporation staffed entirely by Ethiopians, with a sound organization and manage- ment, strong planning capability, and a telephone network run- ning efficiently on sound commercial lines. Facilities are expanding by more than 12% per annum, without any financing from the Government. A more recent example is Colombia, where Bank lending has sup- ported substantial managerial and financial improvements, strength- ening of planning capabilities and integration of small, inefficient operating entities. The telecommunication entities of Upper Volta and Nepal, both in an early stage of development, are being re- organized. In Indonesia, the Bank is trying to help improve network planning and balance in investment expenditures. Procurement For reasons already explained, too frequent changes of switching systems would give rise to serious long-term interworking and pro- curement problems and diseconomies in system growth. This special feature of telecommunication systems requires that the normal Bank requirements for procurement through international competitive bidding not be applied in ways that could force the borrower into a premature change in switching systems. Possible approaches to this problem are outlined below. Where a country is fully equipped to produce its own telecom- munication equipment at reasonable prices, the Bank might finance imported raw material and components purchased internationally through open tenders or other methods of international procurement appropriate to the industry, e.g., metal purchases based on inter- national metal market prices. To date this situation has been ad- dressed only in India, where the Bank Group has financed the foreign costs of materials procurement in recent telecommunication loans; the costs of the Indian equipment were in line with international prices. Where a country does not produce its own equipment, the Bank might find itself considering financing for a project covering require- ments for only two or three years, whereas the country might desire to place contracts for a longer period in the interests of standardiza- tion. It might be possible in such circumstances to procure equip- ment under conditions of international competitive bidding and still preserve continuity of supply by placing contracts for the initial period and obtaining options to renew them as needed in later years at prices subject to suitable escalation and review. A somewhat special case is presented by the country with no local production, which has recently standardized on a new type of switch- ing equipment and has entered into a long term supply agreement with a foreign manufacturer. If the supply agreement were the result of international tender before Bank participation and within the past few years, Bank financing might be appropriate. Where acceptable international tendering did not precede the decision to standardize, Bank financing of switching equipment might still be considered in appropriate cases if the cost of such equipment did not dominate the loan and the Bank were satisfied that the prices paid were reason- ably in line with international prices. Thus the Bank might be willing to finance the foreign costs of switching equipment manufactured locally if the manufacturer were efficient, as judged by prices in comparison with those in developed countries. Where there is no local production, the Bank would ex- pect the borrower to use procurement practices designed to take advantage of the international level of prices, with due regard to the benefits of standardization and the disadvantages of too frequent changes in systems. 210 Annex 1 Telecommunication Loans and Credits Fiscal Years 1962-71(1) Amount Total Project Fiscal Costs Year Country Signed (US$ Millions) 1962 Ethiopia 5/31/62 Bank 2.9 6.2 1963 India 9/14/62 IDA 42.0 122.0 1964 Costa Rica 7/10/63 Bank 9.9 12.6 El Salvador 10/7/63 Bank 9.5 13.6 1965 India 7/6/64 IDA 33.0 228.0 1966 Venezuela 12/13/65 Bank 37.0 100.0 Ethiopia 12/28/65 Bank 4.8 10.8 1967 Jamaica 1/23/67 Bank 11.2 18.3 East Africa 2/17/67 Bank 13.0 26.7 Colombia 6/15/67 Bank 16.0 27.6 1968 China, Rep. of 8/2/67 Bank 17.0 50.0 Singapore 9/15/67 Bank 3.0 9.5 Papua-New Guinea 6/28/68 Bank 7.0 15.4 1969 Malaysia 9/27/68 Bank 4.4 49.0 Upper Volta 2/18/69 IDA 0.8 1.2 Pakistan 3/6/69 IDA 16.0 42.0 Ethiopia 6/3/69 Bank 4.5 25.4 India 6/18/69 Bank- IDA 55.0 361.0 1970 Costa Rica 7/10/69 Bank 6.5 9.5 Nepal 11/10/69 IDA 1.7 4.2 Singapore 12/19/69 Bank 11.0 37.1 Yugoslavia 2/20/70 Bank 40.0 95.0 Pakistan 5/22/70 IDA 15.0 35.3 East African Community 5/25/70 Bank 10.4 28.3 1971 Indonesia 7/13/70 IDA 12.8 22.1 I ran 12/18/70 Bank 36.0 149.2 India 5/3/71 IDA 78.0 290.7 Colombia 5/28/71 Bank 15.0 39.1 Malaysia 6/21/71 Bank 18.7 94.0 Venezuela 6/30/71 Bank 35.0 294.5 Four small loans totalling US$24.25 million were mpde in the period prior to 1962. 211 INTERNATIONAL SECTOR DATA-TELECOMMUNICATION TELEPHONES Gross National Product National Principal Cities National International Service(4) Loog Distance COUNTRY Per Ca ita Growth Rate Population Total No. Per 100 Pop. Growth Rate % Staff per 1,000 % of % of Service Direct Access to: 1970 1965-70 1970 1970 1970 1965-70 Automatic tel.)1) Total Pop. Total Tel.(2) Submarine USS % 000's 000's No. % 9 No. % % Quality03) Satellite Cable U.S.A. 4,235 3.4 206,800 115,222 55.71 5.4 99.9 9 45 51 E Atl. Pac. x Sweden 2,788 3.2 8,050 4,307 53.50 4.9 99.7 n.a. 34 43 E - x Switzerland 2,610 2.4 6,350 2,847 44.83 6.0 100.0 4 37 50 E Canada 2,720 2.8 21,700 9,753 42.28 5.5 98.8 n.a. 41 49 E AtL x France 2,288 3.7 51,000 8,114 15.90 7.3 80.1 19 25 50 E Atl. x Australia 2,169 2.4 12,500 3,599 28.79 6.1 89.7 22 64 67 E Pac. Ind. x New Zealand 2,068 1.7 2,950 1,203 40.77 4.6 86.7 18 54 63 F - x U.K. 1,862 2.0 56,050 13,947 24.88 7.0 98.6 20 33 43 E Atl. Ind. x Italy 1,343 4.6 53,600 8,528 15.91 9.0 100.0 n.a. 23 53 E Ati. x Yugoslavia 553 4.2 20,600 623 3.02 11.0 95.9 19 12 49 G Venezuela 977 1.4 10,650 378 3.54 7.8 97.6 18 35 83 G Atl. x Argentina 836 1.0 24,400 1,668 6.83 2.5 92.7 24 47 86 G Atl. Chile 487 1.8 9,800 348 3.55 6.8 87.2 16 35 76 F Atl. Colombia 319 1.4 21,350 546 2.55 7.4 96.8 n.a. 26 77 G Ati. Brazil 258 1.6 93,500 1,787 1.91 7.2 93.3 n.a. 20 66 F Atl. Mexico 566 3.4 50,950 1,328 2.60 12.9 93.0 n.a. 27 75 G Atl. Costa Rica 469 2.1 1,750 56 3.20 20.5 95.9 13 19 80 G - Guatemala 331 1.7 5,150 38 .73 10.6 95.9 20 15 91 P - El Salvador 292 2.1 3,500 35 1.00 9.7 95.5 33 28 84 G - Singapore 753 3.8 2,100 136 6.47 11.5 100.0 14 100 100 - - x Malaysia 358 4.3 11,050 169 1.52 6.9 89.8 52 3 34 E Ind. x China, Rep. of 305 6.5 14,250 339 2.37 18.0 82.2 39 26 72 F Pac. Iraq 275 2.9 9,100 120 1.31 14.1 81.8 47 54 74 P - Papua-New Guinea 221 2.5 2,400 19 .79 16.1 85.4 34 4 78 P - Ceylon 188 2.3 12,550 61 .48 7.7 98.7 n.a. 5 57 P - Pakistan 106 3.1 129,550 193 .14 9.8 78.2 140 7 75 F - India 102 1.0 550,100 1,160 .21 8.8 75.7 129 4 50 F - Indonesia 102 .8 118,250 182 .15 2.7 49.0 86 8 49 F Ind. Zambia 236 3.6 4,300 52 1.20 10.9 98.0 50 13 62 P Ghana 168 -.7 8,850 54 .61 10.3 66.5 n.a. 14 74 F Kenya 134 1.4 10,800 72 .66 7.1 85.1 (7 73 x Uganda 112 1.1 8,550 28 .32 9.2 79.6 49,5) 5 61 G Ind. x Tanzania 82 1.2 13,150 32 .24 8.8 74.4 3 43 x Nigeria 70 -.3 65,650 81 .12 6.2 79.2 n.a. 3 57 F Atl. Malawi 52 2.2 4,500 12 .26 11.4 89.8 n.a. 3 57 P Upper Volta 51 .1 5,400 1 .02 NIL 82.5 107 3 82 P (i) These figures are not strictly comparable. In some cases international, local and long distance services are provided by different organizations in the same country. Telegraph staff may or may not be included and there are wide variations in the use of contract labor for both maintenance and construction work. The number of telephone operators required also varies with the proportion of manuat to automatic service. ,2) E.g. in the U.S.A., 45% of the people uive in the principal cities and have 51% of the country's telephones. n3) E=Excellent. Most cities and centers connected by microwave or coaxial systems with extensive subscriber trunk dialing. G=Good. Principal cities connected by microwave or coaxial systems with subscriber trunk dialing. F=Fair. Principal cities connected by microwave or coaxial systems but calls handled semi-automatically or manually. P=Poor. Little or no microwave or coaxial. Calls between inter-connected cities handled manually. 4) Countries which have satellite earth stations or submarine cable terminals within their borders. Keepya, Uganda and Tanzania are served jointly by a station operated by the East African Posts and Telegraph. (5) Service provided by a common organization. Sources: Population and G.N.P., World Bank Atlas, 1970. Q> Telephone statistics, The World's Telephones, A.T. & T., 1970. Long distance and international service, World Bank. .J~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~~~~~~~~~~~~~ X Annex 3 SUMMARY OF WORLD BANK TELECOMMUNICATION ACTIVITIES WITH PROJECTIONS THROUGH FY1976 Actual Program Actual(') Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector Studies/Surveys(2) 3 6 5 7 10 18 31 35 Commitments ($M.) 80.7 84.6 195.8 96 123 153.2 580 610 % Total IBRD/IDA 5 4 8 4 4 2.7 5 4 No. of countries 5 6 6 8 9 9 21 22 Lending Operations 5 6 6 8 9 9 34 36 %Total IBRD/IDA 4 5 5 5 5 3 5 4 Projects under Supervision(3) 8 11 14 18 23 4(4) 15(4) 25(4' (1) Including scheduled for FY1972-73. 12) In countries where the sector consists predominantly of a single operating utility, which is the case for most tele- communication enterprises, preparation and appraisal of projects includes a sector review, per se. These numbers referto sector reviews prepared in the course of project work and are thus not comparable with the data on reviews of sectors where the protect and sector work is separate. (3) Borrowers with more than one loan are counted only once. (') Annual average. 714 ELECTRIC POWER Introduction . ............................................... 217 Characteristics of the Sector ........ ........................... 218 Load Variation ........... ............................... 219 Power Systems ........... ............................... 219 Timing of Investment ......... ............................ 220 Investment Requirements ....... .......................... 220 Recent Bank Lending in Power ....... .......................... 222 Planned Bank Activities ......... .............................. 224 Lending ............ .................................... 224 Sector Work ............. ............................... 225 Review of Energy Policy ........ ........................... 225 Areas of Challenge ............ ............................... 226 System Planning ........... .............................. 226 Market Forecasting ......... ............................ 226 Minimizing Costs ......... ............................. 227 Pricing ........... ...................................... 227 Village Electrification ......... ............................ 228 Nuclear Power ............ .............................. 230 The Environment .......... .............................. 230 Annexes 1. Chart: Installed Capacity in Developing Member Countries 233 2. Table: World Bank Group Lending in the Power Sector, FY1948-1971 ......... .......................... 234 3. Table: International Sector Data ...... .................. 236 4. Table: Summary of World Bank Electric Power Activities with Projection through FY1976 ..... .............. 238 I ELECTRIC POWER SECTOR WORKING PAPER INTRODUCTION The World Bank' has been a major participant in the develop- ment of electric power facilities in its less developed member coun- tries for more than 20 years. Its first power loan was made in 1948, to Chile. Since then, power lending has totaled $5,300 million', or 27% of the Bank's total commitments. Through these loans, the Bank has been associated with the construction of about a fifth of the estimated 100 million kilowatts of capacity now installed in 41 developing countries and with an additional 16 million kilowatts still under construction. The earliest objective was to overcome endemic power shortages in many countries, resulting in part from the difficulty of importing equipment and materials during World War II. Subsequently, the focus has been on building a solid institutional, financial and tech- nical base for providing adequate service, without which economic activity would be seriously hampered, especially in large urban com- munities. The Bank has developed relationships with 55 power util- ities in 41 countries. They range from small and simple systems, on the one hand, to very large and complex ones, on the other; thus, they encompass a wide range of problems relating to the develop- ment of power institutions and networks. The Bank's heavy involvement in power development results from the fact that the sector offers an especially appropriate channel for the efficient transfer of a significant part of the capital developing countries require. This is because power systems are so capital in- tensive; because their rapid expansion (doubling every five to eight years) commands such a high proportion of total investment in de- veloping countries; and because in most countries so much of the equipment has to be imported. Technical assistance accompanying power loans can help to bring about large economies in the use of resources by improving system planning and operating efficiency. It can also help to mobilize sav- ings through improvements in utility finances. I All references to the Bank include the International Development Association (IDA) unless otherwise specified. A case for reducing the involvement could be made on the ground that the internationally traded equipment which constitutes such a large portion of power investment is exported by most of the industrialized countries and that financing power projects is one of the most convenient ways of mobilizing development capital from these countries. However, supplier finance is often only a par- tial substitute for Bank funds. Quite apart from differences in terms of lending and in prices obtained through bilateral and international procurement, supplier finance seldom embraces the institution- building objectives which have been central to the Bank's lending in this field and which remain important in many countries. CHARACTERISTICS OF THE SECTOR Electric power, with its unique qualities in terms of convenience and ease of handling, is an indispensable element in the energy economy of all modern societies. In common with other forms of energy supply, electric power production and distribution is accom- panied by special problems: availability and transportation of pri- mary fuels; the effects of modifying the regime of rivers; atmos- pheric and water pollution; encroachment on scarce urban land; infringement on the natural beauties of the environment, etc. Like the industrialized nations, manv of the Bank's less developed mem- bers are seeking an answer to the question: "What is the alternative to providing electricity in the amounts demanded?" But most so- cieties tend to plan on the premise that a power shortage is liable to affect economic growth very seriously, and that all major demand must accordingly be met. Some enthusiasts go further. They hold that an abundant, reliable, low-cost supply of electric power, available in advance of market requirements, automatically stimulates industrial development. There is little basis for this view. In some industries, such as electro-chemi- cals and electro-metallurgicals, electricity is a major raw material. In others, however, electric power generally represents less than 3% of production costs in developing countries. The absence of a reliable central supply may hinder industrial development by forcing producers to install their own generating plants at higher cost, but its presence will not necessarily stimulate this development. The corollary is that the demand for industrial power, which is highly sensitive to levels of economic activity, is within broad limits fairly insensitive to the price of electricity. The same is roughly true of household demand; in this case, the capital cost of wiring houses and purchasing appliances is more likely to affect the growth of demand than the price of electricity. Throughout the developing world today, most of the agencies responsible for the production and distribution of electricity are publicly owned: federal, state, or municipal. Because it is generally recognized that competition among electricity suppliers in a given area results in wasteful duplication of investment, utilities almost everywhere have been granted monopolies which are regulated by the government. The scope of regulation varies widely. Load Variation In common with certain other public services, such as transporta- tion and telecommunication, power systems generally have markets in which demand fluctuates widely from one period to another. Depending on the nature of the market-residential, commercial, industrial, or special electro-intensive users-demand will normally be characterized by a fairly sharp peak or peaks lasting only a few hours each day. The power supply agency must try to meet the peak demands as well as the total energy requirement. When demand exceeds the system's capacity, the quality of service deteriorates: voltages and frequency fall, affecting the performance of all user equipment, especially motors. Beyond certain limits, damage occurs to customers' facilities, and ultimately the power supply itself col- lapses. Because these peak demands are for a short time, and be- cause it is difficult to store electrical energy, some of the facilities remain idle part of the day. Much work has been done, especially in the United Kingdom and France, to develop techniques for flattening the demand curve in order to ensure better utilization of expensive plant. Peak demand has been reduced by restructuring tariffs. The introduction of special off-peak tariffs for clock-controlled electric heaters with thermal storage facilities is one method of moving certain categories of heavy demand off the peak. Another approach is simply to refuse to meet the requirements at peak time by cutting off certain con- su.mers, connected separately for this purpose. On the supply side, specially designed peaking facilities, such as gas turbines and pumped storage plants, have been used to save on the investment in generat- ing plant. Power Systems The modern power system consists of an integrated network of generation, transmission and distribution facilities. The proportion of investment in each varies between countries due to geography, ,)1 type of customer, size of system and the degree of sophistication of the economy. Of the total annual investment, the proportion spent on generation can vary from 30% to 50%, on transmission from 10% to 30%, and on distribution from 30% to 60%. As electrical systems grow, the point comes when it is advan- tageous to interconnect them by transmission lines. This allows economies of scale, since larger generating units can be used (in relation to the total capacity of the interconnected system) and standby plant capacity (for use in case of breakdowns) can be con- siderably reduced. Ultimately the transmission lines and generating stations form a "grid" which integrates the electricity systems of a whole country. There are several instances where systems have been connected across political boundaries, e.g., Kenya-Uganda, Tunisia- Algeria, and Argentina-Paraguay. A regional tie connecting Ghana, Togo and Dahomey is now under construction; in due course the line may be extended to Nigeria. Interconnections between Nigeria- Niger, Nicaragua-Honduras, and Nicaragua-Costa Rica are being considered. Timing of Investment Careful studies are required to determine the optimum timing for investment. This implies that detailed technical and economic con- sideration has to be given to alternative strategies for the develop- ment of a power system (see page227). If the commissioning of plant is delayed, it could have serious economic consequences. The risks of delay need to be evaluated at the system planning stage and kept under continual review as the work progresses. Delay in adding a 200 MW hydroelectric unit in a thermal generating system could cause a loss of as much as $250,000 per month in additional fuel consumption. Heavy losses could also be incurred through delays in commissioning large conventional thermal or nuclear units, and sometimes the interconnecting transmission. Investment Requirements The rewards of investment economies are especially large in power. The table below illustrates the size of investment in power in relation to total investment in the public sector for a selected group of countries. There is no reason to expect that power invest- ment will in the future be less than a sixth of total public investment, the proportion observed in the recent past. 220 Power and Total Public Investment in Three Countries (Millions of U.S. dollars) 1965-1969 1970-1974 Total Total Public Electric Power Public Electric Power Invest- Power as % Invest- Power as % ment Sector of Total ment Sector of Total Colombia 1,450 210 14.5 3,190 550 17.1 Mexico 7,939 974 12.3 12,084 1,580 13.1 India 14,105 2,420 17.2 21,200 3,274 15.5 In developing countries, investment in power facilities has been increasing at an annual rate of about 10% and is expected to con- tinue at a high rate in the future. The chart in Annex 1 shows the actual growth of installed capacity for the three major developing areas of the world during the period 1955-1970, as well as a projec- tion to 1985. The volume of investment needed (at current prices) to accommodate projected demand in the developing countries over the five-year period FY 1971-75 is estimated at about $25,000 million equivalent.' The proportion of foreign exchange required for power invest- ments in developing countries is normally high. In general, such countries can make the largest domestic contribution to hydroelec- tric plant, for which labor and materials such as cement and reinforc- ing steel bars figure prominently in the cost. But even here, since heavy construction equipment is usually imported and the contract management experience required may not be available locally, the proportion of foreign costs for hydroelectric work is typically about 50%. In the case of steam electric plant, which is much more de- pendent on outside fabrication, the proportion of foreign costs is typically 75%. The foreign exchange content of transmission and distribution facilities is also quite high in many countries where the domestic capacity for manufacturing such equipment is limited. Power projects are admirably suited to achieve large and rapid capital transfers where there are properly planned sectors, well- formulated projects, well-articulated institutions functioning with qualified and experienced managements, and enlightened regula- tory policies. One of the principal objectives of the Bank has been to help create such conditions. It has been easier to achieve when there has been a continuing relationship between the Bank, the borrower and government agencies. B Based on an average $500 of investment in generation, transmission and distribution for each kW of new generating capacity. 991 RECENT BANK LENDING IN POWER During the five fiscal years 1967-71, the Bank made 79 loans and credits aggregating $2,046 million for electric power development. Total investment for power in developing countries in the same period was on the order of $15,000 million. The Bank has been one of the major sources of external development finance for this sector, as can be seen from the amount of power loans and credits (exclu- sive of guarantees) made to developing countries by the following agencies during the period: Power Loans by Selected Agencies, 1967-1971 i Millions of U.S. dollars) Agency Amount Remarks World Bank/IDA 2,046 Inter-American Development Bank 456 Including Fund for Special Operations U.S. Export-Import Bank 452 Excluding Europe and Japan U.S. AID 401 Including technical and support assistance Asian Development Bank 50 Power lending began in 1969 African Development Bank 7 Power lending began in 1969 French Government Agencies 33 Africa only Significant amounts of financing have been provided by other bi- lateral agencies, such as the Swedish International Development Authority, the Canadian International Development Agency, the Kuwait Fund for Arab Economic Development and the Kreditanstalt fur Wiederaufbau. Some of these funds have been provided through joint or parallel financing operations with the World Bank. These are devices which have facilitated the mobilization of bilateral funds alongside those provided by the Bank. In joint financing, bilateral lenders agree to provide finance, usu- ally within some specified limit, for supply contracts awarded to their nationals as a result of international competitive bidding. In this way, for example, the 10 largest industrialized countries have committed a third to half of the external cost of certain power sector equipment in Mexico, while the Bank has provided the remainder. In parallel financing, equipment or services for different portions of a project are financed separately by the Bank and bilateral lenders, with international competitive bidding applied to the Bank's portion. An example is the El Chocon project in Argentina, where bilateral "buyer" credits are financing the foreign currency costs of generating and transmission equipment and the Bank is covering those asso- ciated with civil works. The average size of Bank loans for power during 1967-71 was $20 million. The largest single loan was one of $125 million in Mex- ico, and the smallest $2 million for a project in Trinidad and Tobago. Loans to individual countries are shown in Annex 2. Power opera- tions accounted for 24% of the dollar volume of all Bank lending in this period, as compared with 29% since 1947. Four countries- Argentina, Brazil, the Republic of China and Mexico-accounted for 42% of the lending in terms of dollar volume; loans to nations in the Latin American and Caribbean region accounted for 53% of the total. The regional distribution of lending for the five year period is summarized below: World Bank Group Power Lending, 1967-71 Amount Number In US$ Region of Loans % of Loans millions % of Total Africa 17 21.5 256.7 12.5 Asia 19 24.1 537.8 26.3 Europe 7 8.9 127.8 6.2 Oceania 2 2.5 38.2 1.9 Western Hemisphere 34 43.0 1,085.0 53.1 Total Power Lending 79 100% 2,045.5 100% The high proportion of lending in the Western Hemisphere re- flects the continuity of relationships established over the years with major borrowers in the region. Moving soon after World War II into some of the world's most important urban centers where the short- age of power was acute, the Bank supported the physical and insti- tutional growth of these systems. With the passage of time, the emphasis shifted gradually from eliminating shortages to achieving basic policy and institutional reforms. In all countries, the adoption of an adequate tariff policy was a key objective. It was not achieved easily. Lending suffered from time to time in cases where agreement on this basic question could not be reached. In most cases, this is no longer an issue of principle; it has increasingly been recognized that only through adequate tariffs (and therefore revenues) can the timely flow of funds necessary for financing large expansion programs be assured. Brazil offers an in- teresting example in this context. It has one of the largest power investment expenditures in the Western Hemisphere, and at the same time one of the most difficult and persistent problems of infla- tion. The wider economic implications of adequate tariff policies and adjustment mechanisms have thus been even greater there than in the other countries of the region. Brazil has established a virtually automatic tariff adjustment process to help mobilize resources for its large investment program in the power sector. Tariff policy has been a major element, but by no means the only one, in the broad institution-building objectives of the Bank and its borrowers. Here are two examples: * In Mexico, a highly fragmented and inefficient power supply has been consolidated under a strong national agency, which has begun to borrow internationally on its own credit. Tariffs have been unified and a consistent revenue policy introduced. Some progress has also been made towards standardizing the frequency of the power supply in the central region. * In Colombia, the four major power utilities have achieved major economies of scale by integrating the largest systems. They have also adopted a uniform tariff policy. The Bank has helped to achieve objectives such as these by con- tributing substantially to the financing of power development in particular countries. In each case it was the stream of loans, rather than any single one, that created new opportunities for achieving, step by step, important improvements in the utility and in the gov- ernment's regulatory and financing role. Intensive project super- vision helped, but experience shows that major improvements can usually be advocated persuasively only within the contractual rela- tionship associated with new lending. PLANNED BANK ACTIVITIES Lending The level of power lending envisaged by the Bank for the five- year period 1972-76 is $2,540 million, about 24% above the actual figure for the previous five years. As a proportion of total Bank financing, lending for power will fall from around 24% to about 17%. However, substantial increases are expected in the number of borrowing countries and of projects. In many countries, the Bank will be assisting the power sector for the first time, and in some of these two or more loans will be made during the same period. Continuity in lending operations is likely to be especially impor- tant for many of the new borrowers, since most will be in the smaller and poorer countries which often confront even more difficult ob- stacles than have been encountered, and largely overcome, by larger countries with more experience and resources. *)) A Of course, the proper role for the Bank in any country's power sector cannot be defined outside the context of what other financing is in prospect for power and what Bank assistance is being sought for other sectors. Sector Work The continual expansion that is characteristic of most power sys- tems means that forward planning is an integral aspect of general operations. Where it is neglected, the results soon become evident in the form of shortages and deteriorating service. The distinction between pre-investment activities, appraisal of specific investment proposals and analysis of system development plans and related policies is thus less sharp than in most other sectors, except tele- communications. This attribute has important implications for the nature of the Bank's sector work. In countries where the sector con- sists of a single operating utility, preparation and appraisal of loans includes a sector review, per se. The Bank has recently studied 17 of these single-agency sectors in connection with lending operations. A growing effort in sector work will be required as the number of loans and of countries involved increases, despite the fact that power financing will represent a smaller proportion of total lending. The level of investment is likely to be so large-of the order of $25,000 million during FY 1972-76-that even marginal economies will be important. The Bank's contribution is now projected at about 17% of the foreign exchange required, on the assumption that the foreign component will be 60%. With these stakes, the range and depth of the Bank's sector work in power will have to be increased. Review of Energy Policy Decisions in the power sector are greatly influenced by policies affecting the use of energy resources-policies, for example, in the areas of fuel pricing, imports, taxation, etc. Member countries have begun to address themselves to the broader issues of national energy policy, and to seek Bank assistance for this purpose. Such reviews are important not merely because of their implications for the power sector, but also because of the impact of energy policy on the many major investment decisions that are sensitive to the relative costs of fuels. Since the Bank lacks direct experience in this broader field, however, its role will be modest for some time. The aim will be to concentrate on assistance to borrowers in drawing up terms of ref- erence for energy reviews and finding the appropriate persons or firms to help carry them out. AREAS OF CHALLENGE The striking institutional improvements achieved by major bor- rowers might suggest that the Bank's contribution in the power sector has now largely been reduced to financing hardware, albeit on better terms than are otherwise available to most countries. This is hardly the case. Not only do most of these borrowers require continued financial assistance, but the scope of their need for collaboration in solving non-financial problems of the power sector has expanded. Areas of growing concern include system planning, which encompasses mar- ket forecasting and the selection of least-cost alternatives; evalua- tion of the structure, as distinct from the level, of electricity tariffs; village electrification; nuclear power; and environmental problems. The remainder of this paper touches on these various aspects in turn. System Planning Traditionally, the first step in planning the expansion of an elec- tric power system is to forecast the demand for electricity at the time of peak consumption (to determine maximum generating ca- pacity) and for the year (to determine the total amount of energy required annually). This forecast is usually made in the context of existing electricity tariffs, quality of supply, standard of security and plans for adding new consumers to the system. The second step is to identify the various possible sector develop- ment strategies (e.g., varying combinations of hydro and thermal plant in differing sizes and sequences) which would both meet the forecast demand and satisfy the system's technical requirements. The third step is to select the alternative which has the lowest present value of capital and operating costs at a reasonable test rate of discount. Revenues do not enter this calculation since they are assumed to be equal for all strategies. Some features of this system planning deserve more attention than they have received so far. Market Forecasting. The extent to which forecasting can be im- proved probably depends most on the type of country being con- sidered, or more precisely on the nature of demand. The Bank has under way a program to encourage borrowers to examine periodi- cally their standards and methods of demand forecasting and to improve its own capacity to evaluate these demand projections critically. A recent study indicates that the degree of error in fore- casting for developing countries is greater than for developed coun- tries, although the standard of forecasting even for the latter needs 226 much improvement. However, there does not appear to be a con- sistent bias towards either conservatism or over-optimism. Some of the more sophisticated forecasting techniques have yielded the largest errors, while some of the simpler ones have yielded the smallest. Further investigation is planned to determine what meth- ods of demand forecasting will give the best results in particular circumstances. Minimizing Costs. Power projects have a long economic life. If different strategies for developing a power system are to be com- pared, projections of annual investment and operating costs under each strategy must be made for a long period-perhaps as long as 50 years. Frequently, different assumptions have to be made about market development and costs before a particular strategy is chosen. The use of mathematical models is especially suitable for these types of studies. As power systems in developing countries become more complex and the choice between development strategies becomes more difficult, the use of such techniques by borrowers is likely to in- crease. They can be extremely useful, provided proper attention is given to the quality of the input data and to the constraints under which electric power systems must operate. Different models are being tested with a view to improving the Bank's capacity to evalu- ate the quality of this aspect of system planning by major borrowers. Even the most sophisticated system planning techniques will lead to wrong choices if the cost data employed for alternatives do not reflect real economic costs. Rational project selection requires that critical attention be paid to the appropriateness of the cost measures used in the comparisons. For example, because of official pricing policies, the actual cost to a utility of foreign exchange or of capital may be well below the real economic cost; in fact, this is often the case. In this situation, the actual cost must be adjusted if a bias toward capital-intensive and import-intensive investments is to be avoided. Unfortunately, there is seldom unanimity on the shadow prices to be used in particular circumstances. One can thus only test, where appropriate, the sensitivity of particular decisions to variations in key costs. Pricing When forecasting electricity demand, the assumption is normally made that demand is price inelastic. While this is probably broadly true, peak demand could probably be reduced and off-peak demand increased by some form of marginal cost pricing, although the pre- cise amount of the reduction or increase might be difficult to pre- dict. The Central Electricity Generating Board (U.K.) reports peak reduction, attributable to marginal cost pricing, of only 600 MW, or 1.5% of the system demand of 39,000 MW'; and Electricite de France reports a reduction of 800 MW, or 3% of system demand of approximately 24,000 MW'. The scope for changing the shape of the demand curve in devel- oping countries through the pricing mechanism needs further in- vestigation. Consultants were engaged by the Bank in 1970 to exam- ine this question in West Pakistan, and their findings are now being reviewed. The basic issue is whether use of mechanical devices to control demand is more appropriate than the pricing mechanism. Care must be taken to avoid cutting the original peak demand in such a way as to form a new peak at a different time of day and season, which could cause greater diseconomies than before, as has already happened in Britain. However, it is probably true that many developing countries could benefit from tariff structures which re- flected more closely than at present the structure of system incre- mental costs. The problem will be considered further after results of the West Pakistan study are evaluated. Village Electrification' When formulating a demand forecast, it is usually possible to distinguish the increasing requirements of existing customers from the demand created by connecting new customers. In urban areas, these rates of growth can be calculated with a fair degree of accu- racy on the basis of a thorough market survey, because the merit of connecting new consumers to a generally well-articulated urban system is seldom in doubt and can be accomplished with relatively little additional investment. The power authority is usually under an obligation to connect all new consumers within a reasonable distance from the main supply. More remote new consumers (mostly industrial estates) are usually willing to pay a capital con- tribution to the power authority. In both instances, the case for providing service to the customer is clear. Justification for electrify- ing villages in rural areas, however, is more complex. Most developing countries have programs of village electrification. Some are quite comprehensive and involve considerable invest- ment. They reflect a mixture of social, political and economic objec- I Both in the period roughly 196S/1969. 2 "Village electrification" is a more accurate term than the more common "rural electrification" because in developing countries it is the electrification of villages and small towns rather than scattered isolated farms which is at issue; the one major exception is ground-water pumping loads for irrigaton. )911 tives, often concerned with maintaining or changing the rural-urban balance in society. In the past, the Bank has not made loans ex- plicitly for such programs, although it has participated indirectly by supporting utilities involved in them; today the Bank is investigating possibilities for giving more active assistance. In one country recently, for example, the Bank included in a power loan funds specifically allocated for village electrification. The Government has agreed to undertake a study of the villages involved compared with others not scheduled for early electrifica- tion. An attempt will be made to measure the social costs and benefits attributable to village electrification. In two other coun- tries, sampling techniques have been used to a limited extent in connection with Bank-assisted power development programs, in an attempt to measure increased demands-both by type of con- sumer, and total-in the electrified villages. The Turkish Electricity Authority has been encouraged to initiate a sampling review of its village electrification program. Through a simple questionnaire, the Bank has also requested information from certain member coun- tries that have had village electrification programs under way for some time. Villages are basically areas of low-density loading. Even if the tariffs there are comparable to those prevailing in the urban high- density loading areas, financial returns on the facilities serving the low-density areas will be low. Financial returns, however, rarely reveal the full benefits of village electrification. Although research is required to determine more precisely which is cause and which is effect, the introduction of electricity into a village appears to foster "cottage" industries, improve agricultural production through irri- gation, and slow down migration to crowded urban centers. In order to help develop criteria for better allocating scarce rural electrifica- tion funds among rural areas, more of the benefits must be quanti- fied and techniques for doing this need to be developed. A program of research is being prepared for this purpose. The aim will be to study the past, present and likely future development of electrified villages in selected countries, along with "control" villages that are similar in as many ways as possible but have not yet been electrified. The evaluation of the benefits of village electrification programs is only one aspect of the general problem of evaluating the eco- nomic benefits of electric power. As yet there is no reasonable basis for measuring these economic benefits, and thus no basis for defending quantitatively the merit of meeting the expected demand for power at the expense of investment in other sectors of the econ- omy. The financial benefits of power expansion can be readily 229 measured. But revenues are an inadequate measure of economic benefits, and an even less adequate measure of social benefits. Financial returns are thus an unsatisfactory investment test; they reveal more about the effect of pricing policies on (monopolistic) suppliers' ability to cover costs than they do about the merits of particular investments. However, unlike the water supply sector, where the large gap between social and private benefits has prob- ably contributed to serious under-investment, the inability to meas- ure benefits does not seem to have resulted in a tendency to under- invest in power in most countries. Nuclear Power Nuclear power is another form of thermal power, and is subject to the same tests of suitability and economy as other forms of generating plant. Until now, nuclear power has not played a sig- nificant role in the economies of the developing countries; only three have nuclear power generating facilities in operation. How- ever, others will have them soon. Two countries have nuclear facili- ties under construction, one is evaluating offers, and several more are in the planning stage. The scope for Bank financing of nuclear plant is being kept under review. The growth of nuclear power in developing countries will be greatly influenced by changes in the capital and fuel costs of com- peting thermal power technologies, which are also evolving. Since the capital costs of nuclear plants are higher by about a third than those of conventional steam installations, they are economic only where fossil fuels are unusually expensive. Even there, the minimum economic capacity of nuclear plant is unlikely to be much less than 500-600 MW. This means that nuclear plant can be used effectively only in relatively large power systems. Operating experience indi- cates that for protection against malfunctioning of the largest single unit in a given power system's network, no unit should have a ca- pacity greater than about 15% of total system demand. That would limit the use of nuclear facilities to systems of 3,000 MW or more. On this basis, only about 10 developing countries could reasonably plan to acquire nuclear facilities by 1975. The Environment As the supply of electricity in the less developed world doubles every five to eight years, the impact of expanding power systems on the environment can be expected to increase substantially. Greater attention and publicity are being focused on the environ- 230 mental consequences of building hydroelectric projects, fossil and nuclear-fueled steam-electric stations, transmission lines, etc. Many utilities in developed countries have learned too late the deep implications of these questions and their impact on public opinion. Henceforth, environmental considerations will have to be evaluated at an early stage when power projects are prepared. 231 Annex 1 Installed Power Capacity in Developing Member Countries 400 _= / / _ 200 100 80 60 _ _ _ _ __ _ _ _ _ 40 _ _ _ _ _ _ _ _ _ C - ___Total 00__ __ 20 (SEM I-LOGARITHMIC SCALE) 1955 1960 1965 1970 1975 1980 1985 ACTUAL PROJECTED 4 Notes: ChartdoesnotincludedevelopingmembercountriesinEurope. Curveshavebeen drawnthroughtheactualorforecastvaluesatfive-yearly intervals. Sources: Actual capacity, U.S. Federal Power Commission; projected capacity, International Atomic Energy Agency. 233 World Bank Group Lending in the Power Sector, 1948-71* (US$ millions) Total FY 1948-66 FY 1967 FY 1968 FY 1969 FY 1970 FY 1971 FY 1967-71 Africa Algeria 10.0 Botswana 2.5 32.0 34.5 Ethiopia 23.5 23.1 23.1 Ghana 47.0 10.0 6.0 7.1 23.1 Kenya 23.0 23.0 Liberia 7.4 7.4 Malawi 5.2 5.2 Mauritius 7.0 Nigeria 112.0 14.5 14.5 Sierra Leone 3.8 3.9 3.9 South Africa 74.0 20.0 20.0 Rhodesia 28.0 Sudan 24.0 24.0 Swaziland 4.2 2.8 2.8 Tanzania 5.2 30.0 35.2 Uganda 8.4 Zambia & Rhodesia 87.7 40.0 40.0 Sub-Totals 405.6 22.8 39.2 47.5 15.1 132.1 256.7 Asia & Middle East Ceylon 41.5 31.3 31.3 China, Republic of 50.0 44.5 55.0 149.5 India 224.2 75.0 75.0 I ran 42.0 60.0 60.0 Japan 178.2 Malaysia 87.5 37.0 11.5 20.0 68.5 Pakistan 30.2 21.5 23.0 44.5 Philippines 58.7 12.0 12.0 Singapore 15.0 10.0 15.0 20.5 45.5 Thailand 78.6 5.0 46.5 51.5 Sub-Totals 755.9 85.5 15.0 82.0 122.3 233.0 537.8 Europe Austria 81.6 Cyprus 21.0 2.8 5.0 7.8 Denmark 45.0 Finland 67.7 Iceland 2.5 18.0 18.0 Ireland 14.5 20.0 34.5 Italy 239.6 Malta 6.0 Norway 95.0 Portugal 57.5 Turkey 25.7 36.5 31.0 67.5 Yugoslavia 118.0 Sub-Totals 759.6 20.8 51.0 5.0 51.0 127.8 'Totals in this table include the power component In some multipurpose loans. The amount for Botswana represents Industrial Infrastructure, including power. 234 Annex 2 Total FY 1948-66 FY 1967 FY 1968 FY 1969 FY 1970 FY 1971 FY 1967-71 Oceania Australia 100.0 Indonesia 15.0 15.0 New Zealand 53.0 Papua New Guinea 23.2 23.2 Sub-Totals 153.0 15.0 23.2 38.2 Western Hemisphere Argentina 95.0 55.0 82.0 60.0 197.0 Bolivia 15.0 7.4 7.4 Brazil 367.6 100.6 48.9 80.0 70.0 299.5 Chile 65.4 60.0 60.0 Colombia 205.8 18.0 18.0 52.3 88.3 Costa Rica 20.8 12.0 12.0 Ecuador 10.0 El Salvador 25.3 5.6 5.6 Guatemala 15.0 7.0 22.0 Honduras 10.3 11.5 11.0 22.5 Jamaica 22.0 Mexico 374.8 90.0 125.0 215.0 Nicaragua 21.2 5.0 15.2 20.2 Panama 4.0 42.0 42.0 Peru 39.0 10.0 17.5 27.5 Trinidad &Tobago 23.5 2.0 2.0 Uruguay 64.0 18.0 18.0 Venezuela 99.0 15.0 31.0 46.0 Sub-Totals 1462.0 205.6 214.2 189.3 382.3 93.6 1085.0 Totals 3536.8 334.7 268.4 369.8 539.7 532.9 2045.5 235 International Sector Data-Electric Power Country Sector Per Capita GNP' Anonua Annual Population Growth Total 1968 Growth 1968 Growth Mid-1969 USE Rate(%) Production Rate(%) kWh Production Rate (%) Country (000) 1969 1964-69 (GWh) 1963-68 Per Cap. (GWh) 1963-68 1. Argentina 23,983 899 1.7 17,851 7.5 744 13,451 8.0 2. Austrailia 12,296 2,306 3.4 44,531 8.75 3,625 43,067 8.75 3. Brazil 90,840 272 1.7 38,181 6.5 424 34,437 7.75 4. Ceylon 12,240 204 4.6 650 7.5 53 5613 8.25 5. China, Republic of 13,800 304 6.5 10,036 14.0 727 9,802 14.25 6. Colombia 20,463 338 1.6 8,1005 6.5 334 7,1006 9.25 7. Costa Rica 1,695 493 4.2 833 10.0 490 779 8.5 8. El Salvador 3,390 292 0.5 582 11.25 172 554 11.5 9. Ethiopia 24,769 70 5.1 361 15.25 15 318 15.5 10. France 50,330 2,396 4.3 117,925 6.0 2,340 94,271 7.5 11. Ghana 8,600 180 -.6 2,589 53.013 300 2,560 60.013 12. India 536,983 109 0.2 49,520 10.3 93 45,170 11.0 13. Indonesia 116,000 112 1.2 N.A. N.A. N.A. 1,6773 1.6 14. Italy 53,170 1,337 4.3 104,011 7.8 1,950 76,660 7.2 15. Kenya 10,506 134 5.6 402 8.9 38 402 8.9 16. Malaysia 10,583 364 4.6 3,076 12.5 290 2,953 13.0 17. Mexico 48,933 612 6.5 22,484 7.9 477 20,074 8.6 18. New Zealand 2,901 2,141 1.4 12,185 6.4 4,200 12,185 6.4 19. Nigeria 63,870 76 -1.8 1,105 4.4 17 1,090 4.4 20. Pakistan 126,353 112 5.5 4,9913 11.5 40 4,9913 11.9 21. Singapore 2,017 802 6.2 1,639 14.9 800 1,639 14.9 22. Sweden 7,978 2,859 2.4 56,237 6.8 7,000 56,2374 6.8 23. Switzerland 6,230 2,697 2.2 30,552 6.8 4,900 26,676 7.4 24. Tanzania 12,926 87 4.5 315 9.5 24 315 9.5 25. Uganda 8,340 114 4.1 731 8.0 88 731 8.0 26. United Kingdom 55,534 1,906 1.3 223,302 5.2 4,000 204,477 5.6 27. United States 203,216 4,242 35 1,435,512 7.3 7,000 1,329,443 7.8 28. Venezuela 10,035 993 1.1 10,814 9.9 1,004 5,1555 6.8 29. Yugoslavia 20,351 571 2.6 20,641 8.8 1,002 18,801 8.6 o GNP figures are based on the implicit exchange rate used by the Bank. To be consistent, these rates have also been used to obtain the "fuel cost" and "average revenue" figures quoted, although in countries which import much fuel this may be misleading with regard to the costof fuel. 2 Figures for sales are for 1968 unless otherwise noted. The figure given is for 1967. (in the case of production, the relevant growth figures are for the years 1963-67.) ' The figure is forthe sector, since public utilitiesare not ditferentiated in the data. 0 The figure isfor 1966, and the growth figurefor 1963-66. 6 The figure is for 1969. Comision Ejecutiva Hidroelectrica del Rio Lempa only. ' Kenya consumption figure (for 1967) exceeds production;the balance is imported from Uganda. 3 Data for 1968. 236 Annex 3 Public Utilities (excluding 'captive' plant) Average Revenue % of Demand Fuel Cost Consumers Per kWh Production (peak) Per Million Installed Capacity (MW, 1969) Number of Per Sold by Pubiic Sumes 1969 Btu Consumers Employee (US Cents)' Utilities (GWh)Z (MW) (US Cents)' Total Thermal Hydro (1969) (1969) (1969) 75 12,0556 N.A. 23.3 3,965 3,437 528 4,000,000 84 2.44 1. 97 N.A. N.A. N.A. 12,1539 8,647 3,136 4,000,000 N.A. N.A. 2. 90 32,000 N.A. N.A. 7,4559 1,633 5,812 N.A. N.A. N.A. 3. 86 612 143 55 247 56 191 N.A. N.A. 2.62 4. 98 10,0026 1,850 50 2,245 1,524 721 2,100,000 200 1.35 5. 90 5,7506 N.A. N.A. 1,870 470 1,400 N.A. N.A. N.A. 6. 94 678 N.A. N.A. 213 170 43 150,400 70 1.73 7. 95 5007 109 25.6 166 96 70 9,20011 N.A. 1.42 8. 88 2880 N.A. N.A. 180 55 125 94,100 46 3.71 9. 80 110,447 19,5003 N.A. 25,6969 12,210 13,486 15,300,00010 N.A. (11) 10. 99 2,6296 378 97 675 87 588 94,500 17 0.62 11. 91 37,8006 10,400 N.A. 14,503 8,147 6,356 N.A. N.A. 2.25 12. N.A. 1,1586 N.A. N.A. 65114 343 308 N.A. N.A. N.A. 13. 74 92,139 16,7703 N.A. 21,702 9,430 11,060 N.A. N.A. N.A. 14. 100 5768 112 29 157 63 94 100,900 44 3.29 15. 96 1,93912 292 664 265 399 398,600 45 2.1 16. 90 16,705 N.A. 33 4,904 2,471 2,433 4,170,000 122 2.08 17. 100 N.A. 2,544 N.A. 3,023 687 2,336 N.A. N.A. N.A. 18. 99 N.A. N.A. N.A. N.A. N.A. N.A. N.A. N.A. N.A. 19. 100 4,7776 1,090 52 1,873 1,127 746 728,000 127 2.43 20. 100 1,6536 321 29.4 524 524 0 245,000 175 2.07 21. N.A. 50,908 9,6753 N.A. 13,7319 3,008 10,423 N.A. N.A. N.A. 22. 87 22,175 4,0903 N.A. 9,5009 560 8,940 N.A. N.A. N.A. 23. 100 3076 69 N.A. 102 54 48 46,100 15 3.32 24. 100 6646 128 N.A. 170 0 170 59,700 N.A. 1.75 25. 91 192,431 40,596 54 55,6819 53,529 2,158 19,964,000 80 1.72 26. 93 1,317,234 238,000 26.1 309,3489 257,478 51,870 55,847,000 N.A. 1.5 27. 48 6,855 1,125 N.A. 1,715 953 762 N.A. N.A. N.A. 28. 91 17,000 3,500 34.25 4,865 1,665 3,200 N.A. N.A. N.A. 29. 'H Data are for residential consumers only. " US cents 1.4 for high voltage supplies. US cents 3.69 for low voltage supplies 12 National Electricity Board only. 13 The high growth rates and production per capita stem from the heavy power requirements of an aluminum smelter which commenced operations in 1967. This accounted for 75% of total power sales in 1969. 14 Asof June 1968. '5 CEL (see 7 above) sells power mainly wholesale. SOURCES: U.N. Statistical Yearbook, U.N. Annual Bulletin of Electric Energy Statistics for Europe, U.K. Ministry of Power Digest of Energy Statistics, U.S. FPC Electric Power Statistics, Annual Reports of Borrowing Countries, World Bank. 237 Annex 4 Summary of World Bank Electric Power Activities with Projections Through FY1976 Actual Program Actual(M Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector Studies/Surveys(2) 6 10 7 8 11 8 42 50 Commitments ($M.) 370 540 538 550 525 1613 2523 2540 %Total IBRD/IDA 21 25 21 20 20 28 21 17 No. of Countries 15 14 17 18 19 38 49 50 Lending Operations 17 15 18 19 22 68 91 102 %Total IBRD/IDA 14 12 14 13 12 24 14 11 Projects under Supervision 67 71 77 83 90 60(3) 78(3) 95(3) ('MIncluding scheduled for FY1972-73 t2ftncountrieswherethe power sectorconsistsof a single operating utility, preparationandappraisalof projects includes a sector review, per se.These numbers mainly referto sector reviews prepared in the course of project work and are thus not comparable with the data on reviews of sectors where the project and sector work is separate. 'iAnnual Average WATER SUPPLY AND SEWERAGE Water and Sewerage in Urban Areas ....... ..................... 241 Sector Features Significant for the Bank Group ...... .............. 243 Main Features of Bank Group Lending ....... ................... 247 Past Loans and Credits .......... .......................... 247 Future Operations ........... ............................ 250 WHO-World Bank Cooperation ....... ..................... 251 Research ............................................... 251 Annex Tables 1. World Bank/IDA Lending for Water/Sewerage, FY1962-1971 . 252 2. Comparative Data Sheet ......... ....................... 253 3. Summary of World Bank Water Supply and Sewerage Activities with Projection through FY1976 ................. 254 WATER SUPPLY AND SEWERAGE SECTOR WORKING PAPER * Urban communities of any size without adequate piped water and sewerage are not viable and thus seriously compromise national de- velopment prospects. The Bank stated in the 1970 Annual Report i. . . the economic growth of urban areas, where most modern indus- try is centered, determines to a very large extent that of the nation; and the problems involved in urbanization will therefore profoundly influence achievement of national development." The purpose of this paper is to elaborate on one dimension of these urbanization problems in developing countries-the need for adequate water and sewerage facilities. The "sector" discussed below comprises the institutions and physi- cal works associated with the collection, treatment, transmission and distribution of potable water, and the collection, treatment, and disposal of liquid wastes -both domestic and industrial. In the de- veloping countries, these systems usually exist only where population is concentrated even though there is need for similar services in vil- lages and rural areas. WATER AND SEWERAGE IN URBAN AREAS Individuals need a minimum amount of water for drinking and preparation of food. Because this minimum requirement is an abso- lute necessity, people not being served by a piped water system resort to alternatives ranging from carrying water long distances or purchas- ing water from vendors to use of heavily polluted ponds or roadside ditches. These alternatives are not feasible physically in urban areas beyond a certain size. Moreover, in the case of the purchase of water from vendors, the price is so high (see table below) that only very small quantities are bought. Metered domestic rate to customers served by Price Charged by Water Vendors the city water system * (US Dollars) Per 1,000 Per 1,000 US Gal. Per m' US Gal. Per m' Dacca ..20- 8.40 1.11-2.22 .35 .09 Kampala 5.84-14.60 155-3.86 .70 .18 Istanbul 3.52 .93 .35 .09 Singapore - - .21 .06 Typical U.S. - - .25-.50 .07-13 Typical German - - .40-.75 .11-19 241 Domestic consumption is only one of the uses of water in urban areas, although it is the major one, typically accounting for 50%- 75% of total consumption. Industrial, commercial and government (schools, hospitals, etc.) consumption is frequently also important. Water is vital for many industrial processes. While there is sometimes the alternative available to large industry of developing a private supply, it is seldom cheaper than a well-run municipal system. Econ- omies of scale are important in water and usually favor a central system when all costs are considered. The different time patterns of industrial and household demand also make a central system the least expensive way to satisfy differing peak demands. Because of the explosive acceleration of urbanization in many de- veloping countries in recent decades, the typical experience is that service which may have been adequate at one time deteriorates as consumers are connected to the system at a faster rate than its capac- ity is increased. Once a system is operating above capacity, the quality of service deteriorates for all consumers connected to it. A good example of this "network" effect is one large South Asian city which has an extensive and very large water system, but one which has lagged far behind the rapid growth of the urban complex and has not been properly maintained.' As a result, an estimated 40% of the water put into the system is lost in distribution. Service is now avail- able for only a few hours a day because the limited water is allocated by rotation to different areas of the city. Consequently, almost every structure has a roof tank and many have pumps to try to suck more water out of the system. The cost of these facilities is enormous, and may exceed the incremental cost of a proper system. A few years ago, when the seasonal rains were inadequate, the very life of the city appeared threatened. Faced with exhaustion of the reservoirs, con- tingency plans were made to move a part of the population out of the city, and most industry was shut down for weeks. These were the expensive consequences of the failure of the water system to keep pace with the growth of the city. Even in less dramatic circumstances, the cost of inadequate water facilities in terms of debilitating diseases, associated medical treatment and reduced productivity is undoubtedly high, but the magnitude of these costs is not easily measured. The studies which demonstrate in developed countries the public health benefits of safe water (reduction in enteric disease) do not separate the benefits of 'This pamphlet offers certain examples of inadequacies in the planning and implementa- tion of projects, or in the quality of service provided by particular water supply and sew- erage systems. Such examples have not been specifically identified; to do so would seem invidious when the basic purpose is only to highlight the nature of the problems that are often founid in this sector. adequate piped water from those of adequate medical services, shelter, food, etc. Public health benefits are thus rarely included in quantitative estimates of the benefits of improved water services. As a result, water projects are often penalized inappropriately when compared with the projects in other sectors for which economic returns on investment reflect more comprehensive measurement of benefits; this may account in some degree for the widespread inade- quacies of water service in most parts of the developing world. Water supply contributes to a city's existence in another way: it provides the only satisfactory method of removal of human wastes. Inadequate central sewerage not only raises problems of public health and aesthetics, but usually leads to higher costs. As a practical matter, every city has to provide some waste disposal system. For example, a large East Asia city, which has no sewerage system, is forced to haul away most of its wastes by truck. This is an increasingly expensive and unsatisfactory solution, not only because the volume is growing with the city, but also because disposal is becoming more and more diffi- cult and labor for this kind of work is scarce. Waterborne sewerage systems are normally the most effective means of urban waste dis- posal, and water and sewerage facilities should thus be considered as part of an integrated system carrying "new" and "used" water. Unfor- tunately, because sewerage systems may cost as much or more than the water systems and because the value of a sewerage system to an individual is often much less apparent to him than it is to society as a whole, sewerage is often given even lower priority than piped water in developing countries. SECTOR FEATURES SIGNIFICANT FOR THE BANK GROUP The water/sewerage sector shares one basic characteristic of other public utilities such as power, gas, telecommunications, etc.: it sells services to the public. But in the minds of many people, it has a stronger "social service" character. It is frequently classified, even in some Bank reports, as "social" rather than "economic" infrastructure, perhaps because no other public utility service more closely affects the daily lives of people. Because of this attitude and the related political sensitivity of water charges, the idea of consumers paying for the full cost of water service is not yet widely accepted in many developing countries. However, in contrast to most "social" services, water and sewerage systems can be capable of paying their way, even after subsidizing reasonable household consumption for the poor. 243 Because most water and sewerage systems are locally managed and locally financed, and therefore normally depend on overbur- dened municipal budgets, the quality of water supply institutions tends to be underdeveloped in relation to the size of their invest- ments and the importance of their service. This local nature of the service and the relatively low quality of management is one of the major barriers to quick improvement in water supply institutions and systems, and an important reason for central governments tending to pay inadequate attention to improving these services. The facts that benefits are not easily measured and thus tend to be underestimated in production-oriented planning, and that well-prepared programs, including rational study of alternatives and reliable cost estimates, are extremely rare, also contribute to neglect by governments. How- ever, this situation is beginning to change as increasing attention is directed to environmental problems and the quality of life in de- veloping countries. Water and sewerage specialists have been calling attention to growing water pollution problems for years; more wide- spread and better sewage collection and treatment remain the only technically feasible and proven solution to these problems. In every country where we have been able to evaluate sector pro- grams, sizable backlogs of needed water .supply and sewerage in- vestments have been discovered. The World Health Organization, (WHO), which probably has the best worldwide knowledge of the sector, has suggested the following water supply targets for the sec- ond UN Development Decade (1971-1980). These targets underline the present inadequacies of supply in developing countries and the magnitude of the effort necessary to correct them. Urban Population Served by: Now By 1980 Household or Courtyard Service 25% 40% Public Standpipes 26% 60% No Service 49% WHO estimates that new capital investment of US$9 billion would be required between 1971 and 1980 to meet these targets in urban areas and another US$1.6 billion to improve rural supplies. If related sewerage works are added, investment requirements approximate US$20 billion. Resources of this magnitude are unlikely to become available for these services during the 1970s.' But with growing awareness of the problems and the increasing resources being com- mitted to the sector, the benefits of improved allocation of talent 'For comparison, for the period 1961-70, IBRD/IDA and IDB (the largest multilateral lender in this sector) helped to finance projects costing US$1,538 million. ?AA and money for these services are large. There is clearly a larger role for the Bank to play here than the very limited role of the past (Bank lending in this sector started only in 1961): to evaluate and help improve the water and sewerage programs of Bank members, to help strengthen institutions, to improve project selection, preparation and execution, and to help finance the more important and complex programs. Two examples of the kind of benefits that the Bank can sometimes help to achieve are noted below: (a) In June 1966, a Bank mission examined a proposed project in a Caribbean city. Prepared by consultants, the project included a dam, long transmission line, treatment works, etc. Its total cost was estimated at US$41 million and it was expected to supply 22 Mgd (83,000 m3/day) to the metropolitan area. As a result of the mission's discussions with the authority and its consult- ants it was discovered that there were possibilities of develop- ing ground-water sources which had not been explored and that utilization of the existing system could be improved. Studies were made and resulted in a project which will supply 17.5 Mgd (66,000 m7/day) at a cost of about US$9 million-about one- fourth of the original cost. (b) In 1967, a mission visited a prominent African city to make a preliminary study of a project being prepared by consultants. It consisted of a dam, diversion tunnels, transmission and treat- ment works, but nothing was included for the distribution sys- tem. It was designed to increase the city's supply by about 12 Mgd (45,000 m3/day.) The consultants' preliminary cost estimate was about US$25 million; a more realistic estimate was about US$35 million. The mission suggested a re-examination of the scope of the project and the demand estimates. Further studies led to a revised project, which now includes the necessary im- provements to the distribution system and is designed to supply 10 Mgd (38,000 m3/day). The revised project is expected to cost about US$12 million, i.e., about one-third of the original esti- mate. Although the project was delayed by the additional studies, the delay will not cause any water shortage in the city since measures for increasing the utilization of existing facilities were also included in the project. There is also great scope for Bank assistance in improving the operations and financial policies of poorly managed and inefficient organizations. Strengthened management and better management systems (accounting, budgeting, maintenance, etc.) often lead to more efficient use of present facilities which can reduce or postpone future capital needs. By insisting on reasonable financial performance tar- gets, it is possible not only to help relieve local officials of some of the agony of politically difficult pricing decisions but also to reduce large burdens on the national budget. However, the achievement of institutional reform is much more difficult than improving the scope and design of projects. The difficulties-largely human and political -have been more than amply illustrated by the Bank's disappointing experiences in some of its early loans, and by the time needed to prepare this aspect of more recent projects. Probably the largest obstacles to adequate levels of investment in the sector have been the inability to prepare projects and mobilize adequate resources locally for them. Encouragement toward more adequate project preparation and pricing may thus be major con- tributions that an outside lender can make to improve the quality of these essential services. Foreign costs of projects in this sector vary widely (5% to 50% in sewerage, 10% to 80% in water) depending on the stage of develop- ment and size of the country as well as the nature of the project. Also, construction periods are relatively long. Thus, these projects are normally suitable vehicles for large transfers of capital only if lenders are willing to provide substantial local currency financing. Comprehensive data on external financing of water and sewerage projects are not available, but there is reason to believe that the bulk of external finance in this field has been provided in the form of suppliers' credits and other bilateral loans for pumps, pipes and treatment plants. The principal multilateral source of funds for this sector has been the Inter-American Development Bank (IDB). Dur- ing 1961-70, IDB lent US$486 million for 85 water and sewerage projects with a total cost of US$1,117 million. This amounted to about 12% of IDB total lending during the period, and was almost 21/2 times the amount lent by Bank/IDA for the same period. For 1970, the amount loaned by IDB was US$29.9 million and US$80.0 million is projected for 1971; future lending in this sector is expected to be about 10% of IDB's total lending. The Asian Development Bank has made two loans to Malaysia totaling US$12.2 million and one loan to Singapore of US$8.3 million for water supply; and the African Development Bank has negotiated one loan to Uganda for US$3.2 million equivalent for 20 small water and sewerage systems. The principal bilateral lender in the field has probably been the United States, although French assistance has also been substantial. The Swedish International Development Association has participated in three Bank/IDA loans (Lahore, Tunisia I and 11) and has made other loans for rural and village water supplies. In summary, investment in this sector would seem to be much more suited to the Bank Group's development objectives than the limited Bank involvement to date would indicate. Adequate water and sewerage is a necessary, but not sufficient, condition for the solution of pressing urbanization and public health problems in member countries; the scope for institution building, and for im- proving investment decision-making and enterprise finance-and the related economies in resource use and increases in public savings- is very large; and the typical problems are such that the Bank can do much to help resolve them. MAIN FEATURES OF BANK GROUP LENDING Past Loans and Credits The Bank's first loan for water supply was made in 1961 to the Republic of China for the Taipei water system. Through FY 1971, 30 loans and credits for US$391.8 million equivalent were made,' of which 17 (US$255.8 million) were made in the last three years. This amounts to 2.6% of all Bank/IDA lending for FY 1961-71. Breakdowns by Bank/IDA and by country are shown in Table I. There is no par- ticular pattern in past lending; projects were considered as they appeared. However, our approach has been less passive recently, as increasing sector work and closer cooperation with UN Development Programme (UNDP) and WHO have begun to bring a steadier stream of projects into the pipeline. With a water supply staff that has grown from 13 to 26 in the past two years, the Bank is now in a position to deal with a larger volume of projects, and this is reflected in the projected level of lending discussed below. Manylessonshavebeen learnedsincethefirstloan in1961 (FY1962). The most important lesson is that lending in this sector is more diffi- cult than had been expected. Although there is nothing intrinsically difficult about the sector, most borrowers have been poorly orga- nized and elementary principles of public utility management have not been observed. Local and national government officials had first to be persuaded of the need for the fundamentals of good project planning such as thorough study of technical alternatives, design criteria appropriate for the country, financial planning, etc. Then the studies had to be organized and carried out before the project could be appraised. Similar problems have appeared in the procurement and construction stages. Thus delays in the projects have been com- mon, both before and after loans were signed, and more intensive su- 'US$36.4 million has subsequently been cancelled. pervision has been required than for power and telecommunications. As mentioned above, institutional improvement is much more difficult to achieve than engineering improvement, particularly where service is bad and deteriorating, as it has been in many projects. Two early projects involved systems which were being ineffectively operated by municipalities, and the credits included several measures intended to help build viable institutions. New autonomous authori- ties were created, management and engineering consultants were provided, and improved financial policies were established. How- ever, it proved difficult to attract competent management, the con- sultants were not very effective, and the new financial policies were not observed. The projects had problems from the beginning, and had to be substantially revised and reduced. Another early loan was made to a national authority which faced many structural and management problems. Again, extensive efforts were made by the Bank and the government during the development of the project to provide the means for improvement. These included, among other things, a management adviser for two years. Despite intensive project supervision by Bank staff, these efforts were not suc- cessful. The project was delayed and costs went up due, inter alia, to poorconstruction planning and execution. Institutional improvements were blocked by militant opposition by two strong unions, top man- agement was inadequate, and financial planning was given little attention. There were recurring legal actions and disputes with con- tractors and suppliers. The management adviser was ineffective. After these problems were brought to the attention of the government, the government conducted its own investigation and then made certain changes in the water authority; however, major problems still remain. Contact with the project is being maintained but no further lending for water is contemplated until the authority has been effectively improved. Experience suggests that, with some important exceptions, local officials and water managers are less accustomed to using modern economic, financial, and management methods, and are more ex- posed to direct political pressure than in higher levels of governments, and in other public utility sectors. This condition is similar to that which existed in some national power sectors when the Bank began lending for power some 20 years ago. In the work with borrowers the Bank has tried to help develop reasonable financial and engineering solutions to the problems of the projects as they appeared. But, for the reasons already discussed, most projects have appeared in response to a crisis rather than in anticipation of needs. Orderly advance planning is rare, in contrast 94A to power and telecommunications where it is more the norm. One serious lack, both for planning and for appraisal work, is reliable historical data. Borrowers have been encouraged to address them- selves to this problem: some of the results are shown in the Com- parative Data Sheet in Annex Table 2. These limited data confirm the scope of the problems in this field: -Urban populations have been increasing by an average 5.9% per annum, reflecting substantial immigration from rural areas; -In one-third of the cities, the majority of consumers are served by public taps rather than house connections; -In half of the cities, more than 25% of water produced is unac- counted for; and, -Average water rates vary by a factor of almost one to ten (US$0.02/mi in Palmira; US$0.19/m' in Yaounde). To correct the negative bias towards the sector in many countries, which reflects in large degree inadequate appreciation of the impor- tance of water and sewerage, the Bank is exploring how best to initiate studies to help illuminate public health benefits. price and income elasticity and other characteristics of demand for water and sewerage services in member countries. Bank initiatives in this sector have been criticized on occasion be- cause they imply concentrating resources on urban areas, which are believed to be better able to help themselves than the poorer rural areas. There are valid reasons for concentrating on urban areas in this field. The need for water and sewerage service increases more rapidly than population in urban areas. Rising population density renders inadequate traditional water sources such as shallow wells, springs, and small streams, and concentrates human and other wastes. At some point, adequate service becomes absolutely necessary to the continued life of the area. Although relatively well-to-do, urban communities must accordingly confront much larger and more diffi- cult problems than their rural counterparts. Also, if a choice must be made, concentrating on urban areas benefits the largest number of people for a given amount of investment. The public health effects are clearly greatest where population is most densely concentrated. Another reason for the Bank's emphasis on urban areas is that the planning of major water and sewerage projects can have an impact on the broader problems of urban planning. This potentially useful leverage has not been used thus far, but could be available for use as knowledge of and capability for dealing with the complex of urbanization problems increase. The recent Bank urbanization mis- sion to Istanbul found that the proposed water, sewerage, and road ')AQ projects could be used as incentives to make the rapid growth of the city more orderly. Well justified rural water projects also merit Bank support, not- withstanding the more difficult institutional and financial problems that would have to be resolved. To date, none has been presented, although a project in Nigeria was under preliminary discussion just before the recent war. Not many such projects are likely to appear unsolicited, however; the reasons which cause governments to neglect urban water and sewerage needs apply with even greater force to rural areas. The above paragraphs are not intended to suggest that conflicting urban/rural claims on scarce development resources should normally be resolved in favor of urban areas. In certain cases, national develop- ment strategies may well be best focussed on the development of rural areas, if only to try to modify migration patterns which aggra- vate urban problems. The proper urban/rural allocation of infra- structure like water, power, schools, etc., can only be defined in a broad context, to which increasing attention is being given in the Bank's country and sector work. Future Operations Bank/IDA activities in this sector are expected to increase sharply in the 1970s. As in the past, most of these activities constitute re- sponses to urgent problems rather than parts of considered sector development programs. But this situation is beginning to change. Aggregate Projected Aggregates FY 1964-68 FY 1969-73 FY 1972-76 Number of Sector Surveys - 23 40 Commitments (US$ million) 127 517 715 Number of Operations 10 34 46 The proportion of "planned" projects based on a knowledge of sector problems in a country and a deliberate evaluation of the merit of Bank involvement is growing and likely to grow further with ex- panded sector studies. Indeed, it is only through a substantial increase in sector work that country economic missions can evaluate plans for this sector in relation to national programs. Such a perspective is needed whether or not works are to be financed by the Bank. The amount of sector work that would be desirable unfortunately exceeds the Bank's capacity to carry it out in the near future. Increas- ing our knowledge of this sector in some 70 countries where it is 250 needed is obviously not a realistic short-term objective. Our actual goal is to do what is necessary to support planned operations and, in the short run, to provide sector support for Bank country economic work in the countries with the most acute sector problems. Sector reviews in other countries will have to be made more gradually. WHO-World Bank Cooperation For several years, the UNDP has financed pre-investment studies for water and sewerage projects, with WHO as executing agency. The Bank expresses "special interest" in those projects for which Bank lending appears likely, and follows the progress of the UNDP/WHO project in order to help assure that the project is being prepared in a manner likely to facilitate a prompt financing decision. From this experience it has been observed that there is a need to take a more investment-oriented approach to the execution of UNDP projects in the water supply and sewerage sector. As a result, there have been discussions over the past year between WHO and Bank staff aimed at establishing more effective staff cooperation in carrying out the large work program in prospect. In November 1970, WHO established a new pre-investment planning unit in its Environmental Health Divi- sion at WHO headquarters in Geneva to guide all WHO's pre-invest- ment activities in this sector. Since much of the work of this unit could, in due course, replace work that would otherwise need to be done by Bank staff, cooperative arrangements have been negotiated under which the Bank will contribute to the expansion of this group. These arrangements have been approved by the Executive Directors and the Board of Governors and work under the Agreement will begin soon. Research The most pressing problem is to develop better criteria for allocat- ing scarce funds among competing projects in this sector. Progress toward this goal depends on developing better measures of economic benefits, which largely depend in turn on developing better data. The Bank is starting to carry out research to this end and expects to intensify it. Other research objectives are to improve water demand analysis and forecasting and to identify pricing policy options for water and for sewerage. Annex Table 1 WORLD BANK/IDA LENDING FOR WATER/SEWERAGE FY 1962-1971 (US$ Million Original Amount) Country Project Bank IDA FY 62 CHINA Taipei Water 4.4() JORDAN Amman Water 2.0(2) FY 63 NICARAGUA Managua Water 3.0 FY 64 PAKISTAN Dacca Water/Sewerage 26.0() PAKISTAN Chittagong Water/Sewerage 24.0(4) JORDAN Three Cities Water 3.5(0) FY 65 PHILIPPINES Manila Water 20.2(6) SINGAPORE Singapore Water 6.8 FY 66 BURUNDI Bujumbura Water 1.1 VENEZUELA Caracas Water 21.3(7) FY 67 PAKISTAN Lahore Water 1.7 FY 68 SINGAPORE Singapore Water 8.0 COLOMBIA Bogota Water 14.0 FY 69 SINGAPORE Singapore Sewerage 6.0 MALAYSIA Kuala Lumpur Water 3.6 TUNISIA National Water 15.0 JAMAICA Kingston Water 5.0 CAMEROON Yaounde/Douala Water 5.0 FY 70 GHANA Accra/Tema Water 3.5 COLOMBIA Cali Water 18.5 TUNISIA National Water 10.5 FY 71 KENYA Nairobi Water 8.3 CYPRUS Nicosia Sewerage 3.5 CYPRUS Famagusta Sewerage 1.9 BOTSWANA Lobatse-Gaberones Water 3.0 COLOMBIA Palmira Water 2.0 COLOMBIA Bogota Water 88.0 YUGOSLAVIA Ibar Water(8) 45.0 BRAZIL Sao Paulo Water and Pollution Control 37.0(9' NOTE: In addition to the above, a loan to Iceland of US$2 million was made for a hot water supply system used for domestic heating, and a loan to Malta of US$7.5 million for power included US$1.87 million for desalination equipment. () US$0.4 million cancelled. (°) US$0.5 million cancelled. (3) US$12.8 million cancelled. (4) US$17.0 million cancelled. (s) US$1.0 million cancelled. (6) US$0.6 million cancelled. (7) US$4.2 million cancelled. (8) Multipurpose project for water, power and irrigation. (0) Two loans. $22.0 million tor water supply, $15.0 million for pollution control. Annex Table 2 COMPARATIVE DATA SHEET - - - - - - - - - - ZZ-o 0 - - - -0 - - 00 -- I~~~~~~~~~~~~~ = _ O 0 !<, 4 < < 0 < <- - - -4 < < E X -40,R Cr- 0°o1 11I 0r 0 000ISI 0) - 04 --~ ~~~ < 0 - --4 O 0 . _ = I(° DOOcno ss 0 ~ ~ ~ - - - - - - - - - - - o~~~~~~~~~~~~~- E~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~ 0 E <~~~~~ I <= e = o0 00 0 0(0 4 ) 0o 0 (000 - _40 (00 04 R4 04 co oo04 b =00 sx00 (0 0 0( 4 404 000Oa wc 0040404L00404(0to> 040 (0( 04 E _ o o O Or.004- o o _ x X s , _ E < < a X w 3 0040440400400 <<0 (004 004<00Z <<0 0~~~~ < --- ~ c X m O ~~u CD w eZL )r 253 Annex Table 3 SUMMARY OF WORLD BANK WATER SUPPLY & SEWERAGE ACTIVITIES WITH PROJECTIONS THROUGH FY 1976 Actual Program Actual(M) Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector Studies/Surveys,27 - 3 4 6 10 - 23 40 Commitments ($ M.) 34.6 32.5 188.7 123.7 138.0 126.6 517.5 715 % Total IBRD/IDA 2 1 8 4 5 2 4 4 No. of countries 5 3 6 5 13 7 24 35 Lending Operations 5 3 8 5 13 10 34 46 % Total IBRD/IDA 5 3 6 6 7 4 5 5 Projects under 13 14 19 28 37 6 '4 2201 45(4) Supervision(3) (') Including scheduled for FY 1972-73. (2) At the beginning of FY 1969, we had a reasonably adequate knowledge ot the sector in four countries. (3) Borrowers with more than one loan are counted only once. () Annual average over period. 94 EDUCATION Trends in Educational Development ....... ..................... 257 General . ................................................ 257 Quantitative Expansion ......... .......................... 258 Problems of Quality: the efficiency and productivity of education systems .......... ............................. 259 Education and the Labor Market ...... ..................... 260 Financial Constraints ......... ............................ 263 Dilemma for the Government: how to allocate resources .... ... 264 Review of World Bank Policies and Operations, 1963-1971 .... ..... 265 Policies . ............................................... 265 Operations .............. ............................... 266 Organization, Procedures and Criteria ...... ................. 268 World Bank Education Policies and Operations in the 1970s ........ 271 Possible New Areas of Lending .......... .. ................ 271 (a) Non-formal education and training ..... ............... 271 (b) Educational radio and television ...... ................ 272 (c) Programmed learning ....... ........................ 273 (d) Teaching materials and equipment ..... ............... 273 (e) Business administration and management ..... ......... 274 (f) Management of education systems ..... ............... 274 Project Design, Sector Analysis and Project Sequence ... ....... 274 Projections of Lending Operations in FY1972-1976 . ......... 276 Analysis of loans by country categories ........ ............ 276 Analysis of loans by mode and outlay ......... ............ 277 Research and Project Evaluation .......... .. ................ 278 Conclusions ....................... ........................ 279 Annex Tables 1. Enrollment Growth Rates, 1960-1966, and Comparative Enrollment Ratios, 1968, by Regions and Levels .... ......... 281 2. Inefficiency in Primary Education ...... ................... 282 3. Public Expenditure on Education as a Percentage of the Budget and National Income, 1960 and 1965 ..... .......... 282 4. World Bank/IDA: Education Loans by GDP/Capita at Time of Loan, FY1963-1971 ...... ..................... 283 5. Analysis of World Bank/IDA Education Lending, FY1963-1971 283 255 6. World Bank/IDA Education Projects: Student Places Provided or Improved and Boarding Places and Staff Housing, FY1963-1971 .............. ........................... 284 7. World Bank/IDA Education Projects: Cost of Physical Facilities per Student . ........................................... 285 8. World Bank/IDA Education Projects Approved as of July 1, 1971 ................ .......................... 286 9. Comparative Education Indicators ...... ................. 288 10. Summary of World Bank Education Activities with Projection through FY1976 ............................. 290 EDUCATION SECTOR WORKING PAPER * This sector working paper considers World Bank lending programs over a perspective of five years and lending policy over a period of approximately a decade. It consists of three parts: (a) a brief review of trends in educational development; (b) a review of Bank policy and operations during the fiscal years 1963-1971; and (c) a projection and recommendations for Bank policy and operations responding to the changing situation of the '70s. All references to the Bank in this paper are to be deemed to refer also to the International Development Asso- ciation (IDA) unless the context otherwise requires. TRENDS IN EDUCATIONAL DEVELOPMENT General Since about 1950, the movement for political independence, the quickening pace of economic development in the less developed countries with its concomitant emphasis on science and technology, and the related population explosion have provided a powerful im- petus and demand for education. The resulting expansion of educa- tion systems all over the world has been without precedent in history. The achievement is remarkable although far from complete. The popular demand for education continues without slackening but there is mounting evidence, on the one hand, that this rapid ex- pansion has created a new set of formidable problems and, on the other, has failed to achieve many of the benefits which were confi- dently expected of it. Many persons have agreed with Philip Coombs' characterization of this situation as a world crisis.' On all sides it is accepted that the continued development of education during the 1970s must be along very different lines from the largely "linear ex- pansion" of the 1950s and early 1960s. The description of the educa- tional development sector which follows attempts to define not only the salient characteristics and problems of educational development at the present time but also to describe them as continuing trends, to which Bank policy must continue to respond and adapt. 'Coombs: The World Educational Crisis, Oxford, 1968. I- 7 Quantitative Expansion The increase in school enrollments at all levels and among almost all the less developed countries is the most dramatic feature of edu- cational development over the past decade and a half. For the period 1960-1966 in developing countries as a whole enrollment growth rates were 42% at the primary level (reflecting an earlier start), 80% at the secondary level and 93% at the higher (mostly university) level (Annex Table 1). In absolute terms this growth added nearly 55 mil- lion primary students, 15 million secondary students and 2 million students in higher education, or an average of 12 million additional students each year. One result of these increases was to achieve in 1968 primary and secondary enrollment ratios, respectively, of 40% and 15% in Africa, 55% and 30% in Asia, and 75% and 35% in Latin America (Annex Table 1). Two major forces lie behind this dramatic increase in enrollments. First is the strong social and political demand for education. Educa- tion is now universally accepted and demanded as essential to par- ticipation in the development process and to improvement in the condition of individual life. The fact that the individual economic payoff is often illusory does not diminish the demand since there are important social benefits and in any case education is essential to maintaining one's relative position in the social order. Because most curricula are designed to advance the students to the next level, there is an additional escalating force within the educational pyramid. As enrollments grow at one level, the demand increases for accommo- dation at the next level. The second force pushing up enrollments has been the growth of population and, more particularly, the relatively recent increase of population in the developing countries which has resulted in an ex- traordinarily high proportion of young people. The United Nations estimates that in 1970 the percentage of total population under 15 years of age was 27% in the advanced countries and 41% in the less developed countries. One begins to see here the roots of the dilemma which developing countries face. Two significant features of this expansion should be noted. First is the narrow concentration by governments and also by external assist- ance agencies upon the improvement of human resources for the modern economic sector and therefore upon formal education sys- tems. Conversely, all too little attention has been given to the needs of rural populations living in the traditional or transitional economic sectors who, for some years to come, will be the majority in most developing countries. In this connection, President Julius Nyerere has noted of Tanzania: "Although only about 13% of our primary school children will get a place in secondary school, the basis of our primary school education is the preparation of pupils for secondary schools ... The same process operates again at the next highest level, when entrance to the university is the question at issue." It is relatively easy to diagnose the problem and enumerate the obsta- cles to educating rural populations; more difficult, but imperative, is to determine more clearly the characteristics of different learning clienteles, the content of learning which will produce quick and substantial returns and the most appropriate ways to reach hitherto inaccessible learning clienteles. Secondly, despite the very impressive gains in the numbers of young people enrolled in school, enrollment ratios (percentage of the relevant age group enrolled) remain low in a number of countries. Among 51 countries in which the Bank has had recent lending opera- tions in education, the median primary enrollment ratio is 71%; the range is 6% (Somalia) to 99% (Greece). At the secondary level for the same group of countries in the mid-60s the median enrollment ratio is 13% and the range is 2% (Tanzania) to 54% (Ireland). Twenty-five percent of these countries have less than 8% (Annex Table 9). Moreover, the composition of these enrollments remains seriously unfavorable from the point of view of economic development. At the secondary level in 50 countries for which data are available, the median percentage of total students studying vocational courses is 11% and in higher education the median enrolled in courses of engineering, medicine, science and agriculture is 21%. Since almost all countries have a public commitment to achieve universal primary education by no later than 1980, it is likely (whether they achieve it or not) that the '70s will witness a continued pressure for increase at this level and, as a consequence, at the secondary level as well. Hopefully, there will also be a shift toward higher percentages of students following pre-vocational and vocational courses, although as indicated below, there are limits to the ability of the modern sector to absorb the output of technical schools. Problems of Quality: the efficiency and productivity of education systems As enrollments have expanded, it has been difficult, and more often impossible, for education systems to maintain the levels of quality and efficiency of the pre-expansion era. Facilities-classrooms, equip- ment and teaching materials-have not kept pace with the expanding number of students. Teacher training has lagged and the qualification 259 and experience of the teaching force have declined. The most serious gap, affecting everything else, has been in the management capability -organization, planning, evaluation, supervision-needed to meet the challenge of expansion. The effects of all these factors are reflected in the dropout and re- peater rates. In perhaps half of the developing countries, less than half of the students who enter primary school complete the cycle. The resulting inefficiency can be seen in selected cases drawn from the analysis of Bank project appraisals. In the Ivory Coast it requires an average of 12.5 pupil years to produce one graduate of the 6-year primary cycle and 21% of the total primary budget is spent on stu- dents who drop out in Grades 1-3. In El Salvador the comparable figures are 15 years and 37%; in Ethiopia they are 14 years and 32% (Annex Table 2). Of equal importance has been the persistence of institutional forms, school system structures, teaching methods and curricula de- rived largely from European practice of an earlier era and to a great extent irrelevant to present day needs. In different ways, but with comparable effect, both the former colonies and the historically independent developing countries suffer from a lack of responsive- ness to changed conditions. Both types of country have inherited or adopted educational systems designed for an elite-in the historically independent countries usually a landed or commercial upper class; in the former colonies a cadre of civil servants. In both cases, with the change to a mass learning clientele, a substantial proportion of the students in school are being miseducated. The content of primary and secondary courses, heavily dominated by the questions on qualifying examinations, is still drawn to a great extent from the developed world, which is remote from the experience of today's student, espe- cially the rural peasant child, still living in a primitive condition of life. The heavy reliance on examinations encourages rote learning of irrelevant information. Even in technical courses adherence to stand- ards and practices set in the developed countries inhibits the educa- tion of students to solve the problems of their own environments in more appropriate ways. The fact that these practices are often the choice of local people does not diminish their bad effects. Education and the Labor Market Matching educational growth with the manpower needs of devel- oping economies is likely to become more complex in the '70s. Aspirations of the younger generation and the related expansion of school systems have often greatly exceeded the opportunities for productive employment in the modern sector of the economy, re- 260 sulting in growing unemployment of the educated. With hardly 10% of the age group in secondary schools and less than 1% in universi- ties, some education systems are already approaching the point of producing graduates in numbers exceeding the effective demand of employers. A long-term development strategy would no doubt call for a continued expansion of the education system to raise the very low educational attainment level of the working population although the short-term reality of increasing unemployment in the rapidly growing younger age group often reduces the economic benefits of such expansion. When the unemployment problem is of a cyclical or temporary nature, it should be ignored in the development of long-term edu- cational growth targets. Persistently growing unemployment among the educated at progressively higher levels, however, would seem to indicate structural imbalances, which cannot be ignored. In such cases, continued investments in the expansion of education systems without major reforms could become both economically and socially unprofitable. The situation calls for three kinds of action: a continu- ous reassessment of the economic development, potential and effec- tive manpower demand of the countries concerned, improvements in the functioning of labor markets (particularly the system of in- centives and disincentives) and reorientation of the education and training systems. Our knowledge of the problem of unemployment amongst those with advanced education and training (which is of recent origin in a number of countries) is spotty and inconclusive; research in this field deserves high priority and is described in more detail at the end of this section. Closer attention to the functioning of labor markets is particularly important because the vexing unemployment problem is com- pounded by labor market imperfections which obstruct the effective utilization of a country's available manpower resources. Often, un- employment among the educated is accompanied by persistent short- ages of qualified manpower in fields where work incentives are unattractive. Examples of such pockets of "scarcity in the midst of plenty" are industrial technicians, agricultural extension workers and teachers. Educational development in the '60s was no doubt imbalanced when compared with these specific manpower needs. Even in those cases, however, where attention was paid to industrial, commercial and agricultural training the result frequently failed to meet employ- ers' needs. Formal technical education and training in schools were often out of step with informal, on-the-job training and employment opportunities. The systems of wages and other incentives are often 261 not conducive to an optimum deployment and utilization of man- power. Public sector salaries, for instance, are normally insensitive to demand and supply. In Africa the high level of these salaries, com- bined with budgetary constraints, has resulted in a situation where the number of civil service posts expands slowly but at the same time stimulates a demand for education and training for these posts be- yond the saturation point of effective demand. In Latin American and Asian countries the low level of public sector salaries may ease the budgetary constraints but it has often made the civil service, including the teaching professions, unattractive to able and qualified school and university graduates and sometimes created an oversized body of less qualified staff. Within the public sector, the rigidity of salary scales often prevents adjustments for those occupations where scarci- ties persist, particularly for jobs where working conditions are un- attractive or competition for specific technical skills is severe. A reorientation of the education and training systems, with greater emphasis on vocational education and on non-formal training for agriculture and industry, will be required to redress present imbal- ances. A frequent objection against public investments in vocational training is the fact that most public training schemes have been poorly adjusted to employers' needs and have failed to guide students into the vocations for which they were trained. Since vocational training is considerably more expensive than general education, this situation can therefore lead to misinvestment. From this the conclusion is often drawn that vocational training should be left entirely to in-plant train- ing by industry. There is considerable validity in this view but in many countries it is unlikely that the need for vocationally trained man- power could effectively be met by industry itself. The small scale of the majority of industrial undertakings in developing countries, often working with outdated equipment and without adequate know-how of modern production techniques or interest in staff training does not provide the proper basis for modern vocational training schemes. Industrial development would therefore require that the supply of technically trained manpower, which is greatly inadequate in many countries, be strengthened by institutional public training. To avoid the mistakes of the past, however, there is an urgent need for close cooperation between public training institutions and private industry, e.g. through the establishment of apprenticeship training schemes. Examples of such cooperation are the SENA and INACAP industrial training schemes in Latin America.' The basic causes of rising unemployment in the developing coun- 'SENA: Servicjo Nacional de Aprendizale (Colombia). INACAP: Instituto Nacional de Capacitaci6n Profesional ( hile) tries are of course not to be found within their education systems and the recommendations given in the preceding paragraphs would do little to solve the fundamental problem of the traditional (low income) sector of the population aspiring for a place in the modern (high income) sector, with its narrow employment opportunities. Education can help to reduce the unemployment among certain cate- gories of educated at secondary and tertiary levels but otherwise the unemployment problem requires a concerted inter-sectoral approach to rural and urban development. To guide the long-term strategy of Bank operations in this field, the Bank's research program includes a study of the functioning of labor markets in developing economies, which should throw more light on the recruitment, employment and utilization of educated manpower, with particular reference to the unemployment problem. The study, which will be closely linked with research undertaken in the framework of the ILO World Employment Program, is expected to be initiated in fiscal year 1972. Financial Constraints As indicated previously, rapid school-age population growth and the persistent demand for education mean that enrollments inevitably will increase. In addition, there are upward pressures on unit costs (costs per student). These are largely determined by staff costs, which in most developing countries make up over three-quarters of total recurrent expenditures in education. Teachers' salaries, in competi- tion with wages in the modern sector, tend to rise regardless of productivity. Needed improvements in teacher qualifications, the advance toward seniority by young teachers, who often constitute the majority, and the tendency to raise teacher:pupil ratios are other fac- tors which will act to increase staff costs, and unit costs, in the future. In the past, public expenditures on education generally have grown at a faster rate than either total public expenditures, public revenue or national income. A comprehensive study of 30 developing coun- tries showed that between 1960 and 1965 the share of education in public expenditures increased, on the average, from 13% to 15%. Similarly, for the same period public expenditures on education rose from 3°% of national income to 4°f,% (Annex Table 3). A representative sample of 10 recent Bank appraisal reports indicates that this trend has continued in most cases between 1965 and 1970. Given that both enrollments and unit costs are likely to increase, future increases in expenditures on education are virtually inevitable. However, the increase in educational expenditures cannot con- Wlq tinue indefinitely at a faster rate than that of national aggregates. In 1967, of 50 developing countries for which data were available, 18 devoted more than 20% of total public expenditures to education, compared with just six in 1960. It will be increasingly difficult to obtain an ever greater share of a country's budgetary resources for educational purposes. Solutions to this dilemma will have to be sought in four inter- related areas. These will include: -searches for forms of education which will make a more direct and less costly contribution to rural development; -attempts to improve internal efficiency and increase productivity; -attempts to identify new sources of educational finance; and -efforts to help governments plan and control the size and shape of school and university systems. Dilemma for the Government: how to allocate resources The above analysis of the quantitative expansion and the problem of quality represents in some respects an outsider's point of view. In the borrower's calculation social, cultural and psychological consid- erations are merged with the economic to create ultimately a political formulation of the issues in which short-term solutions and benefits sometimes are more heavily weighted than long-term. The public has increasingly come to consider education as a right and a school place as the entrance ticket to a better life in the mone- tary sector of the economy. Parents are generally well aware of the growing unemployment among school graduates, but the gap be- tween the subsistence and monetary sectors is so wide that they may still consider that it pays to provide secondary and higher education for their children on the off-chance that one may secure a job that will enable him to help his less fortunate brothers and sisters. To a very large extent these views have been accepted by the governments and are reflected in their education policies. Cultural and ethnic demands may furthermore force the borrower to establish educational institutions-secondary schools as well as university departments-in areas which for a long time to come will lack the economic resources and the population necessary for a satis- factory development. In some instances schools may be seen as a device for keeping young people out of trouble. Innovations, which are seen by outsiders as necessary to remedy qualitative and financial problems, may often be resisted by the de- veloping country-and particularly by teachers and ministry officials 264 -as changes which would downgrade the system and endanger vested interests. They may refuse to participate in experiments which they feel should be tried out in other countries which could better afford failures. They are particularly sensitive to proposed changes in examination systems since these may seem to imply a retreat from accepted world standards. This demand for international equivalence in examinations is undoubtedly hampering the adjustment of second- ary and higher education to local needs but it represents a politically vocal requirement which most African and Asian governments find difficult to resist. In the face of the surging demand for education of the masses as rapidly as possible, it is not surprising that very few leaders have stopped to ask, as President Nyerere of Tanzania has done-What kind of society do we wish to create? What can we realistically expect to achieve with our limited resources? How can we fashion our edu- cation system so as to maximize its contribution to these ends? Whom shall we educate, for what and how? Instead most governments have sought to replicate what was at hand as widely and as quickly as they could. REVIEW OF WORLD BANK POLICIES AND OPERATIONS, 1963-1971 Policies The basic elements of the Bank's education policy in the initial phase of our educational financing were elaborated in September 1962 when the first education project was presented to the Executive Directors and subsequently in a memorandum from the President on "Proposed Bank/IDA Policies in the Field of Education" of October 1963. In the President's Memorandum, the basic statement of policy with respect to types of projects to be financed reads as follows: "The Bank and IDA should be prepared to consider financing a part of the capital requirements of priority education projects designed to produce, or to serve as a necessary step in produc- ing, trained manpower of the kinds and in the numbers needed to forward economic development in the member country con- cerned. In applying this criterion, the Bank and IDA should concentrate their attention, at least at the present stage, on proj- ects in the fields of (a) vocational and technical education and training at various levels, and (b) general secondary education. Other kinds of education projects would be considered only in exceptional cases." 265 The Bank would thus concentrate on high priority projects within the country's education development plan and fill the most crucial gaps in the system provided they fell within certain a priori defined areas of eligibility. Technical assistance and financing of software- curriculum reforms, education planning, production of teaching ma- terials, etc. would comprise very minor parts of the projects, which would consist mainly of construction and equipment of school buildings. With increased knowledge and experience, the Bank's approach to education widened in the late 1960s. In a Memorandum for the Executive Directors in July 1970, the President reaffirmed the first sentence of the 1963 statement quoted above, but added, "in apply- ing this criterion in future we should broaden the scope of projects considered and we should determine priorities and select projects on the basis of a thorough examination of the education system as a whole rather than by a priori designated areas of eligibility which may not relate to the particular country. We should continue to emphasize projects which, like vocational training, produce trained manpower directly but we should also consider for financing other types of projects . . . which should have important long-term signifi- cance for economic development." Such projects would be "designed to encourage changes which improve the relevance, efficiency or economy of education systems". The Bank's education financing in the fiscal years 1963-1971 has thus developed from hardware projects in restricted education sub- sectors to mixed hardware and software projects based on sector analysis aimed at achieving qualitative improvements and meeting crucial manpower needs. Operations Up to June 30,1971, the World Bank Group has approved financing of 57 education loans in 42 countries for a total amount of US$ 431 million. Africa, including North Africa, has received 27 loans and 44% of the total lending volume; Latin America 15 and 22%; Asia 12 and 25%; and Europe three loans and 9% of the volume. In the beginning most education projects were financed through IDA credits and only 10% of the projects in the fiscal years 1963-1967 were in countries with more than US$ 200 per capita annual income (Annex Table 4). Subsequently, the income range of the borrowers widened and in 1970 and 1971 about 56% of the financing was in countries with more than US$ 200 per capita income. The trend to include countries with a more developed economy has led to in- 266 creasingly complex education projects designed to meet the need of a diversified labor market in a larger monetary sector and it has often involved major reforms and innovations for the modernization of outdated systems. So far 72% of the Bank's education financing has been in secondary education. About 23% has gone to universities and post-secondary education and 4% to adult training, while primary education directly (as contrasted to the indirect effect of teacher training) has received little more than 1% of the funds (Annex Table 5). The post-second- ary non-university component has increased lately from 8% to 13% of the total. By curricula, general education, including pre-vocational options, comprises the most important project component both in number of student places provided and as a recipient of funds (Annex Table 5). The demand for general secondary education has, however, reached a saturation level in some developing countries during the last years, while important needs persist for teachers, technicians and engineers. Consequently, technical education and teacher training have received 48% of student places provided through the Bank proj- ects in 1970 and 1971, as against a 28% average in the earlier period, while student places in general education are significantly reduced from 64% to 45% (Annex Table 6). The provision of student places for agricultural education has remained at 7-8% and most places have been in middle level agricultural institutes which have had a restricted enrollment and therefore a low output of graduates. Most agricultural graduates are being employed in government extension services and one of the reasons for the low output is insufficient budget allocations for such services leading to limited employment capacity. As mentioned, Bank education projects have often made signifi- cant contributions to educational system building. Practical and pre- vocational subjects have been introduced in the general secondary education curricula on an equal basis with academic subjects in 21 countries. Science teaching has been strengthened and the ratio of science graduates to arts graduates will now be compatible with the market demand in countries such as Colombia and Uganda. The in- clusion of curriculum reform studies in recent Kenya and Iran projects will hopefully lead to the practice of continuing reform similar to that in more advanced countries. The instructional television in the Ivory Coast project which is being introduced in the formal education system will be designed to constitute an integral part of the student's learning process as routine as the use of textbooks. The Bank's insist- ence on having full-time teachers in the Bank-financed schools is intended to discourage the unsatisfactory part-time teaching tradition which is so prevalent in Latin America. It should be said that it is still too early to assess finally the outcome of the above reforms. Reflecting the Bank policy of relying on other agencies as the pri- mary source for technical assistance in education, the Bank-financed technical assistance component has always been small and amounts to only 5% of the loans. The number of projects which include Bank- financed technical assistance is, however, increasing. Thirty percent of the projects approved in 1963-1967 contained a technical assist- ance component, while 90% of the 1970-1971 projects do. Disregarding the small technical assistance component, two-thirds of the loans have been for construction of physical buildings and one- third for provision of equipment (Annex Table 5). Therefore, much attention has been given to the refinement of costing techniques in appraising the physical aspects of projects and improving cost plan- ning measures in implementation of projects. As a result, norms for school construction and basic equipment lists for a variety of types and levels of schools have been developed. A recent preliminary study on unit costs in education projects should also prove useful in our future activities, even though firm conclusions cannot be drawn at this stage due to the slender data base (Annex Table 7). With increased emphasis on technical education, the equipment component has increased and, consequently, the demand for foreign currency. The main reason for a drop in local currency financing from 40% of the amount of the loans in 1964-1968 to an estimated 18% in 1969-1973 is, however, increased financing of education in rela- tively more developed countries. Annex Table 8 contains a complete list of education loans by coun- try, type of project and amount as of July 1, 1971. Organization, Procedures and Criteria The organization, procedures and criteria for educational project financing by the Bank are the result of several main factors, from which have developed the length and pattern of the project cycle, the size and composition of educational projects, patterns of disburse- ment, and various other characteristics of financing in this sector. Probably the most important of these factors is the character of the sector itself as described in the previous section. The education sector is made up of many elements which interact organically through their effect on the development of human resources and their common dependence on limited sources of revenue as well as on each other. In its organic aspect education is not entirely different from other infrastructure sectors, such as transportation, where intersectoral re- lations and the "investment mix" are of basic importance. In its eco- nomic character, however, it does differ from these sectors in one or another respect. Not only is it not directly revenue producing; in the present state of our knowledge, its economic return is not accurately measurable. Moreover, educational investment carries with it an un- usually high rate of commitment of recurrent expenditures by the government-perhaps 30 cents per year for each dollar invested. In determining project size and composition, therefore, the recurrent cost commitment rather than the investment cost is often the deter- mining factor. We have noted other constraints which may also limit the number of new student places which should be created at a given time the ability of the labor market to absorb persons with different types and levels of training, the availability of qualified teachers and the need for curriculum, structural or nianagement reforms. It is therefore necessary to view the sector as a whole, within the country's devel- opment as a whole, and put together an investment package which, at any given time, will help to meet that country's needs in a balanced and economical way without overcommitment of scarce resources. In addition, even where agreement exists upon educational objec- tives and needs, there is far less objective and technical consensus upon how to meet those needs than exists in other sectors. In most sectors development assistance is essentially a process of selecting and applying proven technologies which have reasonably predictable results. The results of educational technology even in the advanced countries are difficult to predict and its relation to development is only beginning to be studied. Our organization and procedures have been strongly influenced by the Bank/Unesco Cooperative Program established in 1964, very early in the Bank's history of educational financing. The program provides for cooperation by the Bank and Unesco in assisting member countries to identify and prepare projects for Bank consideration and in appraising, supervising and, when requested by the country, pro- viding the technical assistance component of such projects. Under the agreement Unesco has "primary responsibility" for assisting coun- tries in project identification and preparation, which it exercises through a separate Educational Financing Division, devoted exclu- sively to Bank project work. In 1969-71 this unit assisted borrowers in identifying 65% and preparing 80% of the Bank projects. In the original agreement no provision was made for cost-sharing or other cooperation with respect to services that are closely related to project identification and preparation, such as educational plan- ning, and which might therefore more directly contribute to the 9 1Q efficiency of our operations. During the past year the Cooperative Program has been expanded to provide services of this kind, to be made available by Unesco's Division of Educational Planning and Administration and by its Department of School and Higher Educa- tion. These services will include the provision of country education profiles which should help to increase Qur knowledge of the require- ments of this sector. What then have been the most important practical effects of these factors upon the Bank's organization and procedures for educational financing? (a) Reflecting its traditional role, the Bank has concentrated upon capital financing and has relied upon the cooperation of other agencies to provide the technical assistance related to the educational and institution building objectives of Bank-financed projects. About 90% of such projects have involved necessary technical assistance from bilateral or other multilateral sources. The Bank will continue to seek such cooperation. However, greater attention by governments to improving the quality and efficiency of their education will require, in the short run, larger inputs of technical assistance. The expanding volume of Bank activity has also, in some cases, exceeded the immediate fund- ing ability of other sources of technical assistance. In such cases we have provided the financing through our loans, while other agencies, at the borrower's request, have arranged for the services. (b) Educational project identification has depended from the beginning upon comprehensive reviews of the sector which could establish some relative order of priorities-quantitative and qualitative. These sector reviews have been carried on in some cases as parts of the Bank's economic surveys and more often under the Bank/Unesco Cooperative Program by Unesco project identification missions. Unesco identification reports have been useful to both the borrower and the Bank. While they vary in scope and depth and until recently have been largely confined to formal education, they have provided a valuable pilot exercise and example in project identification through a sector approach. (c) Because of the constraints mentioned earlier and despite the packaging of a number of items in one project, the size of our education loans has been small in comparison with most other sectors. They have ranged from US$ 1.5 million to US$ 20 million and have averaged US$ 7.6 million. The packaging proc- ess has also called for larger than ordinary missions made up of a number of specialists. Preparation and appraisal missions, for example, require a number of specialized educators-general, agricultural, technical-aswell as an architect and an economist. WORLD BANK EDUCATION POLICIES AND OPERATIONS IN THE 1970s Possible New Areas of Lending Bank policies and lending operations have been adjusted in re- sponse to the trends in educational development described above. This adjustment will continue, but changes in the pattern of our lend- ing will be gradual rather than radical. During the next five years the established areas of lending-technical, agricultural, teacher training and improved general secondary-will continue to provide the bulk of our projects. Particular attention will be paid to agriculture and to the close tie between the agricultural training, the agricultural exten- sion activities and the application of agricultural research. We expect also to increase our activity in some areas where we have only re- cently begun, such as management training, and also to initiate lend- ing in some wholly new areas. Our new and increased activities are likely to fall in the four interrelated areas mentioned above, i.e. the search for less costly education through non-formal training; more efficient and productive education by means of curriculum reform and the coordinated use of new technologies such as radio, televi- sion, programmed learning and related teaching materials; explora- tion of new sources of finance; and management studies to improve the planning and control of educational systems. (a) Non-formal education and training As indicated above, the vast numbers of uneducated and misedu- cated young and older people in the less developed countries, prin- cipally in the rural areas, will constitute a major challenge to devel- opment for several decades to come. It is true that universal primary education in the 1980s is a major educational policy objective in many developing countries. There is, however, a long way to go be- fore the goal is reached as 46% of the age groups in the developing world are still left outside the formal education system. An illustra- tive case is East Africa where about 70% of the new entrants to the labor force have had either unfinished primary education courses or no primary education at all. Most of those 70% will remain in the labor force in the year 2000 as unskilled, illiterate subsistence farmers or as equally unskilled, semi-employed urban dwellers, unless major 271 efforts are made to provide non-formal training opportunities. The Bank is therefore concerned to find alternatives to formal primary education and this will include suitable forms of functional literacy programs. The Bank has supported a few projects in non-formal edu- cation, notably farmer training centers in Kenya and Tanzania, mobile training units in Somalia and Tanzania and an industrial training scheme in Chile, and we are supporting accelerated vocational train- ing centers in a number of countries. We have also commissioned a research study designed to discover other types of non-formal educa- tion which might be assisted by the Bank. We expect therefore that the next five years will see a significant expansion of the Bank's activ- ity in this area and two main categories of programs are envisaged: (a) for children and youth outside the formal system, (b) for adults. The objective and content of those two programs would be similar but the teaching methods, structure and financing would differ. The new mass communication media may have a very high potential for non-formal education. (b) Educational radio and television Prudently used as an integral part of the education system, these media can be highly effective in the introduction of new curricula, in upgrading of teachers and in the most efficient use of the best teach- ers for the mass of students. They do not hold out the promise of lower costs per student but because they can be highly effective in- novating forces they can, in some cases, provide a much greater educational return for a moderate increase in expenditure. Two educational television projects which the Bank has considered -in the Ivory Coast and Niger-bring out sharply the characteristics of this type of development and why the Bank should not rush head- long into it. In the Ivory Coast educational television is being intro- duced as part of a multi-faceted program of curriculum reform, pre- service teacher training and in-service teacher upgrading. Existing transmission facilities will be used, obviating the need for major in- vestment for transmission, and the expected number of students and teachers to be reached constitutes a sufficiently large market to keep unit costs at an acceptable level. In Niger most of these favorable cir- cumstances do not exist. Although an experimental project in tele- vision for 20 classrooms in the city of Niamey had success in exploring new teaching techniques and reducing dropouts, expansion of the project for the education system as a whole would require substantial capital and operating expenditure for a small and widely dispersed market which will not be economically justifiable for some years to come. 272 A serious danger in educational television is that hardware instal- lation will precede the program planning and production, organiza- tion and teacher training and equipment maintenance which are essential to the effective utilization of hardware. This danger is being heightened by the promotional efforts of suppliers and by the tend- ency of governments to view these media as easy short cuts to edu- cational development, which makes them susceptible to high pressure salesmanship. Wherever it participates in these projects the Bank will exert a counter influence to the emphasis on hardware. (c) Programmed learning Programmed learning is a method of organizing the content of a course in progressive steps so as to permit a student to proceed at his own speed with a minimum of guidance by the teacher. It is being used in a number of ways in advanced countries, primarily as a means of freeing the quicker and the slower students from the lock-step pace required for an average class. In the advanced countries pro- grammed learning has often been associated with a high degree of mechanization and automation. Therefore, it has been considered in- appropriate for developing countries. Recent research and experi- ence, partly through a Unesco project in Lebanon, indicate that the educational benefits of programmed learning can be delivered through inexpensive workbooks. There may be equally important financial benefits. Although programmed learning requires a sub- stantial investment at the outset for programming and other start-up costs, it promises to reduce the recurrent costs of instruction byallow- ing higher student:teacher ratios. If applied on a wide scale, this approach conceivably could ease the financial burden of education in many developing countries. Thus, programmed learning might lend itself well to Bank financing, especially in more populous nations. (d) Teaching materials and equipment A better and cheaper supply of locally produced textbooks and simple teaching equipment is another of the needs in many devel- oping countries. It should be emphasized that successful production of teaching materials requires more than printing facilities and work- shops. A major effort of this kind requires, first, an appropriate cur- riculum and course syllabi for which teaching aids are to be prepared. On the production side, consideration must be given to the selection and training of textbook authors, equipment designers and audio- visual materials producers. On the consumption side, a market and marketing procedures must be established. In some developing coun- tries it appears that most of these prerequisite conditions exist and that Bank initiative and assistance could provide the catalyst for the solution of a major problem. In a recent loan to Turkey, for example, assistance has been provided to expand and improve an existing center for local production of inexpensive science teaching equip- ment which will in future reduce the country's dependence on costly, often unsuitable, imported equipment. Although in many countries, it may require time to prepare the way for this kind of development, this type of lending may be expected to increase during the next five to 10 years. (e) Business administration and management Only a few institutions in the developing world offer training in business administration and management. A rapid industrialization in some countries, an important development of export and import trades in others, and a replacement of expatriates by nationals in private and public enterprises in most of the former colonies have increased the demand for skilled middle level and high level man- agers. The Bank expects therefore to finance formal as well as non- formal training institutions in business administration and manage- ment, of which the Institute of Management in the aforementioned Turkish loan provides the first example. (f) Management of education systems As indicated throughout this paper, perhaps the most pervasive weakness of the education systems is their management. We have already financed within a number of projects the creation or strength- ening of educational planning units and we propose now to broaden this approach to include improvement of the full range of manage- ment functions at national and provincial levels as well as in the individual schools. This involves the training of supervisors, school principals and other administrative personnel in planning, schedul- ing, budget and accounting techniques, etc. This type of assistance is appropriate for the Bank and would bring important benefits to whole educational systems. Project Design, Sector Analysis and Project Sequence Of major concern in the design of education projects for Bank fi- nancing is the need to make education and training systems more responsive to the borrowing country's economic and social needs. The ill-adapted and costly nature of existing education systems, the persistent imbalances between demand and supply in the labor mar- ket and the severe limitations of financial and human resources often require fundamental and broad reforms. Projects should therefore not be designed exclusively in terms of meeting manpower needs by expanding and improving specific sectors of the education and train- ing system but should also engender such reforms. While continuing its policy to finance educational investment designed to meet the borrowing country's immediate and most pressing manpower needs, the Bank will encourage more long-term, comprehensive and incisive objectives. Projects will, more and more, be selected on the basis of a thorough examination of the education system as a whole and as part of well designed, comprehensive education plans and reform measures. In some cases this may require projects of longer duration -perhaps eight years for the working out of institutional changes. In other cases we might finance long-term educational programs through repeater projects planned in advance to implement succes- sive stages of the program. The sector approach was applied at an early stage of the Unesco/ Bank Cooperative Program but the methods have varied and will vary depending on the circumstances and on our knowledge of the spe- cific country's education system. Any program of action must further- more distinguish between: -sector analysis with the limited objective of serving as a basis for Bank operations only; and -sector analysis with the widest possible objective of offering solu- tions to basic national policy issues. In the first instance our analysis may be based on a numberof sources: information gathered through previous Bank projects, Bank economic missions, preinvestment studies, Bank in-depth reconnaissance mis- sions, Unesco and FAO project identification missions, Unesco's col- lection of country data, reports from bilateral and private agencies, etc. In the second case the in-depth analysis must heavily involve the government of the country if it is to be successful and lead to appro- priate follow-up actions. The exercise might be financed through a UNDP grant, a Bank loan or other means. A recent study in the Philip- pines and forthcoming sector studies in Ethiopia and Northeast Brazil exemplify such comprehensive in-depth analysis. In order to come to grips at an early stage with the complex prob- lems of an in-depth examination of the education systems and of projects of longer duration as described above, the Bank staff expects to devote more time to sector analysis and other pre-appraisal work. Projections of Lending Operations in FY 1972-1976 Our projections for the five-year period 1972-1976 envisage the financing of some 80 projects in 56 countries and a total lending of about US$ 800 million. (See Annex Table 10.) A comparison between the actual and projected lending during the five-year periods of 1964-1968, 1969-1973 and 1972-1976 is shown below: FY 1964-1968 FY 1969-1973 FY 1972-1976 Aggregate number of projects 21 64 80 Total lending 157 610 800 (US$ million) Total project cost 266 1000 1400 (US$ million) Analysis of loans by country categories The content of the projected lending will of course reflect many factors including, most importantly, the borrowers' own investment priorities. On the basis of the approach to educational financing out- lined in the earlier sections of this paper, however, the pattern of the Bank's educational financing might be expected to consist of four broad categories of loans. (a) About 35 loans at $3-6 million each for projects in smaller countries-many of them new borrowers. The projects would form packages of differing items with the objective of meeting overall quantitative as well as qualitative manpower demands. General secondary education of an improved type might be a project item, together with expansion of technical and agricul- tural education and teacher training. In the poorer countries projects in non-formal education would have a high priority. Construction of physical facilities would continue to form an important part of the projects. Technical assistance for planning, improvement of management and project implementation will often be necessary. (b) About 30 loans of $7-10 million each to medium-sized countries, for the most part where the Bank has previously been active. In some of these countries, as previously explained, fur- ther expansion of the general secondary education system may be questionable because of low employment demand and finan- cial constraints. Projects would be more concerned with quali- tative improvements or with special sectors of the education system, such as rural education, use of new media and non- formal vocational training, which would require substantial technical assistance. In many of these countries the strengthen- ing of planning and management capability is urgent. (c) About 10 loans of $15-25 million each to countries with large populations. Because of the size of these countries and, in most cases, their federal structure, individual Bank projects would probably not attemptto finance large-scale national build- ing programs or comprehensive education reforms intended to cover all important aspects of, e.g., secondary education. To achieve impact our lending would have to concentrate on three major types of projects. One type of loan might finance country- wide projects in instructional radio or television, programmed learning and production of teaching materials. Such projects re- quire a large market of the type available in populous countries to justify the investment. The second type of loan might go to construction of institutions related to specific parts of the labor market such as teacher training schools, polytechnic colleges, agricultural institutes or farmer training centers. The third type might involve the financing of diversified education projects in specific states or provinces with particularly urgent needs. (d) About 5 loans of $10-20 million each (or in some cases con- siderably more) to less developed European countries. Countries which see their future within the framework of a common mar- ket, urgently need to bring their education and training systems up to the standards of other European countries. Bank lending for this group would point towards rationalizing and moderniz- ing education in the interest of economy and greater output of qualified workers and managers. Analysis of loans by mode and outlay A further analysis by mode and outlay of the above lending indi- cates that more than one million new or improved student places might be provided through the Bank-financed projects. Approxi- mately 60% of those places might be in technical and agricultural education and in teacher training, compared with 42% in the past, while the number of places in general education might be reduced from 58% to 40% of the total. The shares of the lending for non- formal education and for technical assistance might increase from current 4% and 5% to approximately 15% and 10%. 277 Research and Project Evaluation Useful research is being undertaken outside the Bank which needs evaluation and adaptation to our operational needs. Within the Bank, a case study of cost-benefit analysis in education (Kenya) and a recent survey of educational financing methods have helped to clarify sig- nificant questions involved in our operations. Ongoing studies are: a case study of the external and internal performance of a secondary education system, the use of education planning models and a study of student loan fund schemes. A contract was signed recently be- tween the Bank and the International Council for Educational Devel- opment (ICED) under which the latter is undertaking an assessment of existing experience in non-formal education for rural development. ICED is conducting this research with substantial cooperation by Unesco, FAO, ILO and a number of bilateral and private agencies. A plan for future research, developed last year, covers a wide range of important topics. Important new areas of research which are pro- posed, in addition to the ongoing studies mentioned above, are: -evaluation of the functioning of labor markets in developing countries; -cost/effectiveness of alternative learning technologies; -improved techniques for project implementation and evaluation. The proposed labor market study would investigate existing pat- terns of recruitment, employment, utilization and in-service training of educated and trained manpower in developing countries in the context of existing labor market characteristics: systems of wages and salaries, hiring and in-service training practices, employers' criteria for employment, information deficiencies, job preferences, labor (im)mobility. The objective would be to identify labor market factors which prevent the effective utilization of the output of the education and training system for productive employment. The study would thus form a segment in our knowledge of the design of education for economic development and supplement the manpower forecasting technique used so far. The weaknesses of manpower forecasting are well known and the proposed study would deal with at least one of these weaknesses: that training for correctly assessed manpower needs does not guarantee that students will ultimately be employed in the fields for which they were trained. Two major technologies are currently used in vocational educa- tion: training in vocational schools and on-the-job (in-plant) training. The Bank has so far financed the first category although it has been increasingly felt that the second category might also merit Bank sup- port. In relation to such support the cost effectiveness of the two tech- 278 nologies should be studied and compared. Such comparison would also cover the different teaching methods used, particularly pro- grammed learning, which was originally developed for vocational trainingand has beenwidelyused in industry,the militaryservices,etc. Considerably more attention will be given to the evaluation of Bank projects than was possible in the past. This is not only because more of our projects will have become operational in the near future but also because of growing awareness of the need to integrate evalu- ation mechanisms into the process of project implementation. So far, our practice has been to postpone evaluation until a project has been completed and has become fully operational. To measure the full educational effect of a project as much as eight years may have to pass after its inception. In many cases, however, certain elements of a project (e.g. the results of technical assistance, the training of teach- ers and other staff) could be subjected to evaluation at an earlier stage. Continuous feed-back of information in the stage of imple- mentation could provide early warnings about the need for project modifications. The proposed third major research project would thus provide the Bank with a methodology to evaluate the education proj- ects of the 1960s and define the frames of reference necessary for a proper and continuous evaluation of future projects. CONCLUSIONS A deeper and more continuous dialogue between the Bank and the borrower is necessary if we are to encourage in the borrower a greater willingness to reform and innovate and if we want to succeed with the projected financing of education as indicated in this paper. On the Bank side, more intensive sector analysis, particularly with a longer time perspective, can improve the Bank's understanding of the sector and the credibility of its advice. On the borrower's side, im- provement in management and planning practices should lead to a better appreciation of priorities and readier availability of soundly conceived projects. Unesco's present effort to strengthen its planning services for member countries should contribute to this end as well as financial assistance for planning and management analysis for which the Bank has begun to provide financing in selected cases. In all these activities the prime objective is to develop local capability and self reliance. From this combination of factors, we hope to develop a better basic strategy agreed between the Bank and the borrower within which we can identify a phased sequence of projects covering a period of up to 10 years. In this phased sequence we would expect to find capital and technical assistance financing interspersed so as to constitute a ra- tional and orderly application of resources for development of the sector. Although we anticipate that the volume of Bank lending will con- tinue to increase rapidly as it has done since 1968, it is a major point of this paper that the success of the Bank's efforts in the education field is not to be measured primarily by the amount of money we lend but rather by the effectiveness with which Bank and country resources are deployed to meet crucial needs. We expect the Bank to become the largest financer of educational assistance and to continue in most cases to finance projects of such size that a qualitative and quantita- tive impact is secured. However, since all external assistance to edu- cation will probably continue to be less than 10% of total expenditure by developing countries, the motivating and multiplying effects of external assistance will be the principal test of success. ?An Annex Table 1 Enrollment Growth Rates by Educational Levels and Regions (1960-1966) Number of Pupils (Thousands) 1960/61 1966/67 Percentage ot Increase Academic Year Academic Year 1960-1966 Ist 2nd 3rd Ist 2nd 3rd Ist 2nd 3rd Level Level Level Total Level Level Level Total Level Level Level Total Africa 18,931 2,115 192 21,238 26,748 3,893 334 30,975 +41 +84 +73 +45 Latin America 26,973 3,885 567 31,425 36,653 7,468 978 45,099 +36 +94 +72 +45 Asia (1) 74,645 12,186 1,432 88,261 111,986 21,421 2,911 136,300 +50 +76 +103+54 DEVELOPING COUNTRIES 120,549 18,186 2,191 140,924 175,387 32,782 4,223 212,374 +42 +80 +93 +51 WORLD 248,486 63,927 11,174 323,587 311,700 96,713 19,992 428,405 +25 +51 +79 +32 Note: Figures may not total due to rounding. (') Excluding Japan, Mainland China, North Korea and North Viet-Nam. Source: Office of Statistics, Unesco. Comparative Enrollment Ratios at First and Second Levels 1967/68 (Percentage) Enrollment Rates of Enrollment Rates of Combined Primary and Primary School Age Secondary School Age Secondary Enrollment Population(') Population Rates(') Africa 40 15 28 North America 98 92 96 Latin America 75 35 55 Asia (2) 55 30 45 Europe and USSR 97 65 85 Oceania 95 30 67 (Arab Countries) (50) (25) (38) TOTAL DEVELOPING COUNTRIES 54 23 44 WORLD 68 40 56 1X) Regardless of the School they attend. (t) Excluding Mainland China, North Korea and North Viet-Nam. Source: Estimates of the Office of Statistics, Unesco. Annex Table 2 Inefficiency in Primary Education Years to produce Primary Education Expenditures Primary School spent on Students who drop Graduates out in Grades 1, 2, or 3 Ideal Actual (Percentage of Total) Latin America Venezuela 6 10 19 El Salvador 6 15 37 Africa Ethiopia 6 14 32 Ivory Coast 6 12.5 21 Kenya 7 8 6 Asia Malaysia (East) 6 7 4 Ceylon 7 10.5 10 Source: Bank Economic and Education Appraisal Reports. Table 3 Public Expenditure on Education as a Percentage of the Budget and National Income 1960 and 1965 1960 1965 1960 1965 Percentage Number Percentage Number Percentage Number Percentage Number of ot of of ot National of of National ot Budget Countries(') Budget Countries(l) Income Countries Income Countries Africa 14.5 23 16.4 36 3.0 21 4.3 22 America, North 15.6 10 17.6 18 3.9 15 4.1 14 America, South 12.6 7 15.4 10 3.1 11 4.0 11 Asia 11.8 17 13.2 28 3.3 16 4.0 16 Europe and USSR 13.5 13 15.0 23 4.2 25 5.3 24 Oceania 10.4 4 15.7 10 3.7 2 4.4 2 WORLD 13.5 74 15.5 125 3.6 90 4.5 89 (') Including territories. Source: Unesco questionnaire on statistics of educational finance and expenditure. F. Edding, latereatioral Dereleopaats ofEditcatieoal Expeadityres I1D-1J5IF Unesco. Annex Table 4 World Bank/IDA: Education Loans by GDP/Capita at Time of Loan FY 1963-1971 Number of Loans Millions $ GDP/Capita at 63 64 65 66 67 68 69 70 71 Total Total Average Factor Cost Amount Size Below $100 3 2 2 1 1 2 4 15 101.2 6.7 $100 -200 1 1 1 4 1 1 4 3 16 118.9 7.4 200 -299 1 2 2 3 8 59.6 7.5 300 - 399 2 4 2 2 10 82.3 8.2 400 -499 1 1 2 4 20.75 5.2 500 - 599 1 1 13.5 13.5 600 - 699 700 799 800 -899 1 1 1 3 35.2 11.7 900 - 999 1,000 + Total Number 1 3 3 4 6 5 10 11 14 57 431.45 7.6 Total Amount 5.0 17.6 29.5 33.95 51.8 24.2 81.8 79.9 107.7 431.45 Average Size 5.0 5.9 9.8 8.5 8.6 4.8 8.2 7.3 7.7 7.6 Analysis of World Bank/IDA Education Lending Table 5 FY 1963-1971 A. By Levels Million US$ Percentage Primary 4.90 1.1 Secondary 309.65 71.8 Post-Secondary 56.08 13.0 University 43.24 10.0 Adult Training 17.58 4.1 Total 431.45 100.0 B. By Curricula General(') 190.77 44.1 Technical(2) 126.48 29.3 Agricultural 63.03 14.7 Teacher Training 51.17 11.9 Total 431.45 100.0 C. By Outlay Construction 262.17 60.8 Equipment 148.16 34.3 Technical Assistance 21.12 4.9 Total 431.45 100.0 O) Includes comprehensive education (2) Includes industrial and commercial 283 Annex Table 6 World Bank/IDA Education Projects, FY 1963-1971 Student Places Provided or Improved No. Student Places 1963-1969 1970-1971 1963-1971 Provided or Improved Number % Number % Number % General and Comprehensive 433,000 64 120,000 45 553,000 58 Technical 159,000 24 91,000 34 250,000 27 Agricultural 55,000 8 18,000 7 73,000 8 Teacher Training 30,000 4 36,000 14 66,000 7 Total 677,000 100 265,000 100 942,000 100 Boarding Places and Staff Housing 1963-69 1970-71 1963-71 Number of Boarding Places Provided 82,000 35,600 117,600 Boarding Places as % of Total Places Provided 12 13.4 12 Total Education Project Costs (US$ Millions) 429 245 674 Total Cost of Boarding and Staff Housing 100 38 138 Cost of Boarding and Staff Housing as a % of Total Project Costs 23 16 20 284 Annex Table 7 World Bank/IDA Education Projects Cost of Physical Facilities per Student (Median Cost-US$) Academic and Communal Boarding Facilities Secondary Secondary Post Post General Agricultural, Secondary Secondary and Technical General Technical Adult Post Teacher and and Teacher and Educa- Secon- Secon- Staff Primary Training Vocational Training Vocational tion dary dary Housing Sample Size (No. Institutions) 31 161 30 27 14 11 Construction Cost/Student 365 475 641 661 1381 1164 538 1116 - Furniture Cost/Student 39 58 46 81 96 62 86 123 - Equipment Cost/Student 50 110 829 241 906 1073 - - - Total Cost/Student 450 648 1400 1047 2416 2318 652 1258 14,283 285 Annex Table 8 World Bank/IDA Education Projects Approved as of July 1, 1971 Total Amount of Loan Fiscal Project (US$ Million) Year Country Main Purpose Cost Bank IDA 1963 1. Tunisia I Secondary General, Technical and Teacher Training 9.2 5.0 1964 2. Tanzania I Secondary General 6.0 4.6 3. Pakistan I University Agricultural, Post-Secondary Technical and Teacher Training (T.A.) 9.0 4.5 4. Pakistan 11 University Agricultural, Post-Secondary Technical and Teacher Training (T.A.) 17.0 8.5 1965 5. Philippines University Agricultural 11.7 6.0 6. Afghanistan Secondary Technical, Agricultural and Teacher Training (T.A.) 4.7 3.5 7. Nigeria Secondary General, Technical, Adult and Teacher Training 30.0 20.0 1966 8. Chile I Adult Training 3.8 2.75 9. Morocco Secondary General, Technical and Agricultural 16.2 11.0 10. Ethiopia Secondary General, Technical and Teacher Training 10.7 7.2 11. Pakistan Ill University Agricultural and Post-Secondary Technical (T.A.) 21.7 13.0 1967 12. Kenya I Secondary General, Technical and Teacher Training 9.7 7.0 13. Tunisia 11 Secondary General and Agricultural 19.8 13.0 14. Jamaica Secondary General, Post-Secondary Agricultural, Technical, Adult and Teacher Training (T.A.) 19.4 9.5 15. Thailand Secondary Technical and Agricultural (T.A.) 21.0 6.0 16. Uganda Secondary General 14.3 10.0 17. Malawi Secondary General and Teacher Training 7.0 6.3 1968 18. Malagasy Secondary General, Technical and Teacher Training 7.2 4.8 19. Nicaragua Secondary General and Teacher Training 8.0 4.0 20. Gabon Secondary General and Teacher Training 3.6 1.8 21. Sudan Secondary General, Post-Secondary Agricultural and Teacher Training (T.A.) 15.4 8.5 22. Ecuador Secondary General, Agricultural and Technical and Teacher Training (T.A.) 10.2 5.1 1969 23. Colombia I Secondary General 15.2 7.6 24. Chad Secondary Agricultural and Teacher Training (T.A.) 2.1 1.8 25. Trinidad and Tobago Secondary General and Teacher Training 18.8 9.4 26. Guatemala Secondary General, Post-Secondary Agricultural and Teacher Training 12.6 6.3 27. Guyana Secondary General and Teacher Training (T.A.) 10.0 2.9 2.9 28. Zambia I Secondary General, Technical and Teacher Training 36.2 17.4 29. Malaysia Secondary General, Technical, Agricultural and Teacher Training 16.4 8.8 30. Tanzania II Secondary General and Teacher Training 7.2 5.0 31. Korea Secondary and Post-Secondary Agricultural and Technical (T.A.) 26.8 14.8 32. El Salvador Secondary General, Technical and Post-Secondary Agricultural (T.A.) 8.4 4.9 286 Annex Table 8 continued Total Amount of Loan Fiscal Project (US$ Million) Year Country Main Purpose Coot Bank IDA 1970 33. Cameroon Secondary General, Technical, Agricultural and Adult and Teacher Training (T.A.) 14.0 10.5 34. Zambia 11 University Technical and Teacher Training 7.4 5.3 35. Sierra Leone Secondary General, Technical and Teacher Training (T.A.) 4.5 3.0 36. Chile 11 Adult Industrial and Agricultural Training 3.0 1.5 37. Ivory Coast Primary, Secondary General, Technical, Post-Secondary Technical, Agricultural and Adult and Teacher Training(T.A.) 19.1 11.0 38. Chile Ill Secondary Agricultural and Teacher Training (T.A.) 14.0 7.0 39. Kenya 11 Secondary Technical, University Agricultural and Adult and Teacher Training (T.A.) 9.3 6.1 40. Colombia 11 Secondary General (T.A.) 13.0 6.5 41. China Secondary and Post-Secondary Technical and Agricultural and Teacher Training (T.A.) 15.0 9.0 42. Pakistan IV University Technical (T.A.) 12.8 8.0 43. Spain Primary, Secondary General and Teacher Training (T.A.) 24.0 12.0 1971 44. Iran Primary, Secondary General, Technical and Agricultural; Teacher Training; and University (education) (T.A.) 41.7 19.0 45. Indonesia Secondary Technical (T.A.) 7.6 4.6 46. Greece Post-Secondary Technical (T.A.) 24.0 13.8 47. Dominican Republic Secondary General and Teacher Training (T.A.) 8.1 4.0 48. Tanzania Ill Non-Formal Rural Training and Post-Secondary Agricultural (T.A.) 4.7 3.3 49. Jamaica 11 General Secondary; Teacher Training; Vocational Training; ITV(T.A.) 28.2 13.5 50. Congo (B) Secondary General and Technical Teacher Training; Non-Formal Rural Education (T.A.) 4.1 3.5 51. Ethiopia 11 Secondary General and Secondary Technical and Agricultural 13.4 9.5 52. Brazil Secondary Technical and Agricultural, Post-Secondary Technical (T.A.) 21.0 8.4 53. Chad 11 Secondary Technical and Agricultural 3.1 2.2 54. Somalia Secondary General, Technical; Teacher Training and Non-Formal Agriculture (T.A.) 3.7 3.3 55. Turkey Secondary and Post-Secondary Technical; Technical Teacher Training; Non-Formal Management and Adult Technical Training; Science Equipment Production; Mass Media (T.A.) 17.9 13.5 56. Senegal Secondary General and Secondary Technical and Agricultural 2.3 1.8 57. Uganda II Secondary General and Technical; Post-Secondary and Non-Formal Agricultural; Health and Medical Training (T.A.) 10.4 7.3 TOTAL 755.6 212.65 218.d 431.45 Explanatory Notes. "Secondary Gene,al'=inciudes Comprehensive and Pre-ocational courses Tetchnical" =Includes Industrial and Commercial ' (T.A.)" =Technical Assistance lca7 .C . - - - -CC- - -- - - - - - -a =QUCaCz 0 0 0 0 0 0 C- - - - - -CC > - - - C-0 C'-C - - - - - - - - - -- --- = . - - -C -'C - --- - - -- - - - - -- - - - ------- -------- ~ ~ ~ ~ ~ ~ ~ ~ 0 C = = o 0 _ CC . . b. 0 C >- F.eC C :> > DC C < L AO l O DC 21bD(-C OXaIaaal aaa-(a __o-~-- _ ~ nr'.cs (CC'CC.-. .CCC'CSCC'C (C'C=Cfl'CC'. xOXCCSCSC ~~ NCaCCC CC.CCCCCCCXxI a..~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~a Ca =S OS = L'-_c< _e _. S ( . SaS ( ,o C-n CCJCaCC CN C.C C-C C ,CCC ('CC CCC C.C CCCC CC CCCC- CJ (J 1 1C CC ,C C 1 ( VC C_aa'CflU. C E C o( C CCCC CCCCC.. C'oCCCCCCx CCCC.-.'a Cs- C1 S(CC- ('OiNNX4NCC XNt -X_XOVOXtO - XXNN<7zJ0XsS atZC -- -O t C_ 2 -L - Ca = _ lE _ '_ z C C- Y C-°CC (C(CCCC; = n- ~~~-QD b-CCC)OL Annex Table 9 % of GNP Secondary Devoted Students % Higher Annual Output to Educ. % of Total % Secondary per Enrollment from Higher (Public Public Ex- Primary Primary Secondary Enrollment Teacher in Education Expendi- penditure Enrollment Students Enrollment In (Full- Agriculture per tures Devoted to Ratio per Ratio Vocational time and 100,000 only) Education Net Teacher Net Schools equiv.) Engineering Population (5) (6) (7) (8) (9) (10) (11) (12) t13: 4.3 9.9 99 24 46 60 12 24 77 B 7.2 19.6 C 99 26 51 YD 12 17 9 266 3.5 9.1 99 32 661Y 52 22 25 145C 3.9 20.5 99 26 90 20 21 21 308 7.0 23.3 99 31 64 50C 20 20C 204A 4.4 14.3 99 32 71 19 21 189B 5.5 18.8 99 23 82 22 16 15 64C 7.1 14.6 99 D 16 75 34 11 10 180 5.8 13.8 98 28 58 5C 18 16 188B 4.8 15.1 97 26 80 20 515 C 2.2 11.7 99 33 53 16 31 17D 90 4.8 11.4 95 33 54 YC 25 17 13B 133A 2.4 12.5 E 91 35 29 22C 30 19 31C 10.5 20.0 51 42 9 22 20 6 6 3.0 20.0 74F 48 8 23 25 4 7 4.0 20.0 73X 64 4X 18 30 4.3 N 14.3 D 33 72 2 8 30 5.8 23.2 80X 59 13 10 28 4.0 P 20.0 17 51 5 2 34 10 4 3 3 C 13.7 C 75 40E 10 18 18 E 1 4.7 21.7 Q 74 29 8 4 18 15E 13E 4.2 C 20.4 44 46 6 8 23 6.3 V 18.4 65 32 8 2 22 29 5A 3.7 11.7 50X 31 12X 6 16 12C 9C 4.1 N 17.4 54 35 12 3 22 3 7 A 3.2 N 34X 37 4 18 23 16 6 3.5 20.1 43X 46 8X 27B 25 2C 8A 2.9 19.5 C 32 30 13 7C 25 17 7C 6.7 E 6 30 2 5 20 20 0A 3.5 B 22.0 B 31X 48A 2A 3C 17 A 23B 10C 5.8 V 18.9 35 38C 2 6C 21 B 2 0A 5.2 N 24.3 Q 46 EU 36E 4EU 11E 21 E 10 5 5.7 N 19.0 E 78 44 8 20 24 10 2B 6.2 15.0 84 X 56 11 5 22 2 2 2.9 F 17.0 80 56 17 1 26 10 16C 3.0 B 25.4 CQ 63C 31 15C 32E 15C 7A 2 5 B 17.6 43 38 6 22 26 10 6 4.4 19.1 86X 52 43 9A 19 11A 2.6 17.1 Q 71 46 19 23C 23 26C 22 2.4 19.8 80 36 13 8 23 20C 15 5.1 18.9 95 35 49 11 25 19 14 3.2 14.0 64 31 16 17 18C 35C 5.4 29.0 F 89 40 35 31 32 A 28C 41A 4.0 13.6 B 70 40 21 22C 13 29C 20B 3.2 V 21.5 C 71 38 18 29D 13 A 29C 21C 5.0 14 4 90X 33A 18C 3C 28 3C 2.2 6.3 89X 26 17X 5 15 12 12A 4.9 22.0 80 33 35 32 22 18 59 4.2 19 0 98F 52F 50F 16F 35 F 27F 179F 16.0 71 X 40 13 X 21 22 15 C 3.1 10.8 62 33 23 3 34 23 41 F 6.1 16 3 64 22 25 3 26 17 80 4.4 9.8 Q 90 40 54X 3 25 1 46 3.8 18 4 107X 57 42X 19 38 37 114E 3.3 18.0 79 23U 34 26 18 4 120 4.7 18.5 E 89X 31 29X 3 25 16 74 45X 35 15 38 3 6 29.0 93 30 42 8 36 10 457 3.4 16.1 90 32 40 16 25 18 182 4.1 V 17.1 90X 34 13X 13 25 15 37 3.5 17.1 92X 42 25 X 14 33 26 41 48 49 51 51 51 50 49 45 43 (1.4-6.3) (6.3-29.0: (6-107) (22-72) (2-54: (1-32) :13-38, (1-38) i0-457) 4.8 20 0 86 42 25 20 28 19 41 4.0 18.0 71 36 13 11 24 15 13 3.2 14.3 46 32 8 5 20 10 6 Q = Central government only U = Public only Sources: Columns (1) and (2) : World Tables IBRD) Figures Provisional V = Including private expenditure Columns (3)-(9) & (11) IBRD Missions X = Including overage students Columns (10:, (12) & :13): Unesco Statistical Yearbook Y = Unesco sources and IBRD Missions 289 Annex Table 10 Summary of World Bank Education Activities with Projection Through FY1976 Actual Program Actual() Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector Reviews(2) 2 2 2 2 - 8 14 Commitments ($ M.) 82 80 108 180 160 157 610 800 % Total IBRD/IDA 5 4 4 6 5 3 5 - Number of Countries 10 10 14 14 15 19 52 56 Lending Operations 10 11 14 14 15 21 64 80 % Total IBRD/IDA 9 9 11 10 9 7 9 - Projects under Supervision 24 31 42 56 68 11 44 76 (') Including scheduled for FY1972-73. (2) At the end of FY1968, the Bank had an adequate knowledge of the sector in 15 countries. 290 POPULATION PLANNING Introduction ................................................ 293 The Demographic Situation ....... ......................... 294 World Population Trends ......... ............................. 297 The Demographic Transition ....... ........................ 297 Population Trends in Developing Countries ..... ............. 298 World Population Projections ........ .......................... 300 Economic Effects ............ ................................ 303 Relative Growth Rates of Population and Income .... ........ 303 Effect on Per Capita Incomes ....... ........................ 304 Employment and Income Distribution ...... ................. 305 Estimating the Effects ......... ............................ 305 Other Implications of Population Size ...... ................. 306 Family Planning Efforts .......... .............................. 307 Program Results ........... .............................. 308 Constraints on Family Planning ...... ....................... 310 Program Costs ........... ............................... 312 The Potential for Family Planning ...... ..................... 314 The Bank's Program and Approach ....... ....................... 317 Program Projections ......... ............................. 317 Fact Finding and Institution Building ...... .................. 318 Constraints .............. ............................... 321 Population Projects ......... ............................. 322 Areas of Emphasis .......... .............................. 324 Training .............. ................................ 324 Physical Facilities ......... ............................. 324 Communication Strategy ....... ......................... 325 Commercial Sector and Private Groups ..... ............... 325 Measures for Social Welfare ....... ...................... 325 Reduction of Infant Mortality ...... ...................... 326 Research .............. ............................... 326 Cooperation with Other Agencies ...... .................... 327 Annexes 1. Population Projections, 1970-2100 ........................ 329 2. Specific or General Population Targets in 27 Countries .... ... 341 3. Contraceptive Methods ........ ......................... 343 4. Evaluation of Family Planning Programs ...... ............. 351 5. Bilateral, Multilateral and Private Agencies ...... ........... 361 6. Research .............. ...................... 369 oar POPULATION PLANNING SECTOR WORKING PAPER * This paper describes the Bank's efforts to help member coun- tries reduce population growth rates and sets out its future program of activity in the field, as now envisaged. To give perspective to this discussion, the paper also outlines the economic effects of reducing population growth in develop- ing countries and summarizes available information on the global demographic situation, world population trends and projections, and the accomplishments and potential of family planning programs. References to the Bank or the World Bank Group include the International Development Association (IDA) but not, for pur- poses of this paper, the International Finance Corporation (IFC). Money amounts are expressed in U.S. dollar equivalents. The Bank's fiscal year ends June 30. INTRODUCTION The purpose of economic development is to make possible higher living standards for individual men, women and children. A rising standard of living means a growing ability to afford both the material and non-material benefits which a modernized economy makes pos- sible. For most people in most countries, however, the first require- ments are more and better food, improved access to education and health care, and more opportunity for gainful employment. Despite its limitations, one of the best available measures of eco- nomic progress toward these goals is the growth of per capita income. This is the growth of national income, adjusted for growth of popu- lation. Thus the relationship between the growth of a nation's income and that of its population is fundamental to the improvement of human welfare. While neither the causes nor the effects in this relationship are fully understood, one central fact is clear: the higher the rate of popula- tion growth, the more difficult it is to raise per capita income. Today the world's population is growing much faster than at any time in history. This simple fact led the Pearson Commission to say, in 1969, that "No other phenomenon casts a darker shadow over the pros- pects for international development than the staggering growth of population." 293 The problems created by the large numbers and high growth rates of population concern both the world as a whole and individual coun- tries. Both more and less developed countries confront such universal questions as the ultimate size of population the world can sustain and the rate at which the limit will be approached. The earth can undoubtedly support substantially more than the 3.6 billion people now living on it. But there is great doubt about its ability to sustain unlimited numbers at decent standards of living, which a majority do not have even now. The World Bank's concern, however, is not with ultimate numbers, but with the developmental impact of population growth. Develop- ment does not mean more people, but higher living standards and greater welfare for however many there may be. The Bank entered the field chiefly because it became convinced that the attempt to raise living standards in a great many developing countries was being seriously undermined if not thwarted. The Bank has no fixed ideas as to how large the population of individual countries ought to be. But it is convinced that in the great majority of developing countries, the faster the rate of population growth, the slower will be the im- provement of living standards. Within the last few years, the govern- ments of more than 26 countries have indicated that they share this belief by adopting official policies to slow their population growth. There is another important reason for the Bank's entry into the population field. It concerns human welfare, and particularly health. There is strong evidence that where children have been well spaced, both they and their mothers enjoy better health and experience lower mortality rates. It does not follow, of course, that parents will neces- sarily choose to space their children or to have fewer of them, if given the chance to do so. Many governments feel, however, that people should be given the choice if it is possible to bring it to them, and the Bank is prepared to help them do so. Experience suggests that if couples are given this voluntary choice their own family-size decisions will tend to slow the rate of population growth. Yet no one can predict whether the general response will develop strongly or quickly enough to give governments substantial help in attaining their development objectives. The Demographic Situation It took more than 1,800 years for the world's population to increase from 210 million to one billion. The second billion required about a century and a quarter, and the third only 30 years. It is now taking only 15 years to add the fourth billion (see Table 1). If present growth rates 294 Table 1: World Population Trends, Zero A.D. to 1970 A. Population Size>" (millions) About About Area 0 A.D. 1000 1750 1800 1850 1900 1950 1960 1970 World 210 284 750 960 1,240 1,650 2,518 2,995 3,632 Europe 39 63 120 155 195 293 392 425 462 Soviet Union 30 45 70 130 180 214 243 Asia 138 165 480 630 810 930 1,381 1,660 2,056 U.S. and Canada 1 6 26 81 166 199 228 Africa 100 100 100 150 222 278 344 Latin America 33 56 12 20 35 65 163 213 283 Oceania 2 2 2 2 13 16 19 "'Totals may not check, due to rounding. B. Average Annual Rates of Growth (%) Area 1750-1800 1800-1850 1850-1900 1900-1960 1960-1970 World 0.5 0.5 0.6 1 0 1.9 Europe 0.5 0.5 0.8 0.6 0.9 U.S. and Canada - 3.0"' 2.3`" 1.5 1.4 Soviet Union 0.8 0.9 1.2 0.8 1.4 Asia 0.5 0.5 0.3 1.0 2.5 Africa 0 0 0.8 1.0 2.4 Latin America 1.0 1.1 1.2 2.0 2.9 Oceania 0 0 0 1.6 2.0 ('Includes a high rate ot adult immigration. Sources. UN World Population Conference, Vol. II, pp. 21-22; and Population Council, Reports on Population/Family Planning, December 1969. were to continue, the current population of more than 3.6 billion would double in 35 years, and by the end of this century it would be increasing at the rate of a billion about every eight years. It is becoming increasingly difficult to raise living standards and maintain even the present quality of life in the face of these huge annual increases in population. This is especially true in the develop- ing countries, where two thirds of the world's population live and where five sixths of the 1970-80 increase will occur. Such increases impose heavy economic and social burdens. In the mid-1960s, about two thirds of total annual investment in a sample of 22 developing countries was required to maintain per capita income at a constant level, leaving only about a third to raise living standards. The corresponding figures for a representative sam- ple of 19 developed countries were one quarter and three quarters. It is clear that present rates of population growth in developing countries are penalizing the hundreds of millions who live on the margin of subsistence. If developing countries are to achieve sus- tained social and economic development, population growth must be reduced. 295 The high rates of population growth in most developing countries result from their traditional high birth rates and declining mortality rates. Improved health services and medical technology will cause further mortality declines, which will require fertility rates to be re- duced from present levels simply to avoid further increases in popula- tion growth rates. The evolution of populations through the three stages of (1) high fertility/high mortality, (2) high fertility/low mortality, and (3) low fertility/low mortality is what demographers refer to as the historical demographic transition. Both the first and last are periods of modest population growth; the middle stage is one of rapid growth. It is in this middle stage that the world as a whole now finds itself, because the total rate of growth is heavily influenced by what is happening in the less developed countries where two thirds of the world's population live. The demographic transition describes what happened historically in the now-developed countries. The causes of fertility declines that carry countries into stage three are complex and not yet fully under- stood. They include such influences as increases in the age of mar- riage, urbanization, the gradual spread of education, reduced infant mortality, the high living standards that accompany rising incomes, the spread of old age pension systems, the prohibition of child labor, and rising equality for women. There is no guarantee that the same set of forces affecting deci- sions on family size will emerge with equal force in today's develop- ing countries and lead them inevitably into stage three. They have not done so to date. Eventually they may, although no one knows how long it might take. The provision of family planning services can do much to speed up, perhaps by many decades, a transition that otherwise might occur only very slowly. The difference in the pace of this change might make the difference between development and non-development. It is important to note that the Bank's interest in "population" is wider than family planning, which is directed at the spacing of chil- dren and limiting fertility. The Bank's concern also embraces many other aspects of population and its effects on development. Its peri- odic analyses of the development prospects of member countries normally include the main demographic variables of births, deaths, and migration. Population issues are central to the Bank's long stand- ing interests in education, employment, and rural development. These other aspects of the population question lie outside the scope of this paper, which is concerned only with the problem of limiting fertility and with the Bank's contribution to that objective. 296 However, enabling and persuading man to limit his fertility is itself a much broader problem than the provision of family planning services. WORLD POPULATION TRENDS In the pre-industrial era, world population grew slowly. Disease, famine and breakdowns in the social order resulted in mortality rates which were normally high, and occasionally very high. Epidemics sometimes wiped out large proportions of a population in a few years. For example, in the two years 1348-50, bubonic plague (the Black Death) reduced the population of Europe by 25%. Under these con- ditions, fertility had to be consistently high to ensure the survival of families and of the population. Societies whose cultural, religious and legal codes did not place a high value on fertility would not have survived. The Demographic Transition Population growth rates in Europe began to increase during the late eighteenth century. This was the result of a decline in mortality which accompanied the agricultural and industrial revolutions, but which was not matched for many years by any corresponding decline in fertility. Before the decline, death rates were around 28 to 32 per thousand. By the mid-1800s, death rates in England and Scandinavia were about 10 points lower than they had been a century earlier. The decline spread over most of the continent, and the rate continued to fall until in Europe today it is about 10 per thousand. Thus in many European countries the transition to present death rates took more than 100 years; the transition occurred more quickly in countries where it started later.1 Three factors are considered basic to the historical decline in mor- tality rates: improved nutrition as a result of higher agricultural and industrial productivity; better sanitation and personal hygiene, which reduced parasitic and infectious diseases, particularly water-borne diseases; and improvements in medical care. As a result of the differ- ential trends in mortality and fertility, the long term growth rate of Europe's population doubled, from about .5% to 1%. 'One of the more dramatic cases of a rapid fall in fertility occurred in Japan after World War II. Both birth and death rates rose between 1875 and 1920; thereafter the birth rate fluctuated at the intermediate levels of 26-36 per thousand, standing at 33 in 1949. The death rate, on the other hand, fell to about 12 in 1949, giving Japan a rate of natural increase of more than 2%. Between 1948 and 1952, abortion on medical or social grounds was legalized and contraception encouraged by the passage of new legislation. During the subsequent 20 years the birth rate fell 15 points to 18, and the death rate continued its drop to about seven per thousand in 1969, giving a rate of natural increase just over one percent. In most European countries a decline in fertility did not begin until the second half of the nineteenth century. It has continued, with minor interruptions, until the present. In the mid-1700s European birth rates were high (35-40 per thousand in most countries), but not as high as in many of today's developing countries (often 40-50). The sharpest declines in fertility occurred between 1870 and 1930, when they leveled out at around 20, and most European countries today have rates between 15 and 20. The motivation for reduced fertility appears to have arisen from the spread of education, the progress of urbanization, and a realiza- tion that reduced death rates would lead to larger families unless fertility were checked. Population Trends in Developing Countries Prior to World War II, the developing countries also were char- acterized by high birth and death rates, and thus had low rates of natural increase. The demographic transition began with a rapid post- war decline in death rates unaccompanied by a corresponding decline in birth rates. Growth rates began to increase. Today, the average is 2.8%, with the level in some countries as high as 3% and even 4%. There are wide variations, of course, and different developing coun- tries are at different points along the path of the demographic transi- tion (see Table 2). In some, both fertility and mortality remain high. Table 2: Population Distribution in Developing Countries by Fertility and Mortality Levels, 1970 Deaths Births Population per 1000 per 1000 Africa") Asia'12 Latin America Stages Population Population Number Percent Number Percent Number Percent 1 (a) High High (over 25) (over 25) 27.0 8.0 17.0 1.4 - - (b) Failing High (15-25) (over 40) 307.0 90.1 978.5 82.6 24.0 8.4 2 (a) Low (less High than 15) (over 40) 5.0 1.5 76.0 6.5 112.5 39.9 (b) Low (less than 15) Falling 1.5 0.4 112.0 9.5 145.5 51.7 Totals 340.5 100.0 1,183.5 100.0 282.0 100.0 Note: Population figures are approximations. For purposes of this table, they have been rounded to the nearest 500 thousand. 'MExcluding five African countries with a total population of 42.4 million. (2)Excluding People's Republic of China and four other countries with populations totaling 26.1 million, due to incomplete data. Sources of basic data: United Nations, Population and Vital Statistics Report, January 1971; Monthly Bulletin of Statistics, August 1971; and Population Reference Bureau, 1970 World Population Data Sheet. In others, mostly in Asia and Africa, fertility is high and the death rate is falling. In still others, notably in countries with two fifths of Latin America's population, fertility is high and death rates are already low. In a few countries of temperate South America, East Asia and parts of Oceania, death rates are low and fertility is declining. The postwar demographic experience has differed from the earlier experience of developed countries in several important respects: * The decline in mortality has been much more rapid, occurring over one or two decades. For example, between the five-year periods 1945-50 and 1955-60, life expectancy at birth increased in India from 32 to 45 years, in the Republic of China from 41 to 61 years, and in tropical South America from 44 to about 52 years. Consequently, growth rates in the developing countries today are higher than were ever reached in Europe. The difference is not merely in degree, but in kind. A European country with an annual rate of growth of 1% would double its popu- lation in 70 years; the average developing country with a growth rate of 2.5% will double its population in 28 years. If a 1% rate of growth were reduced to .5%, the doubling time would be extended by an- other 70 years to a total of 140. But an identical reduction of half of one percentage point when the growth rate is 2.5% would extend the doubling time by only seven years-a tenfold difference. * Many of today's developing countries lack the open spaces and wealth of natural resources which characterized the areas of Euro- pean settlement and enabled them to support high rates of population growth more easily. * To some extent, the earlier growth in today's developed coun- tries was attributable to adult immigration, leading to significant differences between their age structure during development and that of today's developing countries. By 1900, for example, only about 44% of the population of the United States and Canada was below the age of 20, while in developing countries today that age group commonly accounts for 50-55%. * When mortality began to decline in many of today's developing countries, the levels of economic and social development were not comparable to those prevailing in Western Europe before the indus- trial revolution. * The postwar decline in mortality has been occurring in many societies in which non-traditional economic practices and social atti- tudes, which normally accompany development, have not yet taken root. Unlike the reduction of fertility, a decline in mortality encoun- ters no cultural and ideological opposition and does not require a large measure of continuous and active participation by the popu- lation. In the earlier experience of today's developed countries, however, the technological means for similarly rapid reduction of mortality did not exist. Thus, the current demographic situation in the developing coun- tries does not have historical precedents from which they might take comfort. The considerably more rapid decline in mortality has created a greater imbalance than ever existed in the developed countries. This is the heart of "the population problem" in today's developing countries. The solution depends on how fast and by how much fertility will decline in the next 20 to 30 years. The objective of population pro- grams is to bring about declines in fertility more rapidly than would otherwise occur, primarily by supplying information and services to those willing to use them. To the extent that additional government policies and activities can influence voluntary fertility decisions, they too deserve to be considered as part of a country's population strategy. Much more needs to be learned, however, about both the effects and the efficacy of other instruments. WORLD POPULATION PROJECTIONS Table 3 suggests a range of possibilities over the next 30 years for the population of the world, the developing countries collectively, and a number of large developing countries individually. These are based on alternative projections showing what is likely to happen under optimistic assumptions (Projection A) and very slowly chang- ing conditions (Projection B). Projection A illustrates a type of population growth which could develop over the next 30 years if maximum efforts were made to develop family planning programs and take any other reasonable measures that might be effective in curbing fertility. On the basis of results achieved in the most successful programs to date, a "net re- production rate" (NRR) of 1.0 could be expected by the year 2000.1 This is a possible but unlikely achievement. Projection B can be conceived of as an illustration of population growth trends if family planning efforts remain as modest as they are at present. Even under these slowly changing conditions, some de- cline in fertility can be expected because of family planning and some 'The net reproduction rate is a measure of population replacement, in terms of the number of girl children born over the reproductive life of a hypothetical age group of women, after allowance for mortality. A population will not stop growing when the NRR of 1.0 is reached. It will maintain momentum for two or three generations, depending mainly on its age structure. Usually there is a relatively higher proportion of women in the child bearing ages than would be required for a stationary (non-growing) population. Therefore, even with a sustained NRR of 1.0 these populations would continue to grow for about 60 to 70 years until their age structure became stationary. 2 (vI improvement in socio-economic conditions. In this case a NRR of 1.0 would be reached by about 2040. These alternative projections, discussed in detail in Annex I, have the following major implications: (a) Population size and distribution. World population, which to- taled more than 3.6 billion in 1970, would reach 5.9 billion under projection A and 6.7 billion under projection B by the year 2000, a difference of about 770 million. Under projection A, world popula- tion would level off during the last quarter of the next century after having reached 8.4 billion. Under projection B, population would become stationary half a century later, at about 15.3 billion, or 7.4 billion more than under projection A. The figures for developing countries alone are not very different. These countries would increase their population from 2.5 billion in 1970 to 4.5 billion in the year 2000 under projection A and to 5.3 billion under projection B. The difference is in the neighborhood of 800 million. The ultimate levels for today's developing countries would be about 6.7 billion people under projection A and twice as many, or 13.4 billion, under projection B. (b) The key role of large countries. What happens to world popu- lation will depend very much on fertility trends in a few large de- veloping countries. Comparing projections A and B for developing countries, half the difference in population size is accounted for by 12 of them: seven in Asia (India, Iran, Malaysia, Pakistan, the Philip- pines, Thailand and Turkey), two in Africa (Kenya and Egypt), and three in Latin America (Brazil, Colombia and Mexico).' This suggests the importance of giving priority to efforts to reduce fertility in coun- tries where maximum impact can be achieved. (c) Fertility trends and births to be averted. If projection A were achieved, the birth rate would be 7.1 per thousand lower in 1995-2000 than under projection B (21.1 instead of 28.2). For developing coun- tries, the difference would be 9.3 per thousand (23.7 instead of 32). To achieve projection A rather than projection B, it would be neces- sary to avert about 840 million more births between 1970 and the year 2000 in the developing countries than projection B assumes; almost half of these would have to be in the 12 countries listed above. While present programs give little promise of achieving a NRR of 1.0 by the year 2000, a substantially increased effort now should bring this goal within reach by about 2025. To reach a NRR of 1.0 by the year 2000 would require a maximum effort, something it is probably not realistic to expect. 'Two large countries, Indonesia and Nigeria, are omitted from this list due to lack of sufficient data. Table 3: Alternative Population Projections,() Birth Rates, and Births Omitted Population Birth Rate (in millions) (Per 1,000 Population) Births Omitted Regions and Projec- c. 2075 A 1995- 1970-2000 Countries tion 1970 2000 c. 2125 B 1965-70 2000 2020-25 (in millions) World Total A 3.652 5,916 8,348 34.0 21.1 16.3 882 B 3,652 6,690 15,306 34.0 28.2 23.1 Developed A 1,122 1,388 1,622 18.8 15.8 14.0 45 B 1,122 1,431 1,931 18.8 17.4 15.6 Developing A 2,530 4,528 6,727 41.8 23.7 17.2 837 B 2,530 5,259 13,374 41.8 33.0 24.6 Asia India A 536 948 1,402 41.4 23.6 17.1 B 536 1,100 2,799 41.4 32.9 24.6 174 Iran A 28 56 88 45.0 24.3 17.2 13 B 28 68 213 45.0 36.5 26.8 Malaysia"z) A 9 18 27 36.0 21.2 16.1 4 B 9 21 54 36.0 30.1 22.8 Pakistan A 126 260 408 42.3 24.2 17.0 62 B 126 316 982 42.3 36.4 26.6 Philippines A 38 79 122 45.2 24.1 17.3 1 B 38 95 282 45.2 35.9 26.2 8 Thailand A 37 72 108 41.1 22.4 16.7 15 B 37 86 234 41.1 33.1 24.7 Turkey A 34 62 91 36.0 22.0 16.5 10 B 34 71 173 36.0 30.6 23.2 Africa Egypt A 33 62 94 44.5 24.0 17.1 13 B 33 74 205 44.5 34.8 25.6 Ghana A 9 17 25 49.2 25.6 18.0 4 B 9 20 58 49.2 37.7 27.4 Kenya A 11 21 31 49.6 25.1 17.6 5 B 11 25 70 49.6 36.7 26.5 Tunisia A 5 10 15 45.3 24.0 17.4 2 B 5 12 34 45.3 35.8 26.1 Latin America Brazil A 94 181 273 38.6 22.1 16.5 31 B 94 209 529 38.6 31.1 23.5 Colombia A 21 41 62 38.0 22.1 16.6 8 B 21 48 120 38.0 31.3 23.6 Mexico A 51 109 172 44.6 23.7 17.0 27 B 51 131 397 44.6 35.7 26.1 tFor Projection A, a linear decline in gross reproduction rate (GRR) is assumed to a level which corresponds to a net reproduction rate (NRR) = 1 by the years 2000-2005; this decline is equated with maximum effective fertility control. Projection B represents population trends under the assumption that the GRR will decline linearly to make NRR = 1 in the year 2045; this situation is considered likely to occur if efforts for fertility control continue at the present levels. Both projec- tions assume the same mortality which is supposed to have different declines in the future, for different countries. The full explanation of the assumptions is found in Frejka's introduction to his plojections. t2)Excluding Sabah and Sarawak. Source: The table is computed on the basis of data from Tomas Frejka (Population Council), in "Alternatives of World Population Growth," a monograph in process of publication. ECONOMIC EFFECTS Relative Growth Rates of Population and Income Rapid population growth is a comparatively recent phenomenon. It has accompanied economic development, and it is clear that the possibility for more people to live longer and fuller lives has been one of development's more important results. Past history, however, is a misleading guide to action, because the present situation does not offer a comparable possibility. As noted above, new forces are producing unprecedentedly high rates of population growth, while special circumstances which gave peculiar impetus to economic growth in the earlier period do not prevail. There is no reason to believe that current rates of growth will fall fast enough to relieve the pressures on developing countries arising from the need to use significant and rising proportions of their re- sources simply to maintain the average standard of living of growing numbers, leaving less for further improvement. It is not that coun- tries cannot have both growing per capita incomes and growing populations; the growth record of many developing countries in the last two decades shows that this is not impossible. What is at issue is the maintenance of per capita income growth at acceptable levels over longer periods, when the population may be doubling every two to three decades. The decline of mortality in most developing countries has resulted in the survival of more adults, who would otherwise have had a shorter life span, and an increase in the number of surviving infants. The respective contributions of these two groups to the postwar population increase cannot be accurately calculated, but probably they are about equally significant. The economic impact of their survival is quite different, however; more adults living longer increase the potential labor force and create an immediate demand for jobs and supporting services. They also add to the numbers in the reproductive age group, with a potential impact upon fertility. More children surviving mean a rise in the de- pendency burden and, at a later stage, a further relative increase in the reproductive age groups. Thus while any fall in the infant mortality rate is to be welcomed on humanitarian grounds, it adds burdens to weak economies which can be lessened only by reducing fertility. To do so, and thus lessen the dependency burden, leads to large economic benefits. This is the heart of the economic case in favor of programs to limit fertility. The other component in high population growth rates has been the continued high level of fertility. In the longer run this can be expected to decline everywhere. However, cultural and social factors are significant enough to make for important differences in fertility levels among areas of the world. In the face of continuing success in reducing mortality, no developing country has yet experienced a fertility decline sufficient to reduce the rate of population growth to the average level of 1% per annum characteristic of the developed economies of Europe, North America and Japan. Falling fertility is the only factor that can accomplish such a reduc- tion-except, of course, a return to much higher mortality rates. The issue is not whether the reduction will take place, but how soon, by what means and at what cost. The question is whether it can be initi- ated and accelerated through appropriate policy actions, in order to reap more quickly the economic benefits that can be linked with lower rates of growth. Effect on Per Capita Incomes The most certain, immediate, and measurable benefit of slowing population growth is the increase in per capita income. The imme- diate impact of falling fertility is a decline in average family size, reflected throughout society in a smaller dependency ratio.' In the short run there is no change in the labor force or other resources, so that the same national income will be available to a smaller number of people. At the same time, proportionately less of the national in- come will have to be used to maintain the capital stock per person at a constant level, making it possible to apply more resources to increasing capital per worker, thus raising productivity and per capita income. The higher per capita incomes permit higher savings which could finance higher levels of capital accumulation, both physical and human. This, in turn, leads to further increases in the national income. There is nothing automatic about such a process, however; it is made possible by falling fertility, but the possibilities have to be seized and used for purposes which promote economic development. Such de- velopment may be accelerated as much by the investment in human resources-notably improvements in the quality of education-as by other kinds of capital improvements. The effects of the decline in fertility will be felt in the labor market, which will have fewer entrants approximately 15 to 20 years later. The impact of this decline upon the national income will depend 'The ratio of people not in the labor force to the total population. Since all persons depend on the production of those in the labor force, a lower ratio means that producers do not have to share their output with as many non-producers, yielding both producers and their dependents a higher per capita income. (5 A mainly upon whether the opportunities made possible by lower population growth in the previous 15 years or so have been used to increase the quantity and quality of the capital stock. Many factors are involved, including labor productivity, the composition of the labor force, and improvements in health and education made pos- sible by rising per capita incomes. Employment and Income Distribution Problems of unemployment and inequality in the distribution of income will always be eased by reductions in fertility. Continuing high fertility results in large numbers of young people entering the labor force each year. Employment opportunities have to expand fast enough to absorb them. At high rates of growth of population, where the numbers involved may be doubling every 25 years, the absorp- tion problem is severe. Any country with a problem of long run unemployment cannot fail to benefit from the slower growth in the labor force which results from reduced fertility. Where large numbers of people are entering the labor force to compete for jobs, wages are depressed, while those who own or control capital earn high returns, as do the owners of land and other resources in fixed supply. In any social or political system, high fertility tends to worsen the distribution of income and wealth. A reduction in the rate of growth of population makes it easier to redress these inequities. Estimating the Effects The longer term cumulative economic effects of lowering popula- tion growth rates are clearly profound, although it is difficult to isolate them from those of other economic forces. Since there is no basis for estimating such effects historically, attempts have been made to indicate their orders of magnitude by the construction of simulation models. With these models it is possible to work out the implications of varying fertility reductions over several decades, and to compare the results with the situation assuming no fertility decline. While the quantitative results depend upon assumptions made about the economic relationships involved, they indicate substantial benefits in growth of per capita incomes, with a cumulative effect over time. Typically, if fertility is halved in a generation, by the end of that period per capita incomes can be 20% to 40% higher than if fertility had remained constant. The indicated benefits become more impressive as projections are extended into the future, but results in the shorter period are more persuasive in terms of current policy. This account of the economic benefits to be expected from a re- duction in population growth rates places its main emphasis on the attainment of a higher per capita national income. The national in- come, however, has long been recognized as an incomplete measure of welfare, even in strictly economic terms. It needs to be supple- mented by taking account of other benefits of reduced fertility. Some of these appear as social benefits, but they have economic consequences which may themselves be measurable, at least in prin- ciple. They include an increase in the spacing between pregnancies, bringing benefits to the family in the form of improved health for mothers and children, fewer maternal deaths, and fewer retarded and handicapped infants. Improved nutrition and family care is more likely with smaller family size. A reduction in the number of illegal and unsafe abortions also follows as knowledge of contraception spreads. It is possible to place many of the above conclusions in a cost- benefit framework, relating the value of the benefits to the costs of programs required to bring about reductions in fertility. There is as yet no fully agreed basis for estimating some of the key values to be employed, and even the use of this approach has aroused controversy. But there can be no question that the economic benefits of lower rates of population growth are considerable, and would prove to be all the greater if proper account could be taken of those which are not easily quantified. Calculations have consistently shown a level of benefits that exceed costs by a very wide margin. Other Implications of Population Size Much of the concern about current population growth stems from anxietyabout its implication for the future size of population, whether in the world or in a particular country, in relation to the availability of natural resources. The growth of population, however, accounts for only about half the growing annual drain on the world's resources. The other half, or more, arises from the growth of per capita incomes. Thus, except perhaps in the case of food, it is both rising levels of income and expanding populations that create pressures on the use of natural resources. These seem bound to intensify even if popula- tion growth slows down, despite man's ingenuity in overcoming technological problems with new agricultural methods, new sources of energy, new ways of combating pollution, and new, increasingly dense modes of urban living. Even if many of the difficulties concerning the environment and resource availability stem as much or more from income growth as from population growth, the wealthier a country is in per capita terms, the easier it will be to mobilize resources in order to cope with such problems. For example, an economy of 20 million people with an average income of $2,000 might have the same problems of pollu- tion or scarcity of resources as one of 400 million with a per capita income of $100. There is little doubt that the smaller country could more easily tackle such problems, because of its greater taxing power and the more developed research and production capabilities that are associated with higher incomes. Against the disadvantages of larger population size one potential advantage is sometimes mentioned: countries with large populations may be able to take advantage more readily of the economies of scale that undoubtedly exist in many activities, particularly in manu- facturing. Market size, however, is more a question of aggregate in- come than of population size per se. There may historically have been countries which could have been considered under-populated, in terms of the economy's ability to make effective use of its natural resources. Perhaps the United States was in this position at some point in the past. However, instances when the addition of more people to the labor force led to increases in labor productivity and income per head must have been few in the past and are virtually nonexistent today. Developing economies will be able to increase per capita in- comes more rapidly if their population growth is low than if it is high. FAMILY PLANNING EFFORTS While efforts to reduce population growth rates are still dwarfed by the magnitude of the problem, there has been a notable even dramatic-increase over the last decade in both public and govern- mental interest, concern and action. In 1960, only three countries had official policies designed to slow the rate of population growth; by 1971, 26 countries, with more than two thirds of the population of developing areas, had announced such policies or started official programs; and some 24 others, with 12% of the developing world's population, supported private family planning programs without an- nouncing official policies (see Table 4). By 1970, 30 heads of governments, including those of 19 develop- ing countries, had signed the U.N. Declaration on Population, which characterized "unplanned population growth" as one of the world's "great problems" and called on national governments to recognize family planning as one of their "vital interests." Table 4: Official Positions of 48 Developing Countries on Family Planning Population Size Support but No (Millions) Policy and/or Program Announced Policy 400 and more People's Republic of China (1962) India (1952, reorganized in 1965) 100-399 Indonesia (1968) Pakistan (1960, reorganized in 1965) 25-99 Egypt (1965) Iran (1967) Republic of Korea (1961) Nigeria (1969) Philippines (1970) Thailand (1970) Turkey (1965) 15-24 Morocco (1965) Colombia South Africa 10-14 Republic of China (1968) Ceylon Kenya (1966) Tanzania Malaysia (1966) Venezuela Nepal (1966) Less than 10 Barbados (1967) Bolivia Botswana (1971) Chile Dominican Republic (1968) Costa Rica Ghana (1969) Cuba lamaica (1966) Dahomey Mauritius (1965) Ecuador Puerto Rico (1970) El Salvador Singapore (1965) The Gambia Trinidad and Tobago (1967) Guatemala Tunisia (1964) Haiti Honduras Hong Kong Nicaragua Panama Rhodesia Senegal Western Samoa Source Lapham, R. J. and Mauldin, W. P., "An Assessment of National Family Planning Programmes," unpublished paper presented to OECD's Fourth Annual Population Conference, October 1971. Program Results About 20 million women in 18 countries with programs, or 10% of the married women of reproductive age in those countries, have become "acceptors" during the last five years, the average period the programs have been in effect (see Table 5). It is important to distin- guish between the total number of "acceptors" and the annual in- crease in that number (and to distinguish between the gross and the net increase, i.e., after allowing for those who cease their participa- tion). The figure of 20 million acceptors overstates the number of women practicing contraception, because in every country many qR acceptors leave the program and those who reenter are again re- corded as "new acceptors." Countries with population programs usually establish population goals, or targets. They may aim at reducing the existing crude birth rate to a specified lower rate over five to 10 years, or at recruiting specific numbers of acceptors into the program (sometimes even classified by methods) over a certain number of years.' An examination of these targets will show that a number of the programs which began five or more years ago, after declines in fer- tility for a few years, appear now to have reached an annual peak of acceptors; the number of new acceptors seems to have stabilized, while the decline of the birth rate appears to have slowed down or 'See Annex 2 for population targets. Table 5: Number of Acceptors by Method, and Coverage Achieved by Programs of 18 Countries (OOOs) All Program Methods as Oral Other All a Percentage Contra- Sterili- Program Program of Women Country Year lUDs ceptives zation Methods Methods 15-44"' Ceylon 1966-69 68 55 18 19 160 >8.2 Colombia 1965-70 209 97 u 11 316 u Ghana 1969-70 5 3 0 3 11 1-2 Hong Kong"' 1964-70 76 55 4 149 284 51. India 1964-70("' 3,799 0 8,659 2,098'" u 14.9 Indonesia 1968-70 88 59 0 28 175 u Iran 1966-70 36 627 0 u 662 >9.3 Kenya 1969-70 49 24 u u u 2. Korea, Rep. of 1964-70 1,713 597 150 u u 42. Malaysia'2 1967-70 3 196 9 12 220 8. Morocco 1964-70 41 25 0 0 67 3. Pakistan(') 1964-70 3,277 6 189 u u 4. Philippines") 1970 84 193 - 85 362 u Singapore 1965-69 10 82 4 51 147 35. China, Rep. of 1964-70 779 150 1 u 979 44. Thailand 1964-70 248 207 34 0 489 >9.6 Tunisia 1964-70 66 24 9 20 108 12. Turkey 1965-70 250 32 0 0 282 u Symbols: u, unknown; >, greater than total. (')Non-sovereign territory. 'Excludes Sabah and Sarawak. "'Annual nuimber of acceptors is an estimate based on units of contraceptives supplied, the assumed relationship being one acceptor equals 130 units per year. Because of the arbitrary bias of this estimate, the cumulation of acceptors is not considered warranted. "'Acceptors for years prior to 1970 when a population policy was announced represent clients attend- ing clinics that currently participate in the government program. (5'To March 31, 1971. "'Users of conventional contraceptives, based on figures of distribution. ')This column refers to current acceptors of all program methods as a percentage of women aged 15-44 as of January 1971 except for India, for which the date is January 1970. Source: Population Council, Population and Family Planning Programs A Factbook, No 2 (1971 edition), June 1971, New York. even to have been reversed. The long-established programs in the Republic of China, Hong Kong, Republic of Korea and Singapore are in this situation. In a number of countries-India, Malaysia, Pakistan and Tunisia among others-programs have been in operation for five or more years and a substantial operational infrastructure has been built up; however, performance has been uneven and there has not yet been a significant or demonstrable impact on their fertility rates. But even in the case of the more successful programs, it is clear that more effec- tive education and motivation efforts and more and better family planning services could increase significantly the proportion of women practicing family planning, with consequently lower fertility levels. Other countries are just beginning their programs and the number of acceptors is still increasing. These newer programs include those in Ghana, Indonesia, Iran, Philippines and Thailand. In addition, many countries are moving gradually from limited voluntary efforts in urban centers to a larger scale of services but without any national population policy, although government health facilities may offer family planning services. As their scale of activities expands, these countries can learn from the experience of others and develop not only their delivery systems for contraceptives but also their programs of information, education, and the evaluation of results. Such countries include many of the 34 which have pioneering private associations affiliated with the International Planned Parent- hood Federation (IPPF). One estimate of the number of births averted by family planning programs was made by the Development Center of the Organization for Economic Cooperation and Development (OECD) in 1970. It indi- cated that 2.3 million births had been averted in 1968. Compared with what is needed to bring down the rate of population growth to acceptable levels within a reasonable period, this is far from adequate. Approximately 4.8 million births would have to be averted annually between 1970 and 1975, and more than ten times that many, or 60.5 million annually, between 1995 and 2000 if the population of de- veloping countries were to reach a net reproduction rate of 1.0 by the year 2000 (Annex 1, Table 6). Constraints on Family Planning There is considerable unevenness in the strength of commitments to population control in developing countries, ranging from mere pronouncements to firm policies and programs with varying degrees of budgetary support. This may reflect in part some of the difficulties which have limited the effectiveness of many family planning programs. Politically, there is sometimes concern that support of population programs may be a liability, especially since the results of any pro- gram will not be immediately demonstrable but unfavorable mis- conceptions may be widespread: a belief, for example, that there is a correlation between population size and military power; a fear that ethnic balances within a country will be upset; or a suspicion that the advocacy by rich countries of lower growth rates for the poor is merely a new form of colonialism and an excuse for not providing adequate development aid. Cultural and religious objections to family planning are still serious, even in areas where national programs exist. In some countries a family planning program would imply a complete reversal of prac- tice, tradition and mores. Furthermore, where infant mortality is high and children are needed for the family work force, high fertility is to be expected. Various administrative and organizational difficulties commonly beset governments in launching new programs. These are often seri- ous limitations. No matter how earnest the political commitment, a family planning program cannot be effective unless there is an organi- zational structure capable of bringing the available technology to those prepared to make use of it. Family planning programs are inher- ently difficult to administer, since they must maintain continuing con- tact with married couples over a long period. The difficulty is increased when services must be carried to a widely dispersed, and often illiterate, rural population. Moreover, there is seldom enough trained personnel to carry out a program of the re- quired magnitude. Finally, there is the handicap of limited adminis- trative experience on the part of many persons in the medical and social service professions, who in most developing countries are re- sponsible for administering family planning programs. There are also important technological constraints. Although con- traceptive technology has made considerable progress in the last 15 years, so far there is no perfect contraceptive: highly effective, safe, inexpensive, easily used and reversible, and one which would not necessarily have to be delivered under medical surveillance. Oral contraceptives and the intrauterine devices (IUDs) are likely to be the best available means of contraception for some time to come, al- though experience has demonstrated that both have their limitations. While research is producing variations and refinements of both "the pill" and the IUD, any radically new technique of fertility control must come from basic research in reproductive physiology, which is ex- pensive, complex and uncertain. For the present, family planning programs will have to operate with essentially the same methods known today. Improvements are on the horizon, but radical break- throughs toward the "ideal" are not expected soon (see Annex 3). Family planning programs normally include education and infor- mation components, although the exact type of activities carried out must be carefully tailored to allow for cultural sensitivities and reli- gious beliefs. Education consists of the preparation of curriculum materials on family life and sex and their introduction into school curricula after suitable testing and training of teachers. The target group is principally the next generation of potential acceptors. Information, or communication, activities are directed mainly to the present generation of potential acceptors. These may be reached through various forms of mass communication (radio, cinema, news- papers, posters, etc.), as well as through face-to-face contacts estab- lished by health personnel, social workers, or specially trained field workers. The education and information components are vital parts of any well conceived family planning program. A final constraint on effectiveness is inadequate evaluation. Al- though considerable work has been done on the measurement of results of family planning programs, there is everywhere a long way to go before adequate reporting systems exist to provide information for management decision-making and program evaluation. Inade- quate evaluation machinery limits assessment of a program's effec- tiveness in reducing fertility, the ultimate test of its success. An adequate information system for evaluation would address itself to three questions in particular: (a) The structure and level of demand, to provide guidance for determining the priorities of the program's efforts among areas and population groups; (b) The effect of varying the level and mix of various program inputs, to suggest the combination likely to achieve best results; and (c) The effect of the program on reducing fertility, to provide a measure of effectiveness.' Program Costs Financial expenditures on family planning programs have been modest in relation to national budgets (averaging about 1%) and, as noted in the previous section, very low in relation to the economic benefits of reduced fertility. To date, external assistance has carried a sizable proportion of the costs, usually more than 30%. Tables 6 and 7 give data for selected countries. The relatively low costs are 'See Annex 4 for discussion of a desirable management information and evaluation system. partly explained by the use of existing health services as the main delivery system. In many cases, even when a population policy exists, program ex- penditure is too little in relation to the need. Because national family planning efforts are so recent, cost trends are available for only a few countries; in some (e.g., India, Republic of China) expenditures are increasing; in others the data show decreases. Cost projections and the level of foreign assistance needed in the next decade cannot be assessed with confidence on the basis of such limited experience, but informed estimates have been made which suggest orders of magnitude. Table 6: Budgetary Position of Family Planning Programs in Selected Countries, 1968-1969 (US$ millions) Foreign Family Foreign Aid as ° Family Planning Family Aid for of Family Planning as 04, of National Health"' Planning Family Planning as %, o, National Country Budget Progiam Program Planning Budget Health Budget China, Rep. of 823.0 64.4 0.6 0.5 87.1 1.0 0.1 India 3,141.9 70.8 49.3 18.9 38.3 41.1 1.6 Indonesia 646.3 n.a. 4 .0(2 n.a. 92.5 n.a. 0.1 lamaica 218.3 18.5 0.5 0.2 31.7 2.9 0.3 Korea, Rep. of 984.8 7.7 4.1 2.0 48.8 52.8 0.4 Pakistan 996.6 46.0 19.1 14.9 77.9 41.5 1.9 Trinidad and Tobago 188.6 13.5 0.1 0.03 30.2 0.8 0.1 i)These data are not comparable on a country basis. Some countries do not show tamily planning costs independent of expenditures tor health programs or do) so to varying degrees. In some countries it is considered politically inadvisable to publish precise figures Also, the data do not include the costs of private programs. '-'Data tor 1970-71. In a staff study, the United Nations Fund for Population Activities (UNFPA) has estimated that it would be necessary for developing countries, if they were to carry out reasonably complete family plan- ning programs on a broad national basis, to spend approximately 65 U.S. cents per capita per year. This figure increases to $1 per capita if the costs of educational, motivational and system evaulation activi- ties are included. These amounts are not small when translated into the proportions of national budgets they would represent, e.g., some- where around 5%. Five percent is a much lower proportion than most governments normally spend on education (15% to 30%) or on national security, but it is about the proportion of total Ministry of Health expenditures in many low income countries. Not even the most committed governments with the most suc- cessful programs are spending as much as half the per capita figures mentioned in the UNFPA study, and most are spending much less. Table 7: Annual Per Capita Expenditure of Selected Family Planning Programs"' (US cents) China, Republic of 6.0 (1971) India 7.72 (1968) Indonesia 4.1 (1970) Jamaica 37.0 (1968) Korea, Republic of 10.4 (1969) Pakistan 9.4 (1969) Tunisia 16.8 (1969) "'Based on funds from all sources-government, international, bilateral, and private. Source: Population Council, Population and Family Planning Programs: A Factbook, No. 2 (1971 edition), lune 1971, New York. It is clear that population programs require important shifts in gov- ernment priorities, which will frequently require fairly significant changes in budget allocations. Successful programs require substan- tial increases in both national and international inputs, public and private. The Potential for Family Planning How successful can family planning efforts be in reducing fertility over the next generation? No one knows. Not enough experience to serve as a guide has been accumulated in existing programs, most of which began only four or five years ago. Large areas of ignorance surround such key questions as the potential number of acceptors, since relatively little is known about the determinants of family size decisions and how open to change they may be among various social groups. There is much debate among experts as to whether the present numbers of acceptors can be increased (a) primarily through the ex- tension of services (an approach which assumes the existence of large numbers waiting to participate); (b) whether much more intensive in- formation, communication and family life education activities would be more productive; or (c) whether much larger numbers of accep- tors can be recruited only after basic socio-economic changes have taken place. Both research and further experience are needed to throw light on this fundamental question. Some inferences may be drawn, however, from recent experience: * There does appear to be a correlation between a national pro- gram's supply capacity (i.e., the number of service facilities) and the number of women who make use of the program. This suggests that as a program's capacity is increased, it does gain additional acceptors, particularly in the early stages when those who can be most easily recruited are becoming acceptors. But there is also evidence that ,)1 A programs in which insufficient attention is given to education and motivation do not succeed in recruiting anything like the proportion of women needed to reach national population goals. It seems clear that governments will have to give equal attention to both the supply and demand sides of the problem. * Everywhere there are constraints to program expansion because of the already heavy demands on the medical services, inadequate numbers of trained personnel at all levels, inadequate and insufficient physical facilities, etc. But given the political will, such problems are not insurmountable over the long run (and, as indicated in the next section, these are areas in which the Bank can assist). Over the short run, there are possibilities for making use of non-medical personnel, in particular social workers, and using paramedical personnel more effectively in providing family planning services. * As to long term demand for family planning services, three kinds of evidence suggest that it exists or can be developed: (1) Various knowledge, attitude, and practice (KAP) surveys indi- cate that most couples in developing countries want fewer children than they now have. The average number of children that a family "desires" is between four and 4.5, compared to actual family size of five to 5.7. In particular, many families which already have at least three children do not want more: in Hong Kong and seven develop- ing countries,' 60% of respondents who already had three children and 70% of those with four said they did not want to increase their families (see Table 8). These surveys have their limitations. For example, the replies of older respondents are influenced by their actual reproductive his- tory. Moreover, an expressed preference for a given number of children does not necessarily mean that contraceptives will be used on the requisite sustained basis. Nevertheless, studies suggest that younger and better educated women want smaller families than their mothers did. If the evidence of the studies is accepted, it does appear that there is an unsatisfied demand for expanded family planning services. (2) The appallingly high incidence of illegal abortions in many countries, particularly in Latin America, clearly indicates an unmet need for family planning services. (3) The most successful family planning programs to date-in the Republic of China, Hong Kong, Republic of Korea and Singapore- have been conducted in countries were social pressures and the level of socio-economic development had already led to some decline in 'Hong Kong 1967; Thailand 1965; Philippines (urban) 1969; Turkey 1963; Republic of Korea (urban) 1964; Tunisia 1964; India 1960-61; Indonesia 1963. 31 C fertility. Preliminary results of the 1970-71 census in many countries of Asia and Latin America indicate a population size smaller, though admittedly to a minor degree, than had been projected by the United Nations. To the extent that the lower-than-expected population level indicates that a decline in fertility has begun in these countries, it may be an encouraging indication of the results that might be achieved by inaugurating or intensifying family planning programs there. Table 8: Survey Findings on Actual and Desired Family Size A. Average Desired Family Size and Percentage of Persons Reporting They Do not Want More Children, For Selected Countries Percent not wanting more Average children among those with Number Area Type of Sample Children 4 5 6 Ceylon Rural 3.2 69 88 44 China, Rep. of Urban 3.9 76 88 - Ghana Urban 5.5 - - - India 75 85 37 Central India Urban/Rural 3.8 Mysore a Urban 3.7 b Urban 4.1 c Rural 4.7 d Rural 4.6 New Delhi 4.1 Indonesia Rural 4.3 37 41 19 Jamaica Urban/Rural 3.4-4.2 80 84 - Korea, Rep. of Rural 4.3 - - - Pakistan"'1 Urban 4.0 East 57 54 45 West 68 71 41 Philippines Urban/Rural 5.0 68 85 50 Thailand 3.8 86 96 70 Turkey National Population 3.8 44 66 - B. Completed Family Size by Region Region Completed Family Size12) World Total"3' 4.7 Developing Regions"3) 5.5 Africa 6.1 Asia"3) 5.5 Central and South America 5.7 Developed Regions 2.9 Europe 2.7 United States and Canada 3.7 °)Before political events of 1971-72. (2)Average number of children born per woman living through reproductive period. (M)Excluding the Soviet Union and People's Republic of China. Source: Mauldin, W. Parker, "Fertility Studies: Knowledge, Attitude, and Practice" in Studies in Family Planning, Number 7, The Population Council, New York. 316 In the long run, fertility reduction can be achieved only with the right combination of social and economic development, cultural and political attitudes, and easy availability of contraceptive facilities. No one yet knows the required mix, which will probably differ from country to country. However, the limited experience available has already shown that if an adequate service can be provided, including public information and a variety of acceptable methods, the results will be demographically significant even if inadequate to achieve the desired reduction in fertility. The initiation of movement in the right direction also provides hope that more rapid progress will be possible as programs are extended and improved. THE BANK'S PROGRAM AND APPROACH Program Projections Table 9 summarizes the Bank's actual and projected lending for population projects in the seven fiscal years 1970-76. It also shows the number of sector missions already mounted and contemplated. Table 9: Summary of World Bank Population Activities By Fiscal Years, with Projection through FY1976 Actual Program 1969 1970 1971 1972 1973 1972-76 Sector Missions 1 3 3 4 5 24 Commitments ($ million) 2.0 8.0 29.0 30.0 150.0 % Total Bank and IDA 0.1 0.3 1.0 1.1 1.0 Number of Countries 1 2 3 4 19 Lending Operations (No.) 1 2 3 4 20 % Total Bank and IDA 0.8 1.6 1.9 2.2 2.0 Projects under Supervision"l) 1 3 6 10 14(2 ')End of Fiscal Year. :MAnnual average. The number of projects, the number of countries involved and the volume of lending anticipated are all relatively minor when compared with the Bank Group's total operations. This should not, however, be taken to reflect the importance attached to the population sector by the Bank or by increasing numbers of its borrowers, nor do such projections serve as a full measure of the Bank's growing activity in this field. As now envisaged, projects assisted by the Bank and IDA by the end of FY1976 will be in countries with about 70% of the population 317 of the Bank's developing members. Many of the projects will be con- siderably larger than the first few that have been financed while the Bank has built up its staff, gained experience in the field, established close working relations with governments, other international institu- tions and private organizations concerned with population questions, and mounted nearly a dozen missions to accumulate detailed knowl- edge of the sector in individual countries. Beginning with FY1972, Bank Group operations will focus increas- ingly on the larger countries where a downward shift in fertility rates would be most significant. For planning purposes, it is assumed that roughly half the 20 projects foreseen over the period 1972-76 will involve commitments of between $5 million and $10 million, while half the remainder will be above and half below that range. The Bank's strategy in the five year period will be to establish the usefulness of its project approach in dealing with the population problem, primarily through experience with about 25 family plan- ning programs, including as many as possible in countries with large populations. At the time of drafting this paper, for example, projects in India and Indonesia were nearing the final stage of consideration, and a number of others involving major population groups were already well advanced. The projections in Table 9 are based on an assessment of practical possibilities in specific countries, in the light of both government and popular attitudes prevailing in 1972. They are unlikely to be re- vised downward, but experience may well show them to be overly conservative. Attitudes toward population questions have already changed dramatically since the Bank entered the field three years ago, thanks chiefly to the pioneering work of a few governments and several leading private organizations. The momentum of change is increasing. As awareness of the problem spreads and the Bank's capacity to assist becomes better known, further acceleration of the five year program may become possible. Fact Finding and Institution Building Leaving that possibility aside, however, it should be emphasized that project numbers and amounts in money terms are a less accurate barometer of Bank involvement in the population field than are similar figures for most other sectors. This is because, in most cases, a lack of foreign exchange is not the principal constraint on effective implementation of population policies. Often, in fact, the greatest need and the Bank's most useful contribution is not finance, although this can be critically important, but technical assistance in any or all 318 of its many forms. One of these is education, in the most strategic sense. In many countries, the general implications of population growth in terms of development may be known by the professional and political elite, but understanding may not be sufficiently widespread to support decisive policies or action. In these cases, the Bank can often help governments through its capability for fact-finding and analysis. It can do so effectively, of course, only at the Government's invitation or, at the least, with its acquiescence. Given one or the other, however, the Bank's help in collecting, organizing and analyzing facts about a country's demographic position and its meaning in terms of develop- ment can sometimes be crucial. One medium for this is the Bank's regular economic reports on borrowing countries. As a matter of policy, each of these reports is now supposed to include an analysis of the country's demographic situation and of its population policies or program. Unfortunately, this policy objective has not yet been fully realized; in fact, demographic analysis was included in only half the country reports completed in calendar years 1969 and 1970. This situation can be expected to im- prove, however, steadily increasing the effectiveness of such reports in providing the basic facts governments need to identify their own population problems and generate the political will on which action must depend. Far more detailed knowledge of the demographic situation and of population problems, policies and programs in specific countries is provided by the Bank's sector missions. Because of staff limitations and the shortage of qualified independent consultants in this field, the number of such missions will increase only gradually over time. They can be extremely valuable, however, both to the Bank itself and to the countries concerned. As of January 1, 1972, population missions had been sent to Colom- bia, Egypt, India, Indonesia, Jamaica, Malaysia, Mauritius, Trinidad and Tobago, and Tunisia. Missions to Ghana and the Philippines were planned during the remainder of fiscal 1972. In addition, pre-invest- ment studies in population planning have been completed for Iran, Kenya, Tanzania, and Uganda; hopefully, at least two of these will lead to projects in the near future. Each of the three projects thus far financed by the Bank Group, in Jamaica, Tunisia, and Trinidad and Tobago, was preceded by a sector study. These three relatively small countries have given the Bank invaluable experience in a new sector. A fourth successful sector mission was one in 1969 to Indonesia, involving the United Nations, the World Health Organization (WHO) and the Bank. This mission's 319 work led to the Government's adoption of a five year plan and facili- tated reorganization of the Government's Family Planning Board. In Mauritius, the work of a fifth sector mission led to an increase in the Government's financial support, and provided the basis for an integrated family planning maternal and child health program which was subsequently financed by UNFPA. Proposals for a large scale pilot project in India, including a man- agement information and evaluation system, have been accepted by the Government and will form the basis of a Bank project which has already been appraised. No project has emerged as a result of the mission to Colombia, since the Government does not feel able to accept a project justified solely in terms of family planning. Results of some of these sector missions illustrate another facet of the Bank's technical assistance capability which may often be more important than external finance: help to governments in building the exceptional types of institutional structures which are required to plan and administer successful family planning programs. This institution-building form of assistance involves a wide range of activities. It may include help in the organization and top-level staffing of national family planning agencies; the assessment of a pro- gram's manpower needs; the planning or provision of training for medical, social and demographic personnel; the design, collection and processing of service statistics; the evaluation of a program's performance; the design of research projects; the conduct of attitude surveys; the development of education materials for use in school systems, adult education activities or public information programs; the functional and economical design of buildings used in providing family planning services; the organization of recruitment activities; the procurement and distribution of contraceptive supplies; or other facets of planning, administration or training. The Bank, of course, is not equipped with either the staff or the expertise required to provide assistance in all such areas directly. A number of these matters are the primary responsibility of other agencies in the United Nations system. In some specific fields, the principal reservoir of competence and experience may be found in other international institutions, bilateral agencies or private organiza- tions. The Bank maintains close working relations with all these sources of expert assistance, and can often play an important coordi- nating and catalytic role, with or without financial involvement. The relative emphasis to be placed on fact-finding and analysis, on the one hand, and institution building, on the other, will vary from country to country, depending upon the government's attitude toward population planning. The first role is likely to be dominant in 320 countries which are hospitable or permissive with respect to family planning activity but which have no official programs; institution building will play a larger role in countries which have adopted population policies or programs and which welcome the Bank's assistance. As in other sectors, the Bank's lending for population projects will also be accompanied by help in building institutions and by other forms of technical assistance. One of the most basic of these is assistance in identifying those projects which command the highest priority and preparing them in sufficient detail for the Bank, IDA or other agencies to consider their financing. Many other aspects of technical assistance will be incorporated in the projects themselves, with their costs included in Bank loans or IDA credits. Constraints Government commitment to population planning is a prerequisite to Bank activity. In those of its member countries which have not yet recognized that a population problem exists or which have expressed opposition to family planning programs, and which accordingly do not seek the Bank's assistance, the Bank cannot expect to operate. It may seek to educate and persuade, but it cannot hope to develop projects. This constraint, very real where it exists, may nevertheless become less severe with time, since political support for population planning appears to be spreading while commitments already made are becoming stronger. At present the lead time for project development is long, since the Bank does not possess the basic sector knowledge in most coun- tries which allows projects to be identified rapidly. Every project is a "first," in a new country. The field itself is a new one so far as large scale government programs are concerned, and there is no firm body of experience to give clear leads as to what activities are needed and will be successful. When project proposals have been worked out, they frequently have a number of components that require prepara- tion with several ministries and agencies which may have no experi- ence of working together in the population field. The large number of external agencies providing assistance in the field also presents a problem of coordination in each country as well as internationally, especially since some of the basic relationships among key agencies are still in the process of being sorted out. Finally, almost all governments are new in dealing with population problems; they too are inexperienced and uncertain about the kinds of projects they want and need and about what types and sources 321 of external assistance can help them most. For all these reasons, the number of Bank and IDA lending operations is bound to grow fairly slowly. Population Projects Population programs typically embrace much more than the pro- vision of family planning services. In addition, they may include in- formation and educational activities, research on the determinants of fertility and family-size decisions, adjustments in the social and welfare legislation affecting the age of marriage or the size of families, the improvement of vital statistics, and training in demography, nutrition and related activities. Nevertheless, the core of population projects is the provision of effective family planning services, includ- ing not only supply-oriented activities but also demand-oriented activities designed to motivate and recruit acceptors. Family planning services are provided through a system that can be described in terms of the following nine components: (a) Program organization, management and administration: i.e., designation of the responsible program agency and its location within government, the qualifications of key individuals, and the agency's internal organization and functioning. (b) Physical facilities: buildings, equipment and vehicles used for dispensing services and for training, research, and program adminis- tration; (c) Staffing and training: the medical, paramedical and other staff required and their training in family planning work; (d) Contraceptive methods and supplies: the methods made avail- able in the program and the existing arrangements for procuring and distributing necessary supplies; (e) The "delivery system": the channels through which contracep- tive methods are made available to acceptors. This will depend mainly on how the facilities and staff used for family planning are related to the regular health services, and the way the latter are organized. The use made of the private sector, including voluntary agencies, physi- cians, pharmacists and other retail distribution channels, is an impor- tant aspect of the delivery system; (f) Information, education and the recruitment of acceptors: the kinds of public information and communication programs used to explain program objectives, plus the mechanisms relied on to estab- lish contact with the target population (husbands and wives in the child-bearing years); also family life education built into school curricula. ~9) (g) Evaluation: trends in activity levels as revealed by service sta- tistics; demographic trends as revealed by census data and special research studies; bio-medical effects as revealed by special studies; (h) Finance: program costs, including capital and operating ex- pense, and arrangements for meeting them; (i) The program: the action plan, or strategy, for expanding and improving family planning services. While this is not an exhaustive list, it serves to identify those aspects of a program which the Bank examines when making a sector survey as background for project identification. Weaknesses or gaps in the set of activities outlined above, as found in a particular country, automatically suggest the contents of a project. They may also sug- gest steps the borrower will be asked to take before the Bank is prepared to finance a project. Projects as well as programs normally include both tangible and intangible elements. The tangible-or "hardware"-elements usually account for the principal capital expenditure. These are items such as buildings, vehicles, furniture and equipment, office machines, training aids, printing equipment and machinery for contraceptive manufacture, etc. Most "software" items also require additional expenditures. They may include training, the preparation of materials for schools and adult education, applied contraceptive research, attitude surveys, demographic research, foreign study fellowships, technical assistance services, or additional operating costs arising from program expansion. Project components which may or may not require funds include necessary legal or organizational changes, the improvement of service statistics, the installation of improved accounts, revisions in arrange- ments for procuring contraceptive supplies, or the working out of an action plan for three or four years in the future. The mix of project components in the Bank's initial population projects has already shown considerable variety. The first three, in Jamaica, Tunisia, and Trinidad and Tabago, include a high construc- tion component (about 80% of total project costs), although there are significant differences in the types of facilities financed in the three projects. The much larger Indonesia project consists of only 40% construction, the rest being distributed among advisory services, technical assistance, vehicle purchase, demographic research, training stipends, and incremental operating costs. "Software" components of the first three projects are also widely varied, and there are considerable differences in the organizational and administrative changes agreed upon with the borrowers. In the proposed India project, an experimental nutrition component will 323 be introduced in order to test the relationship among nutrition, infant mortality and fertility. Thus, while population projects consist of a certain number of identifiable "building blocks," the relative importance of each ele- ment can differ greatly from project to project. Furthermore, the importance of a particular component may bear little relation to the amount of money earmarked for it in the project cost table. While facilities financed by the Bank for family planning purposes will frequently include buildings used for other health needs as well, their justification is solely in terms of their contribution to family planning objectives. In other words, the Bank does not currently finance health facilities per se. Areas of Emphasis Certain aspects of the problems of broadening the scope and improving the effectiveness of population programs demand special attention, and will be given greater emphasis in the Bank's work. These include: * Training. The effective use of family planning facilities depends to a large extent on the quality of the training of personnel available to staff them. Most developing countries suffer from shortages of adequately trained personnel in the three most relevant fields: medi- cal, paramedical and social service. Paramedical personnel play a key role in family planning through provision of services in clinics, in the field, and in maternity hospitals. In rural areas they are often the only persons with whom villagers come in contact; they are therefore a critical class of family planning workers. In addition, voluntary social workers can be trained and mobilized for field work. Besides the local training needed for most operational workers, support frequently needs to be given for external training of per- sonnel in the various disciplines at suitable overseas centers. Bank projects have provided for such training of selected personnel to strengthen the programs. * Physical Facilities. Most developing countries have a substantial need for the development of physical infrastructure. This is true of health services and institutions for training, education and research. In the current stage of contraceptive technology, health facilities will continue to be the main vehicle for delivery of services and for the training of personnel involved in family planning. A large number of governments with population policies have committed themselves to provide family planning services only within the context of maternal and child health (MCH) services, for practical as well as political reasons. The successful programs in East Asia are in countries with 324 developed MCH services. Thus the provision of physical facilities in population projects will often be designed to enable family planning services to be delivered within the existing framework of a government's health services. Recent studies have suggested a close relationship between declines in infant mortality and, with some time lag, declines in fertility. Some of the contraceptive methods now in use and others likely to become available in the future require the support of a health infrastruc- ture. These facilities also provide an important base for postpartum motivation. There will also be need to expand other types of physical facilities required for such functions as communication, information and education, the production of educational materials, and research on socio-economic and technical problems. * Communication Strategy. More emphasis will be given to the stimulation of demand for family planning services, especially in the larger of the less developed countries. These include the provision to individuals of information about family planning, group motiva- tion, and face-to-face communication, particularly through social workers. Some existing programs will need a review of their com- munication strategy, since present approaches, adequate to reach the most responsive acceptors, appear less effective in penetrating deeply. Information may be provided through existing channels of com- munication, for example by incorporating family life education into the school curriculum. In addition, influential older people may need to be reached by incorporating family planning information in literary, handicraft, or community development programs. In some countries, and especially for rural areas, it may be more effective to develop a program of home visitors-trusted local per- sons trained in social work and community development - who would work closely with local family planning and MCH centers, serving as a link between them and the community. * Commercial Sector and Private Groups. Often the commercial sector and private groups have not been adequately utilized in gov- ernment programs, though they have been in some and much can be learned from their experience. The utilization of these groups for delivery of contraceptives and for promotional work will be built into programs where opportunities exist. * Measures for Social Welfare. There are indications that reductions in fertility and acceptance of the small family concept may be pro- moted by measures which have little or no relation to the direct provision of family planning services. These include steps to improve 325 the status of women and expand their opportunities for education and a wider choice of occupational and intellectual pursuits. Other examples are changes in the lawful age of marriage, in benefits pro- vided by old age insurance, in the payment of children's benefits under insurance or welfare programs, and changes in allowances or exemptions for children provided by personal income tax law. Al- though research on these topics is now going on, the field is new and little firm guidance yet exists. The Bank will try to keep abreast of find- ings in the field and will use them as appropriate. * Reduction of Infant Mortality. In assessing attitudes toward fam- ily planning, it is important to differentiate between women in the subsistence and monetary sectors of the economy. Among the former especially, a convincing reduction of infant mortality rates may be essential before the majority can be effectively motivated to practice family planning. The Bank intends to take full account of govern- ment policies and plans to achieve this result, for example through nutrition programs and the provision of maternal and child health services. * Research. In virtually all areas involving population questions there is need to improve the state of knowledge by means of well conceived, well implemented research. It is the Bank's intention to promote research in areas where the need is most pressing. Research in the population sector covers a wide variety of topics. These can, however, be classified into four main groups. There is first of all basic demographic research, aimed at improving the data base for the main variables-fertility, mortality and migration. Second, there is economic research into the implications of population trends and their interrelationships with other economic variables. Included in this group would be studies of the economic effects of population programs. The third area is bio-medical research, involving contraceptive technology and reproductive biology, basic to the increase in knowl- edge required before improved contraceptives can be devised and tested. Finally, there is a variety of operations research topics, all of which are directed to questions in various disciplines involving popu- lation policies. These include, for example, research on management questions, evaluation procedures, the role of communications, and motivation and educational techniques. One subject that needs con- siderably more research in nearly all countries is the set of motiva- tions that determines the fertility and family size objectives of dis- similar socio-economic groups. In view of the nature of the Bank's involvement in the financing of family planning programs, its main interests in research are likely I?R)R to be those related to the operations of such programs. The method- ological aspects of family planning programs, including, for exam- ple, cost-effectiveness and cost-benefit questions, or techniques of evaluation, are likely to be of particular concern to the Bank. In addition, analysis of the wider relationships of population growth and movement to economic development requires support through appropriate research. In carrying out general economic analyses of the development prospects of member countries it is necessary to study the demographic aspects which form the underpinnings to economic planning. Population research is usually best pursued in the country con- cerned, by domestic institutions which can draw upon well informed local research workers. The cross-disciplinary nature of the work requires skills and knowledge which normally go beyond the com- petence of the Bank. In these cases, such as biomedical research, the Bank is unlikely to be involved directly. It will maintain a continuing interest, however, and will encourage other appropriate national and international organizations to pursue such work. In such areas as economic and operationally oriented studies, the Bank's involvement may be more direct. It might initiate research projects or encourage national institutions to carry out studies. It is also hoped that the Bank's own operations will yield material which can help to illuminate research undertaken by others. Cooperation with Other Agencies The Bank works with a number of other agencies operating in the family planning field (Annex 5). This cooperation has been both formal and informal. It is certain to increase, in depth and in scope. The United Nations, WHO, UNFPA, the Pan American Health Organization (PAHO), the International Planned Parenthood Feder- ation (IPPF), the International Development Research Centre of Can- ada, the Population Council and the Ford Foundation, as well as sev- eral academic institutions, have cooperated with the Bank in pro- viding experts for missions. The Bank itself has participated in missions with other institutions. The UN-WHO-Bank mission to Indonesia, which the Bank initiated, helped the Government establish its population policy and prepare its Five Year Plan. The UNFPA participated in a second mission to Indonesia. The Swedish International Development Authority (SIDA) sent an observer on an appraisal mission to India. Cooperation with other agencies has extended beyond sector studies into the project implementation stage. In the Jamaica project, two studies bearing on the management aspects of the program were ')7 financed by the U.S. Agency for International Development (U.S. AID). PAHO will provide some of the advisory services in the Trinidad and Tobago project. WHO may undertake some health studies related to the project in Tunisia. In Indonesia, the project will be jointly financed by UNFPA. SIDA may join in financing the Indian project. Within the United Nations system, the major expertise in the popu- lation field is not with the Bank but is spread among a number of agencies: WHO for the medical aspects, the United Nations for demography, Unesco where education is concerned, etc. Accord- ingly, the Bank's work in this sector must be carried out in close cooperation with the other international agencies concerned, as well as with bilateral agencies, such as AID, SIDA, the Canadian Interna- tional Development Agency (CIDA), and with private organiza- tions such as the Population Council and IPPF. Cooperation with the large number of development assistance organizations involved presents some difficulties, particularly in cases where basic approaches to the subject may be different. Moreover, there are many kinds and degrees of cooperation, ranging from projects formally organized as joint financing operations to informal arrangements for separate but complementary projects in the same country. Only after considerably more experience has been gained will it become clear precisely which patterns or sets of patterns of cooperation are likely to be most effective. 328 Annex 1 POPULATION PROJECTIONS, 1970-2100 Future size of population in the various regions of the world will depend ultimately on the trend of fertility in the next 20 to 30 years. Other determinants are the present age and sex structure and the mortality schedule. As for mortality, after a further substantial decline in developing countries over the next decade or so, it may reasonably be assumed that the downward trend will level out, as it has in the developed countries. For this analysis, a recently developed model' has been used to prepare illustrative projections for the entire world, the less devel- oped countries (LDCs) considered as a whole, and for large countries and regions of Asia, Africa and Latin America. This model permits calculations and demonstrates the interaction, at five year intervals, of future population size, age structure, and several demographic indices and rates, such as the average annual growth rate, crude birth and death rates, total fertility rate (sum of age-specific fertility rates multiplied by interval length), expectation of life at birth, and gross and net reproduction rates. The basic data for these projections are the age distribution of the population of a region or country in or around 1965 and the mortality and fertility levels of the late 1960s. Only one set of mortality assump- tions, implying an orderly decline, is used for each region or country; in the calculations, a series of assumptions has been used, differing from each other by speed of fertility decline. For the purpose of this illustration, two projections for each popu- lation under consideration have been selected. These are provided for the whole world, more developed regions, less developed regions, and for regions and countries selected on the basis of their absolute size and their present favorable disposition to measures of fertility control. Projection A assumes that fertility will decline to a level correspond- ing to a net reproduction rate (NRR)' of unity by the period 2000- 2005. For this decrease to be possible, an effort of "maximum effec- tiveness" is assumed. Projection B represents population trends in the future if efforts to achieve NRR of 1 by the year 2000 are unsuccessful, and the level of replacement is not achieved until 40 years later (i.e., by 2040-2045). The measure of the effort required to decrease fertility to NRR of 1 by the year 2000 is the result of comparing absolute numbers of births 'Tomas Frejka (Population Council), "Alternatives of World Population Growth," monograph in process of publication. 'For definition see tootnote, page 10. 329 Annex 1 through the years, under assumptions A and B and applying the same mortality rates. The arbitrary choice of NRR of 1 as a goal of fertility decline is based on the generally accepted view that populations will have to stop growing some time in the future. Fertility patterns corresponding to NRR of 1 lead eventually to stationary populations. A NRR greater than 1 results in compounded population growth. The consequences of continued growth for a long period are extremely large popula- tions, difficult to conceive as functioning societies in terms of present standards. Table 1 presents the results of projections A and B for the entire world and for the less developed regions to the year 2100. Other demographic indicators of the resulting populations at different dates between 1970 and 2100 are included in the table, as supplementary evidence to describe the consequences of rapid or slow fertility decline. World population was estimated to be at a level of about 3,700 million in 1970. Projection A, which resulted from assuming that fertility would decline after 1970 and achieve NRR of 1 by 2000, indi- cates that the size of the population in that year would reach 5,900 million. This growth may be considered a minimum. It is a target which can be attained only if considerable effort is applied to chang- ing present attitudes toward high fertility among peoples of less de- veloped countries, where world population trends will be chiefly determined. The contribution of the developed countries to population growth may be considered negligible. As a group, they had 30% of the world's population in 1970 and a net reproduction rate of 1.25. By 2000, under the assumption of effective fertility control in the world as a whole, their share of population would be reduced to 23%. Therefore, in analyzing the alternatives of world population growth in the next 30 to 70 years, it will suffice to concentrate on alternative changes in fertility in the less developed countries, and especially on likely developments in a few large countries. The total population of developing countries in 1970 was estimated to be about 2,530 million people. Under assumption A (effective con- trol) for fertility decline, by the year 2000 there would be 4,500 million inhabitants of those areas, an increase of almost 80%. This would require a decline in the crude birth rate (CBR) from the level of 42 per 1,000 in 1970 to 24 in 2000 (Table 2), oran annual average decrease of 0.6 per 1,000. If fertility trends decline less rapidly during the next 30 years-that is, if efforts to achieve NRR of 1 by 2000 are unsuccess- ful and the population reaches this rate some 40 years later-the 22n Annex 1 Table 1 Population Projections and Other Demographic Indicators-World, Developed Countries, Developing Countries, 1970-2100 Assump- Region tion'l) 1970 1975 1980 1985 2000 2050 2075 2100 World Total Population (millions) A 3,652 4,019 4,402 4,796 5,916 8,136 8,348 8,386 B 3,652 4,042 4,475 4,956 6,690 13,444 15,306 15,815 Percent Increase (base, 1970) A - 10 20 31 62 124 129 130 B - 11 22 36 83 268 319 333 Average Annual Growth A 1.99 1.91 1.82 1.71 1.20 0.28 0.06 0.00 Rates (%)(2, B 1.99 2.03 2.04 2.04 1.96 0.82 0.31 0.04 Percent under 15 yrs.( ) A 37.0 - - 34.5 29.1 20.3 19.8 19.8 B 37.0 - - 36.6 34.7 23.7 20.3 20.3 Dependency ratio'4' A 0.74 - - 0.66 0.55 0.52 0.57 0.57 B 0.74 - - 0.72 0.60 0.52 0.53 0.53 Developed Countries A 1,122 1,169 1,217 1,263 1,388 1,610 1,622 1,623 B 1,122 1,171 1,222 1,274 1,431 1,853 1,931 1,952 Developing Countries Population (millions) A 2,530 2,850 3,185 3,533 4,528 6,525 6,727 6,763 B 2,530 2,871 3,253 3,682 5,259 11,591 13,374 13,863 Percent Increase (base, 1970) A - 13 26 40 79 158 166 167 B - 13 29 46 108 358 429 448 Average Annual Growth A 2.6 2.4 2.3 2.1 1.4 0.3 0.1 0.0 Rates (%)(2, B 2.6 2.6 2.5 2.5 2.3 0.9 0.3 0.0 Percent under 15 yrs. (3' A 41.6 - - 38.2 31.4 20.8 19.9 20.0 B 41.6 - 40.6 38.2 24.8 19.7 19.7 Dependency Ratio(4' A 82.0 - 72.0 56.3 50.4 56.1 56.0 B 82.0 - 79.2 73.0 48.2 51.1 51.1 0) For projection A, a linear decline in the GRR is assumed to a level which correspondsto NRR=1 by 2000-2005; this decline is equated with effective fertility control. Projection B represents population trends under the assumption that N RR= 1 will not be achieved until 2045; this situation is considered likely to occur if efforts for fertility control continue at the present levels. 12) Annual growth rates for preceding five-year period. ,n) For the female population. 4') Ratio of female population under 15 and over 65 to population 15-64 years. Source: Tomas Frelka (Population Council), "Alternatives of World Population Growth," monograph in process of publication. LDCs will be inhabited by 5,300 million people in the year 2000. The possibility of an additional 800 million in the year 2000 under assumption B, however, is less significant than the difference in total numbers under the two assumptions by the time a stationary popula- tion is arrived at. By 2050, assumption A would result in total popula- tion in the LDCs of 6,500 million, compared with 11,600 million under assumption B. It would still grow to 6,700 million under assumption A, or to 13,400 million under assumption B, before becoming stationary by approximately 2075. Population trends for the years 1970-2075 are presented for main world regions and large countries in Table 3. The consequences of slow fertility decrease have to be measured in terms of the unavoidable ultimate size of the population. Although 121 Annex 1 Table 2 Projected Birth Rates(') per 1000 Population-World, Developing Regions-Selected Countries, 1965-2025 Assump- Region tion(n) 1965-70 1970-75 1975-80 1980-85 1985-90 1990-95 1995-2000 2020-2025 World Total A 34.5 32.4 30.4 28.5 26.6 24.3 21.8 16.5 B 33.6 32.9 32.1 31.3 30.5 29.6 24.0 Developed Countries A 18.8 18.6 18.2 17.8 17.2 16.4 15.8 14.0 B 18.8 18.9 18.7 18.2 17.7 17.4 15.6 Developing A 41.8 38.2 35.3 32.5 29.8 27.0 23.7 17.2 Countries B 39.9 38.3 36.9 35.7 34.4 33.0 24.6 East Asia A 34.5 32.2 29.8 27.5 25.3 23.3 21.2 16.4 B 33.3 31.9 30.4 29.2 28.3 27.4 21.3 South Asia A 44.1 39.9 36.6 33.7 30.8 27.6 23.8 17.1 B 41.7 40.1 38.7 37.5 36.1 34.5 25.4 India A 41.4 38.1 35.4 32.9 30.1 27.1 23.6 17.1 B 39.7 38.4 37.3 36.0 34.4 32.9 24.6 Iran A 45.0 40.8 38.5 36.4 33.1 28.9 24.3 17.2 B 42.8 42.5 42.1 40.8 38.6 36.5 26.7 Malaysia(3) A 36.0 34.4 32.9 30.6 27.7 24.5 21.2 17.0 B 36.1 36.0 35.1 33.4 31.6 30.1 22.9 Pakistan A 42.3 41.1 40.0 37.3 33.2 28.8 24.2 17.0 B 43.2 44.0 43.0 40.9 38.4 36.4 26.6 Philippines A 45.2 41.3 38.2 35.6 32.5 28.5 24.1 17.3 B 43.4 42.1 41.2 39.9 37.9 35.9 26.2 Thailand A 41.1 37.9 35.2 32.4 29.5 26.2 22.4 16.7 B 39.8 38.6 37.5 36.2 34.7 33.1 31.5 Turkey A 36.0 34.2 32.9 31.2 28.5 25.3 22.0 16.5 B 35.7 35.8 35.4 34.0 32.2 30.5 23.2 Africa A 45.8 41.9 38.5 35.7 32.7 29.2 25.3 18.1 B 43.8 42.0 40.6 39.2 37.5 35.7 26.2 Egypt A 44.5 40.3 36.7 33.7 31.1 27.9 24.0 17.1 B 42.1 40.2 38.7 37.6 36.4 34.7 25.6 Ghana A 49.2 44.7 41.0 38.3 34.7 30.3 25.6 18.0 B 46.7 45.2 44.1 42.2 39.9 37.7 27.4 Kenya A 49.6 44.3 40.2 36.5 32.9 29.3 25.1 17.6 B 46.3 44.0 41.9 40.1 38.1 36.7 26.5 Tunisia A 45.3 41.2 37.9 35.1 32.0 28.3 24.0 17.4 B 43.3 41.7 40.6 39.3 37.6 35.8 26.1 Latin America A 39.4 36.4 33.8 31.5 28.9 25.9 22.5 16.6 B 37.9 36.9 35.9 34.6 33.3 31.9 23.9 Brazil A 38.6 35.9 33.5 31.1 28.4 25.4 22.1 16.5 B 37.4 36.5 35.5 34.1 32.6 31.1 23.5 Colombia A 38.0 36.4 34.9 32.8 29.5 25.8 22.1 16.6 B 38.0 38.1 37.4 35.4 33.1 31.3 23.6 Mexico A 44.6 40.6 37.4 34.4 31.5 27.9 23.7 17.0 B 42.6 41.2 40.0 38.7 37.3 35.7 26.1 (') Female birth rates resulting from projections multiplied by factor 1.025. (2) Assumptions used for projections: See Footnote (1), Table 1. (3) Not including Sabah and Sarawak. Source: Tomas Frejka (Population Council), "Alternatives of World Population Growth," monograph in process of publication. Annex 1 Table 3 Alternative Population Projections 1970-2075-World, Developed Countries, Developing Countries, Selected Large Countries (Millions) Assump- Region tion(') 1970 1975 1980 1985 1990 1995 2000 2050 2075 World Total A 3,652.1 4,019.3 4,401.9 4,796.4 5,191.9 5,571.9 5,916.1 8,135.7 8,348.4 B 4,041.7 4,474.8 4,956.1 5,485.9 6,064.2 6,690.1 13,443.6 15,305.6 Developed A 1,122.2 1,169.4 1,216.7 1,263.2 1,309.0 1,350.6 1,388.1 1,610.3 1,621.7 Countries B 1,171.1 1,221.9 1,274.0 1,327.6 1,379.8 1,431.3 1,852.9 1,931.2 Developing A 2,529.9 2,849.9 3,185.2 3,533.2 3,882.9 4,221.3 4,528.0 6,525.4 6,726.7 Countries B 2,870.6 3,252.9 3,682.1 4,158.3 4,684.4 5,258.8 11,590.7 13,374.4 East Asia A 940.8 1,034.6 1,129.6 1,223.5 1,315.2 1,403.5 1,485.0 2,001.7 2,053.4 B 1,039.9 1,146.4 1,259.2 1,379.1 1,507.8 1,645.5 3,010.9 3,358.0 South Asia A 1,102.7 1,253.1 1,413.0 1,582.1 1,755.4 1,925.0 2,079.3 3,076.6 3,153.5 B 1,263.4 1,447.0 1,658.1 1,898.2 2,168.7 2,468.9 5,915.8 6,896.0 India A 534.3 598.3 666.7 739.0 812.4 883.5 948.3 1,365.5 1,402.0 B 602.5 680.5 769.6 869.3 979.5 1,100.2 2,431.8 2,799.0 Iran A 27.9 31.9 36.5 41.5 46.7 51.8 56.2 85.9 88.1 B 32.2 37.4 43.7 51.1 59.4 68.4 179.8 212.9 Malaysia(2) A 9.4 10.7 12.2 13.8 15.4 16.9 18.3 26.8 27.4 B 10.8 12.5 14.5 16.6 18.9 21.3 47.3 54.2 Pakistan A 126.2 144.9 166.7 190.7 215.1 238.4 259.6 398.9 408.4 B 146.2 171.3 201.3 235.3 273.3 315.8 831.0 981.8 Philippines A 37.8 44.0 50.6 57.8 65.1 72.3 78.6 118.6 121.9 B 44.4 52.0 60.9 71.2 82.6 95.2 239.7 282.0 Thailand A 36.6 42.1 48.0 54.2 60.5 66.6 72.1 105.5 107.6 B 42.4 49.2 56.9 65.6 75.4 86.0 202.2 233.7 Turkey A 34.5 38.5 43.0 47.8 52.7 57.4 61.6 88.4 90.8 B 38.8 43.8 49.7 56.3 63.4 70.9 151.4 173.3 Africa A 344.2 386.9 432.4 480.5 529.6 577.2 620.2 899.2 929.0 B 389.9 442.2 502.3 570.2 645.9 729.1 1,663.6 1,931.6 Egypt A 33.5 37.9 42.7 47.6 52.8 57.8 62.4 91.7 93 8 B 38.3 43.7 49.9 57.0 65.1 74.0 176.3 205.1 Ghana A 9.0 10.1 11.4 12.8 14.2 15.6 16.8 24.8 25.4 B 10.2 11.7 13.4 15.5 17.7 20.2 49.8 58.2 Kenya A 10.9 12.5 14.1 15.8 17.6 19.2 20.7 30.3 31.0 B 12.6 14.5 16.6 19.1 21.8 24.8 59.7 69.5 Tunisia A 5.0 5.8 6.5 7.4 8.3 9.2 9.9 14.7 15.1 B 5.8 6.7 7.8 9.0 10.4 12.0 29.5 34.5 Latin America A 283.2 323.8 366.7 411.3 456.4 500.0 539.7 797.3 818.8 B 326.2 374.5 428.7 488.8 554.7 625.9 1,396.4 1,614.3 Brazil A 94.0 107.6 122.1 137.3 152.7 167.5 180.9 265.9 272.8 B 108.3 124.7 143.0 163.3 185.4 209.3 459.3 528.9 Colombia A 21.1 24.1 27.5 31.1 34.7 38.1 41.2 60.3 61.7 B 24.3 28.1 32.5 37.2 42.3 47.7 104.8 120.4 Mexico A 50.7 59.5 68.9 79.1 80.4 99.6 108.7 167.9 172.5 B 60.0 70.8 83.3 97.5 113.6 131.3 335.4 396.8 (') Assumptions used for projections: See Footnote (1), Table 1. (2) Not including Sabah and Sarawak. Source: Tomas Freika (Population Council), "Alternatives of World Population Growth," monograph in process of publication. 22't Annex I assumption B implies also a decline in the CBR of developing coun- tries from 42 in 1970 to 33 in the year 2000, and a drop in total fertility' from 5.7 in 1970 to 4.3 in 2000 (see Table 4), this would not be enough to stop growth for at least 60-75 additional years. By then, the absolute size would exceed 13,000 million persons-twice as many as under assumption A. Accepting the desirability of achieving population trends nearer to projection A in the LDCs, one can elaborate on the process of fertility change which would be necessary and on the likelihood of such change taking place. In the LDCs, under assumption A the population would achieve NRR of 1 by 2000 and would stabilize some 75 years later at approxi- mately 6,700 million. For this situation to occur, the total fertility rate would have to decrease from a level of 5.7 in 1970 to less than three in 2000 (2.8 children per woman, see Table 4). As indicated above, the resulting birth rate would decline at a speed of 0.6 per 1,000 per year, reach a level of 24 by 2000, and further stabilize at about 17 per 1,000 25 years later (Table 2). What would make the difference between a situation conducive to changes in fertility from assumption B to assumption A? This question may be answered by showing how many live births would take place annually in both situations, indicating the magnitude of the difference, and estimating the number of women who would need to practice contraception yearly in order to prevent the excess births. Table 5 shows projected live births for the world, the LDCs, and several regions and countries. The difference between the numbers of births under assumptions A and B may be taken as an estimate of the additional number of births which would have to be averted annually by expanded popula- tion planning programs to accomplish population trends similar to projection A. This has been summarized in Table 6. Thus, in the five year period 1970-75 it would be necessary to avert 4.8 million births per year in the LDCs; these annual averages would increase to 11 million by 1975-80 and up to about 60.5 million births by the year 2000. Since population growth in the developed countries is already slow, and in these projections it is assumed that it will stop in the near future, it may be expected that after the year 1985 birth preven- tion beyond the levels of present contraceptive practice will have to take place in the LDCs. During the period 1970-2000, half the magni- tude of birth prevention to achieve projection A instead of B would 'Sum of five-year interval age-specific fertility rates multiplied by five; it is interpreted as an indicator of "children per woman." V4 Annex 1 Table 4 Projected Total Fertility Rates per Woman(')-World, Developed Countries, Developing Countries, Selected Large Countries, 1965-2025 Assump- Region tion(') 1965-70 1970-75 1975-80 1980-85 1985-90 1990-95 1995-2000 2020-2025 World Total A 4.74 4.38 4.02 3.66 3.30 2.94 2.58 2.15 B 4.56 4.39 4.21 4.04 3.86 3.69 2.81 Developed A 2.66 2.58 2.50 2.42 2.34 2.26 2.18 2.08 Countries B 2.63 2.59 2.55 2.51 2.47 2.43 2.24 Developing A 5.74 5.25 4.77 4.28 3.79 3.31 2.82 2.20 Countries B 5.50 5.26 5.02 4.78 4.54 4.30 3.09 South Asia A 6.15 5.60 5.04 4.49 3.93 3.38 2.82 2.16 B 5.88 5.61 5.35 5.08 4.81 4.54 3.20 India A 5.55 5.08 4.62 4.16 3.69 3.23 2.77 2.19 B 5.32 5.09 4.86 4.63 4.41 4.18 3.04 Iran A 6.76 6.12 5.47 4.83 4.18 3.54 2.89 2.15 B 6.46 6.15 5.84 5.53 5.22 4.91 3.36 Malaysia(3) A 5.31 4.86 4.40 3.95 3.49 3.04 2.58 2.10 B 5.09 4.88 4.66 4.45 4.23 4.02 2.94 Pakistan A 6.53 5.91 5.30 4.68 4.07 3.45 2.84 2.14 B 6.23 5.94 5.64 5.35 5.06 4.76 3-.29 Philippines A 6.77 6.12 5.48 4.84 4.19 3.55 2.90 2.17 B 6.46 6.15 5.84 5.53 5.22 4.91 3.35 Thailand A 6.15 5.58 5.02 4.45 3.88 3.32 2.75 2.15 B 5.88 5.61 5.34 5.07 4.81 4.54 3.19 Turkey A 5.30 4.86 4.43 3.99 3.55 3.11 2.67 2.16 B 5.09 4.88 4.66 4.45 4.24 4.02 2.95 Africa A 6.35 5.80 5.24 4.68 4.12 3.56 3.01 2.28 B 6.08 5.80 5.52 5.24 4.96 4.68 3.28 Egypt A 6.14 5.59 5.04 4.49 3.94 3.39 2.83 2.17 B 5.88 5.61 5.34 5.07 4.81 4.54 3.20 Ghana A 6.97 6.32 5.68 5.03 4.38 3.74 3.09 2.26 B 6.65 6.33 6.01 5.69 5.37 5.05 3.45 Kenya A 6.75 6.13 5.51 4.89 4.27 3.65 3.02 2.23 B 6.45 6.14 5.84 5.53 5.22 4.92 3.38 Tunisia A 6.77 6.13 5.49 4.85 4.21 3.56 2.92 2.19 B 6.46 6.15 5.84 5.53 5.22 4.91 3.36 Latin America A 5.54 5.07 4.60 4.13 3.66 3.19 2.72 2.20 B 5.31 5.09 4.86 4.64 4.41 4.19 3.06 Brazil A 5.38 4.93 4.48 4.03 3.58 3.13 2.68 2.20 B 5.16 4.95 4.74 4.52 4.31 4.09 3.02 Colombia A 5.56 5.08 4.61 4.14 3.67 3.20 2.72 2.20 B 5.33 5.10 4.88 4.65 4.43 4.20 3.07 Mexico A 6.54 5.92 5.31 4.69 4.07 3.46 2.84 2.19 B 6.25 5.96 5.66 5.37 5.08 4.79 3.33 (1) Sum of age-specific fertility rates multiplied by 5; it is interpreted as an indicator of "children per woman." (2) Assumptions used for projections: See Footnote (1), Table 1. (3) Not including Sabah and Sarawak. Source: Tomas Frejka (Population Council), "Alternatives of World Population Growth," monograph in process of publication. it l Annex 1 Table 5 Projected Annual Average Number of Live Births(')-World, Developed Countries, Developing Countries, Selected Large Countries, 1970-2000 (Thousands) Region Assumptions(2) 1970-75 1975-80 1980-85 1985-90 1990-95 1995-2000 World Total A 124,090 128,160 131,260 132,758 131,052 125,200 B 129,170 139,954 151,238 163,560 176,272 188,548 Developed A 21,255 21,761 22,120 22,147 21,808 21,608 Countries B 21,511 22,566 23,287 23,726 24,001 24,485 Developing A 102,835 106,399 109,140 110,611 109,244 103,592 Countries B 107,659 117,388 127,951 139,834 152,271 164,063 East Asia A 31,789 32,276 32,320 32,137 31,628 30,643 B 32,991 34,847 36,616 38,536 40,835 43,149 South Asia A 46,966 48,780 50,501 51,485 50,738 47,611 B 49,354 54,313 60,153 66,707 73,366 80,098 India A 21,593 22,366 23,125 23,376 22,946 21,593 B 22,547 24,657 27,052 29,482 31,836 34,213 Iran A 1,220 1,318 1,420 1,460 1,424 1,312 B 1,287 1,480 1,709 1,934 2,135 2,332 Malaysia(3) A 345 377 398 404 396 373 B 365 419 474 520 561 606 Pakistan A 5,572 6,228 6,668 6,738 6,531 6,023 B 5,877 6,980 8,021 8,928 9,774 10,718 Philippines A 1,689 1,808 1,928 1,996 1,958 1,817 B 1,781 2,030 2,326 2,633 2,916 3,190 Thailand A 1,492 1,584 1,655 1,693 1,668 1,557 B 1,571 1,770 1,990 2,216 2,450 2,672 Turkey A 1,249 1,341 1,414 1,432 1,394 1,311 B 1,307 1,474 1,653 1,804 1,927 2,051 Africa A 15,327 15,786 16,283 16,516 16,166 15,157 B 16,063 17,482 19,166 20,995 22,809 24,523 Egypt A 1,438 1,479 1,522 1,559 1,542 1,441 B 1,512 1,647 1,813 2,011 2,221 2,417 Ghana A 426 443 464 469 452 415 B 449 495 554 610 662 714 Kenya A 518 534 545 549 539 501 B 544 596 652 715 788 855 Tunisia A 222 234 244 251 247 229 B 234 260 294 330 365 400 Latin America A 11,043 11,677 12,241 12,539 12,400 11,722 B 11,556 12,929 14,407 15,895 17,382 18,817 Brazil A 3,616 3,850 4,027 4,117 4,069 3,857 B 3,784 4,251 4,747 5,228 5,683 6,130 Colombia A 822 902 961 972 940 877 B 863 999 1,134 1,232 1,316 1,407 Mexico A 2,236 2,402 2,548 2,509 2,509 2,455 B 2,360 2,695 3,080 3,502 3,938 4,368 (') Births estimated on the basis of projected population and resulting birth rates in model (shown in Table 2), under assumptions A and B. (2) Assumptions used for projections: See Footnote (1), Table 1. (3) Not including Sabah and Sarawak. 1 t# Annex 1 Table 6 Estimated(l) Average Annual Number of Births Omitted if Projection A Instead of B is Achieved, 1970-2000 (Thousands) Region 1970-75 1975-80 1980-85 1985-90 1990-95 1995-2000 World Total 5,080 11,794 19,978 30,802 45,220 63,348 Developed Countries 256 805 1,167 1,579 2,193 2,877 Developing Countries 4,824 10,989 18,811 29,223 43,027 60,471 South Asia 2,388 5,533 9,652 15,222 22,628 32,487 India 954 2,291 3,927 6,106 8,890 12,620 Iran 67 162 289 474 711 1,020 Malaysia(2) 20 42 76 116 165 233 Pakistan 305 752 1,353 2,190 3,243 4,695 Philippines 92 222 398 637 958 1,373 Thailand 79 186 335 523 782 1,115 Turkey 58 136 239 372 533 740 Africa 736 1,696 2,883 4,479 6,643 9,366 Egypt 74 168 291 452 679 976 Ghana 23 52 90 141 210 299 Kenya 26 62 107 166 249 354 Tunisia 12 26 50 79 118 171 Latin America 513 1,252 2,166 3,356 4,982 7,095 Brazil 168 401 720 1,111 1,614 2,273 Colombia 41 97 173 260 376 530 Mexico 124 293 532 993 1,429 1,902 O Estimates obtained by difference between the number of births under projections A and 8, shown in Table 5. () Not including Sabah and Sarawak. have to take place in South Asia. In India alone the share of birth prevention would be more than 20% of that required for the entire group of LDCs. Table 7 gives the total number of live, births during the 30-year period 1970-2000 in the world as a whole, in the less developed countries, and in several large countries under alternative assump- tions A and B, and the number that would need to be averted to achieve projection A instead of B. During that period about 840 mil- lion live births would have to be prevented in the LDCs. Of that number, almost half (46%) would have to be prevented in the 14 countries mentioned in Table 7. In order to accomplish projection A, the number of women who would have to practice contraception each year through the period 1970-2000 has been calculated and presented in Table 8. In 1970-75 it is estimated that an average of about 19.3 million women (or 3.3% V47 Annex 1 Table 7 Projected Number of Births and Births Omitted, 1970-2000 (Millions) Total Number of Births 1970-2000 Region Projection A(') Projection BC') Births Omitted Percent(2) World Total 3,862 4,744 882 100.0 Developed Countries 653 698 45 5.1 Developing Countries(3) 3,209 4,046 837 94.9 South Asia 1,480 1,920 440 49.9 India 675 849 174 19.7 Iran 41 54 13 1.5 Malaysia(4) 11 15 4 0.4 Pakistan 189 251 62 7.0 Philippines 56 74 18 2.0 Thailand 48 63 15 1.7 Turkey 41 51 10 1.1 Africa 476 605 129 14.6 Egypt 45 58 13 1.5 Ghana 13 17 4 0.4 Kenya 16 21 5 0.6 Tunisia 7 9 2 0.2 Latin America 358 455 97 11.0 Brazil 118 149 31 3.5 Colombia 27 35 8 0.9 Mexico 73 100 27 3.1 1) Assumptions used for projections: See Footnote (1), Table 1. (2) Ratio of births omitted in each subdivision to world total. (3) During the 30-year period, there would be 4,325 million births in the LDCs under conditions of almost maximum human fertility (constant fertility at the levels of 1965-70, or about 200 live births per 1,000 women 15-44 years); in all, 1,116 million births would have to be averted to make projection A a reality. Of that number, it is expected that about 279 million will be avoided if present contraceptive practices are maintained. Prolection A instead of B will be achieved in the LDCs only if about 837 million additional births are avoided in the three decades. (4) Excluding Sabah and Sarawak. of all women 15-44 years of age) would need to be practicing contra- ception each year in the LDCs in order to prevent 4.8 million live births. The number of women practicing contraception in a year would have to increase to 117 (13.9% of those 15-44 years old) in 1985-90 and to 242 million (22.5%) in 1995-2000. These calculations were made on the basis of the projected age structures produced by the model, and on the assumption that four women receiving protection for one year are needed to prevent one live birth. If fertility is assumed to continue constant' at present levels to the year 2000, it is estimated that the proportion of the female 'Almost equivalent to a low estimate of maximum human fertility, or 200 live births per 1,000 women 15-44. 3389 Annex 1 Table 8 Total Average Annual Number of Women 15-44 and Estimated Proportion of Contraceptors Required to Achieve Projection A, 1970-2000 (Thousands) 1970-75 1985-90 1995-2000 Women Contraceptors.'- Women Contraceptors r Women Contraceptors a. Region 15-44'-, Number % 15-44 Number % 15-44 Number 7u World Total 820,840 20,320 2.48 1,123,684 123,208 10.96 1,381,432 253,392 18.34 Developed Countries 240,618 1,024 0.42 272,010 6,316 2.32 293,040 11,508 3.93 Developing 576,982 19,296 3.34 841,728 116,892 13.89 1,076,165 241,884 22.48 CountriesO' (26,700) (5.0) (130,000) (16.2) (320,000) (30.8) South Asia 249,125 9,552 3.83 375,470 60,888 16.22 494,530 129,948 26.28 India 120,622 3,816 3.16 176,860 24,424 13.81 225,312 50,480 22 40 Iran 5,920 268 4.53 9,790 1,896 19.37 13,528 4,080 30.16 Malaysia,4 2,125 80 3.76 3,358 464 13 82 4,355 932 21.40 Pakistan 26,702 1,220 4.57 45,450 8,760 19.27 62,748 18,780 29.93 Philippines 8,365 368 4.40 14,010 2,548 18.19 18,938 5,492 29.00 Thailand 8,245 316 3.83 13,048 2,092 16.03 17,232 4,460 25.88 Turkey 7,628 232 3.04 11,508 1,488 12.93 14,698 2,960 20.14 Africa 77,315 2,944 3.81 115,150 17,916 15.56 148,178 37,464 25.28 Egypt 7,498 296 3.95 11,300 1,808 16.00 14,875 3,904 26.24 Ghana 1,988 92 4.63 3,085 564 18.28 4,035 1,196 29.64 Kenya 2,522 104 4.12 3,775 664 17.59 4,988 1,416 28.39 Tunisia 1,108 48 4.33 1,778 316 17.77 2,388 684 28.64 Latin America 63,888 2,052 3.21 98,268 13,424 13.66 126,842 28,380 22.37 Brazil 21,420 672 3.14 33,205 4,444 13.38 42,505 9,092 21.39 Colombia 4,802 164 3.42 7,715 1,040 13.48 9,812 2,120 21.61 Mexico 11,185 496 4.43 17,705 3,972 22.43 25,725 7,608 29.57 v Female population 15-44, estimated on basis of projected age structures as obtained by model projection Al. * 2 Number of women who would need to practice contraception during year(couple-years of protection), in order that projection A is achieved instead of B, is obtained by assuming that four couple-years of protection are needed to avert one birth ,a, Numbers in parentheses represent the total contraceptors needed to achieve projection A, as compared to a situation of con- stant fertility at 1965-70 levels. r Not including Sabah and Sarawak. Source- World Bank. computed on the basis of data from Tomas Frejka < Population Council, "Alternatives of World Population Growth," monograph in process of publication. population in reproductive ages who would have to use contracep- tion during the year to achieve projection A in the LDCs as compared to potential maximum human fertility would be 5% in 1970-75 (26.7 miilion), 16.2% in 1985-90 (130 million), and 30.8% in 1995-2000 (320 million). The estimates in Tables 7 and 8 give an indication of the magnitude of the task required in the next 30 years to reduce fertility in the less developed regions of the world to the levels implied by assumption A. A second important consideration is the changes in age structure q Annex 1 of the population which alternative fertility declines A and B would bring about. In Table 1 the proportions of female children under 15 are given for the world and the LDCs. Also, the dependency ratios which would result at different points in time are presented. The proportion of the female population under 15 is a consequence of prevailing fertility conditions. In the LDCs the proportion is high- on average in excess of 40%. If in the next 30 years fertility decreases are of the magnitude stated in assumption A, the proportion of fe- males under 15 will gradually decrease to 38% by 1985 and about 31% by 2000, stabilizing at about 20% in the following years. If fer- tility declines are closer to assumption B, changes in age structure will be much less noticeable, and by 2000 the population under 15 years will still be 38% of the total. The slow change in age structure will also have implications for the dependency ratio. In the developed countries this ratio indicates an age structure more favorable to the growth of productive activi- ties; therefore, changes in age structure resulting in a decrease of this ratio are considered desirable for development and improvement of economic conditions. If assumption A can be made a reality in the next 30 years, the dependency ratio in the LDCswill drop from 82% at present to 72% in 1985, and to 56% in the year 2000. These ratios would be similar to those prevailing now in the developed countries. If assumption B dominates the trends, however, the change in the dependency ratio will be considerably more moderate, remaining almost unchanged up to 1985 and decreasing to 73% by 2000. RAn Annex 2 Specific or General Population Targets in 27 Countries A. Countries with Specific Goals or Targets 1970 Population Estimated 1970 Specitic Country (Millions) Crude Birth Rate(') Goal or Target Africa Egypt 33.3 37 Reduce CBR by one point per year in 1970s. Mauritius 0.8 26 Reduce CBR from 30 to 20 by 1975. Morocco 15.5 50 Reduce CBR(from about 50in 1968) to 45 by 1972 and 35 by 1985. Tunisia 5.0 41-42 Reduce CBR from 43 in 1968 to 34 in 1975. Western Hemisphere Barbados 0.3 30 Enlist 60,000 women in program in 3 years. Dominican Republic 4.3 45-48 Reduce CBR to 28 in 10 years. Jamaica 1.9 32 Reduce CBR from 34.2 per 1,000 in 1968 to 25 per 1,000 by 1976. Trinidad & Tobago 1.1 28 Reduce CBR to 19 per 1,000 by 1977. Asia China, Republic of 14.7 27 Initial target was to reduce the rate of growth from 3% to 1.9% by 1973. Current target is to reduce the CBR from 27.1 at end of 1970 to 24.4 by 1976. Korea, Republic of 31.9 30-32 Achieve a 2% population growth rate by December 1971 and 1.5% by 1976. India 552.0 40 Reduce CBR to 32 by end of Fourth Plan (1974) and to 25 six to eight years later. Indonesia 118.0 40-45 Reduce birth rate. Current five-year plan sets a target of six million acceptors. Iran 28.7 46 Decrease annual population growth rate from 3% to 2%. Malaysia 10.4 37 Achieve 2% population growth rate by 1985. Nepal 11.1 n.a. Tentative target of 2% population growth rate. Pakistan 134.0 45-50 Recommended goal for Fourth Five- Year Plan (1970-75) is to reduce CBR from estimated 41-43 in 1970 to 33.2 in 1975, with 31% of fertile couples effectively using contra- ception. Singapore 2.1 22 Reduce CBR from 32 in 1964 to below 20 in the five year National Family Planning Program, 1966- 70. Thailand 35.7 40-45 Reduce population growth rate from 3.3% in 1970 to 2.4% by 1980. (continued) tA1 Annex 2 B. Countries with General Goals 1970 Population Estimated 1970 Country (Millions) Crude Birth RateM" General Goal Africa Botswana 0.7 n.a. Reduce population growth rate. Ghana 9.0 50 (1969) Reduce the population growth rate and give Ghanaians choice of family size. Kenya 10.8 50 Provide family planning services to combat malnutrition, improve ma- ternal and child health, and slow rate of population grqwth. Nigeria 66.4 50-55 Provide information, facilities, and services to families to enable them to achieve desired family size. Western Hemisphere Puerto Rico 2.8 26 Provideservices through the Family Planning Association. Asia Ceylon 12.5 32 Government's position has shifted from reducing birth rate (1965) to support for family planning pro- grams in the interest of maternal and child health. China, People's Rep. of 750-950 36-40 Degree of expressed official interest varies, but fertility control services are widely available. Government advocates late marriage and small families. Philippines 38.4 44-50 All agencies providing services are tocooperatein makingfamily plan- ning efforts effective. Turkey 35.5 40 Population Planning Law No. 557 states that every Turkish couple should have the number of children it wishes. (1) All crude birth rates (CBR) are per 1,000 population per year. na. =Not available. Sources: United Nations, 1970 Demographic Yearbook. Nortman, Dorothy, "Population and Family Planning Programs: A Factbook," 1971 edition, Reports on Popularion/a8milyPlanning. The Population Council, New York, June, 1971, tables 4 and 5, pp. 5-15. A few of the crude birth rate figures and goals are based on recently available information. Annex 3 CONTRACEPTIVE METHODS Of all the species, only man can control his fertility. This unique ability is exercised by interfering with the biological consequences of sexual intercourse at any one of several steps in the process of conception, fetal development and birth. Various methods of preventing conception have long been known and practiced: coitus interruptus, male sterilization through castra- tion, and various folk preparations of uncertain effect and reliability. Within recent years, there have been significant advances in knowl- edge of reproductive biology and in the development of contracep- tive technology. The two most widely known and used of the newer methods are, of course, "the pill" and "the loop," but other promis- ing methods are under trial or development. With increased research, still more and better methods can be expected. Family planning, or birth control, is the application of scientific knowledge by couples to regulate the number and spacing of their children. It makes birth a process of choice rather than chance. All scientific methods of accomplishing this-whether surgical, medical, mechanical or behavioral-rely on interference at some point in a complex series of events which must occur in perfect harmony to result in pregnancy. This can be done by blocking the passages through which the ovum and sperm must travel to meet, by changing the hormonal balance to prevent ovulation or sperm formation, by altering the wall of the uterus so that it becomes nonreceptive to the ovum, or by limiting sexual intercourse to times when no ovum is exposed to fertilization. Some methods, such as male or female sterilization, diaphragms, oral contraception and intrauterine devices, require medical super- vision. Others, such as the condom, rhythm, vaginal creams, jellies and suppositories, do not. A good contraceptive must be reliable, effective, and simple to use. It should not interfere with sex activity or be harmful to either partner. It should not affect fertility unless used as a terminal method. Finally, it must be inexpensive and easily supplied to large numbers of people. All methods now known have some disadvantages, and all have their failures. There is no perfect contraceptive. The method chosen must therefore be a matter of compromise and personal pref- erence. Each of the main categories of methods is described below. Annex 3 Oral Contraceptives (The "Pill") From ancient times, there has been speculation about the use of medicines taken by mouth to prevent childbirth. It was not until the early 1950s, however, that a scientific oral method became available. Pincus and Rock had worked on experiments with hormone prepara- tions and developed their "pill," which was first tried out in Puerto Rico beginning in 1956. This work was based on experience in the use of certain hormones in the treatment of women's diseases, for regulation of the menses, for severe pain during menstruation, and for subfertility. During normal pregnancy, the ovary produces certain hormones which help the fertilized egg to continue to develop in the womb. These hormones in the blood stream decrease the capacity of the pituitary gland to produce other hormones which cause the ovaries to develop eggs; the absence of the pituitary hormones causes ovula- tion to be suppressed. The "pill" produces a hormonal condition similar to pregnancy, in that no egg cells are released. The original "pill" was a combination of two types of hormones-one stops egg production, the other helps menstruation to occur. Several brands now on the market vary slightly in composition but also employ two hormones to achieve the same results.' The routine requires that 20 or 21 tablets be taken orally every month, beginning on the fifth day of the menstrual cycle (counting the first dayof the period as day one). A "sequential" pill has also been developed which is taken every day of the month without interruption. When the course is finished, menstruation will occur within two to three days, and the course will be repeated again from the fifth day of its start. For women who faithfully take these pills, success is 100%. Side effects such as nausea and vomiting, fullness of breast, and headaches have been reported, but most of these cease to occur in time. Weight changes may also occur. Medical complications in association with the use of these oral progestagens, e.g., thrombo- embolic conditions, etc., are worth attention, but it has still to be shown that they occur more frequently among pill users than among other women of similar age and history. Other contingencies, largely hypothetical, are also much discussed, but these questions will not be resolved until enough experience of prolonged use by women and other species has accumulated. The "pill" is already aesthetically acceptable, however, because its use is removed from the time of the sex act. It is the most widely used con- traceptive among women in developed countries, and in some of the 'There are newer varieties being developed with other action; these are discussed below under "Prospective Methods." Annex 3 national programs of developing countries. The discovery of oral contraceptives has revolutionized family planning methods. The Intrauterine Device (IUD) The use of small devices inserted and left in the womb is a revival and modernization of older methods, such as metal buttons, stems and Grafenburg Rings, which were used about 40 years ago. They were frowned upon by most medical people on the ground that they might cause infection or cancer. Modern IUDs are made of stainless steel or plastic impregnated with barium sulphate, which makes them visible under x-rays. They are made in a variety of shapes. The more well-known are the Margu- lies Spiral, Lippes Loop, Birnberg Bow, and Ota Ring. These devices are threaded through an inserter which straightens them for easy introduction into the uterus; after they are released in the womb, they regain their original shape which helps to keep them in place. Inser- tion is usually quite easy, and no anesthetic is needed for women who have had children (IUDs are not recommended for others). Re- moval normally requires medical attention. The IUD is fairly effective, with a pregnancy rate of 4% at the end of one year of use. The failure rate declines with subsequent years of use. It is superior to more conventional forms of contraceptives and is inexpensive, with no recurrent cost. Once fitted, it can be left in place for at least two years; in fact, no ill effects have been reported from cases of much longer use. The IUD is a simple method, once inserted, and requires no fur- ther motivation. It is also the only reversible form of contraception which does not rely on the user for its success, and therefore is more suited for couples who are not constantly motivated. The IUD, like the pill, does not require any precoital or postcoital preparation. The main disadvantage, aside from the rather small chance of pregnancy, is that some women who cannot tolerate the presence of foreign bodies in the uterus suffer cramps and various degrees of bleeding. Some women also cannot retain the IUD, and expel it. Permanent Methods Sterilization renders a person incapable of reproduction, and the result is usually permanent. Although there are many ways to render a person sterile, in practice the methods most commonly employed are to cut and tie the fallopian tubes in women (tubal ligation) or the spermatic ducts in men (vasectomy). Sterilization may be advised in cases of ill health as well as a means Annex 3 of family limitation. Because of the method's finality, it never gained much acceptance until the recent awakening to the population prob- lem. Even today, the acceptability of both male and female steriliza- tion varies greatly among different countries. Both popular and professional views as to how many children a couple should have before considering sterilization have changed in recent times. Whereas formerly doctors in many areas were loathe to agree to such operations for families with less than six or eight children, it is likely that most today would accept three or four as a more realistic figure. Sterilization below the age of 30 is generally discouraged. Each case must be considered on its merits, taking ac- count of all relevant factors: the number, sex and health of the fam- ily's children, the couple's economic circumstances, and experience with other family planning methods. The exact legal position varies in different countries, but generally voluntary sterilization is permissible if the couple fully understand the nature and permanency of the procedure. Cases considered for sterilization are those of couples who have reached the desired family size and where both partners give their unequivocal consent. The operation must be done in good faith and for the welfare of the family concerned. In men, sterilization is performed by blocking the seminal duct through which the sperm move out of the testicles to join the seminal fluid. This simple operation is called vasectomy and does not require a general anesthetic or hospitalization. It can be completed within 10 minutes under a local anesthetic on an outpatient basis. Only a half inch incision is made on each side through the skin in the scro- tum to cut and tie off the vas deferens, just under the skin. Nothing is removed and there is no change in sexual drive or capacity. In women the operation, though not a serious one, does necessi- tate hospitalization and is done under general anesthesia. It is con- veniently performed within 24 hours after delivery-ideally as a postpartum procedure. A small abdominal incision is made in order to tie the fallopian tubes which carry the ovum to the womb. The patient is usually discharged from the hospital in four to six days, and can gradually resume her normal activities after the usual recovery period advised after childbirth, i.e., about two weeks. The woman's chemistry is not affected and she will continue to have her menses normally. Nothing has been removed and her physical well-being and sexual characteristics are not affected. It is possible to carry out this operation intra-vaginally, but the procedure is more complex and requires highly competent surgery; it is not yet suited for general use. Research is under way to develop )3 AC Annex 3 techniques for blocking the tubes by chemical injection through the cervical canal. Traditional Methods Other contraceptive devices of varying efficiency, depending on the motivation, knowledge and interest of the user, are the condom for men and the vaginal diaphragm and spermicides for women. The diaphragm, which was used most successfully in developed countries before the advent of the pill, has several disadvantages, such as the need for a suitable housing environment, pelvic exam- ination and fitting by a physician, and elaborate preparations for use. Spermicides are chemical products in the form of jellies, creams and foams which have the effect of immobilizing the sperm on con- tact. These are preparations relatively easy to use, but they are not so effective. Rhythm Method In virtually all societies, it has been known that women could con- ceive only during a certain part of the menstrual cycle. This fact is the basis of the rhythm method of avoiding conception. Early ideas about the fertile and sterile periods of the menstrual cycle, however, were often mistaken. Much has been learned about them in recent years. Two varieties of the rhythm method are now practiced: calendar rhythm and temperature rhythm. Calendar rhythm, developed in the 1920s, depends upon estimation of the day of ovulation by a formula based on the individual woman's menstrual history recorded over a number of months. Abstinence is prescribed for a few days before and after the estimated day of ovulation. More recently the temperature rhythm, based on the principle of the rise of basal body temperature at the time of ovulation, has been used to determine that ovulation has occurred; marital relations are permitted during the postovulatory phase only. The rhythm method is based on the avoidance of coitus when it could result in the simul- taneous presence of a fertilizable ovum and mobile sperm. The contraceptive effectiveness of the rhythm method has been the subject of much controversy. Correctly taught and understood, and consistently practiced, it may be quite effective, especially the method based on temperature records. Successful practice, however, requires considerable self discipline and an equally strong desire to control fertility. Self-taught rhythm, haphazardly practiced, is a very ineffectual method of contraception. It has never been used on a mass basis. Annex 3 Abortion Abortion is any termination of pregnancy before the 28th week, the stage before the fetus is regarded as viable and capable of an independent existence. Pregnancies can be medically terminated for medical and social reasons. Pregnancies can also abort sponta- neously (if before the 28th week, an abortion; if later, a miscarriage); it is estimated that about 10% terminate this way, usually in the third month or earlier. The traditional method of terminating early pregnancy (before 12 weeks) is by the operation termed dilation and curettage (D and C), which involves the scraping of fetal contents from the uterine wall. This operation is usually done under surgical operating conditions, though it can be done as an outpatient procedure if the pregnancy is of less than 12 weeks. More modern is the use of a "suction method" to evacuate the uterus. This relatively simple and safe out- patient method was developed in Eastern Europe. Its use is increasing rapidly in many countries where abortion laws have been liberalized. Therapeutic abortion is sometimes permitted if doctors of consult- ant status agree that a woman's life might be jeopardized or her health endangered or seriously impaired should the pregnancy con- tinue. Social abortions are usually authorized by regulatory tribunals, although recent legislation in the United Kingdom and some jurisdic- tions in the United States allows the procedure to be practiced almost "on demand." Since the last war, Japan has brought down its birth rate to a very low level by legalizing abortion on medical or social grounds. The idea behind more liberal abortion laws is that a woman should have the right to decide whether to give birth, and that legal abortion in a hospital will be much safer than an illegal one. The rate of illegal abortions is difficult to estimate but ranges from 10% to 30% of births, depending on the country. The incidence of such abortions, which cause much maternal morbidity and even death, can be reduced by more liberal attitudes towards abortion itself and by encouraging contraceptive practice. In some countries, there are so many illegal abortions that strong efforts are now being made to encourage family planning as an alternative. Prospective Methods There is common agreement that additional research on the bio- medical basis of contraception is needed. The world total of expen- diture in this area is now in the range of $50 million a year. Biomedical 348 Annex 3 research is of two main types, both of which are essential to further progress: 1. Reproductive biology covers the anatomy and physiology of reproduction. This work, which is largely fundamental and not ap- plied in nature, is often unrelated to population control objectives. It nevertheless provides the basic knowledge of the full chain of events involved in reproduction. Such knowledge is far from complete. 2. Contraceptive development is goal-oriented applied research. Even though not all links in the reproduction chain are delineated, many areas are sufficiently well understood to support present con- traceptive techniques and to suggest some promising new leads. These can be translated into contraceptive products if they meet standards of safety and acceptability. They point to the following future possibilities: a) Hormonal method: a monthly pill, a postovulatory progestin that induces menses; a weekly pill which affects the lining of the uterus and prevents ovum implantation; a precoital pill which acts by causing thickening of cervical mucus which acts as a barrier to sperm passage; a long-acting injection which provides a more satis- factory performance than reported with medroxy progesterin acetate and related hormones; and a long-acting skin capsule, a progestin implant which causes infertility for up to one year. b) Mechanical method: a high-performance, low-side-effects IUD -e.g., the copper T intrauterine device. c) Postconceptive method: Prostaglandins-a vaginal insert which causes termination of pregnancy. A number of these developments are in various stages of trial and testing They include: * Progestagen Skin Implants. Delivering progestagens to the system on a long term basis by under-skin implant offers promise. A method now in the intermediate stage of development requires the injection of capsules containing progestagen under the skin of the arm or hip. A quick, painless outpatient procedure, this method offers promise because it is economical and convenient, requiring implantation only once a year. * The Copper T IUD. This is a T-shaped IUD with a strand of fine copper wound around the stem; in trials it seems not to have some of the main disadvantages of other IUDs. Its users have been reported to be free of bleeding and cramps. Because of its shape it is not ex- pelled and the reported pregnancy rate is close to zero. The exact mechanism of copper's effect on the endometrium to prevent preg- nancy is not known. Like all IUDs, the "T" does not interfere with either ovulation or menstruation. 149 Annex 3 * Prostaglandins. A new type of contraceptive is provided by pros- taglandins, a body chemical of many uses. One of its characteristics is that it acts as a chemical abortifacient; it is therefore a postcon- ceptive method. Its action in producing abortion has been amply demonstrated, but it seems on the basis of present information that side effects and toxicity still need to be regulated. Prostaglandins offer the best promise when used intravaginally. One advantage of this or a similar contraceptive is that it can be used after a pregnancy has occurred, when motivation is high. It also requires no regulation of sex activity and greatly reduces the need for education. Medical Supervision and Assistance Required Contraceptive methods differ in the extent to which they need to be supervised and carried out by medically trained personnel. Some methods (e.g., the condom, foams, jellies, etc.) require no medical supervision and can be made available in outlets unconnected with health services. Other methods, e.g., abortions and sterilizations, re- quire highly trained, highly specialized medical personnel working under hospital operating room conditions. Between these extremes are methods for which some degree of supervision and assistance by medical or paramedical personnel (nurses, midwives, health visitors, etc.) is necessary, and contraceptive delivery normally relies heavily on the general health system. These conditions apply to both "the pill" and IUDs; well trained personnel are required to examine for possible contraindications, and to insert the IUDs. There is room for judgment, however, as to the degree of medical supervision and assistance required at delivery points in the system. This is an important point that affects the planning and staff- ing of a program's various delivery systems and the design of training courses for the different categories of staff involved in the provision of services. qr,n Annex 4 EVALUATION OF FAMILY PLANNING PROGRAMS As with any action program designed to produce attitude changes and influence behavior in population groups, the evaluation of family planning programs is based on the following general principles: a) Definition of program objectives; b) Selection of appropriate cri- teria to judge achievement; c) Decision on the design of evaluation; d) Collection and analysis of data; and e) Interpretation of findings for use by program administrators. The frame for the evaluation process is built upon the definition of a hierarchy of objectives. It comprises the ultimate objective, fer- tility reduction, and an array of descending and branching subsidiary objectives. The ultimate success of the program will depend on the extent to which the established targets are met and the validity of assumptions on which the various goals are based. In the evaluative process of a family planning program, the follow- ing progression may be identified: a) Evaluation of program efforts or inputs (personnel-years, visits, clinics, costs, etc.); b) Evaluation of intermediate results (number of new acceptors, continuing users, etc.); c) Evaluation of program efficiency (relation of intermediate results to various physical or financial inputs; this stage corresponds to an evaluation of process); and d) Evaluation of effect on the ulti- mate goal, fertility decline. Independent of the administrative arrangements under which an evaluation system operates, both continuous and periodic data col- lection will be required to develop quantitative measures and indi- cators of program achievement. The system should provide a steady flow of information for judgment of program performance, for stimu- lating new approaches tending towards program improvement, and for testing procedures, on a demonstration basis, before adopting them as standard for the whole program. In order to carry out the different stages of evaluation, a system combining continuous data collection with periodic sample surveys will be required. This Annex describes the technical means to accom- plish evaluation at all indicated levels. The accompanying diagram relates the various levels with the data collection mechanisms and gives an indication of time frequency and phasing of activities. A clear understanding of the functions of all the means of measurement indicated in the diagram will help considerably in designing and ad- ministering family planning programs in which decisions will be based on concrete knowledge about what is more efficient in rela- IS1 Components of Evaluation of a Family Planning Program TYPE OF EVALUATION MEANS FREQUENCY INITIATION SOURCE OF DATA INPUT DATA i (Costs, suppies, ac- CONTINUOUS 'counting) A EVALUATION OF EFFORTi (Visits, personnel, expen- ditures-cost-effective- ness) OUTPUIDAT ness) _ ~~~~~(Women attended, home CONTINUOUS _ PROGRAM STATISTICS| (Wori ~~~~~~~~~~~~~~~~~~~~~~~~~(EVALUATION UNIT) t (A & B) / visits, group meetngs) .LONGITUDINAL 2nd B. EVALUAIO OFBI MONTHLY YA INTERMEDIATE EFFECTS VFLLOW-UP SAMPLE I IEWS (Acceptors, continuing O ACCetrs users) ~ ~ ~ ~ ~ Logtrmlnitdnl ERSPECTV users) or periodic r-TIMEtSTUDIES OF OR rn1 C. EVALUATION OF TASKS WINICS t'i ~~~PROGRAM EFFICIENCY -INTERVIEWS WITHRIANL (Samples among eea VR ER pplation) D. EVALUATION OF ( FIC.IA VITAL CNIUU ULTIMATE EFFECTS STATISTICS AND (Fertility decline) \ CENSUS S BUREAU, OR ~~~~~~~~~~~~~~~~~~~~~~~~VITAL REGISTRATION \ OR OFFICE SAMPLE SURVEYS OF3r VITAL EVENTS 'SURVEYS OF KNOWLEDGE, ATTITUDE AND PRACTICE CONCERNING CONTRACEPTIVES AND FERTILITY CONTROL. x -ASSUMING THAT A SURVEY WAS CARRIED OUT PRIOR TO INITIATION OF FAMILY PLANNING PROGRAM. gv Annex 4 tion to the intermediate and ultimate goals of program acceptance and fertility decline. Evaluation of Program Effort Statistics maintained on clinic operation, number of visits, sessions, personnel time, materials and expenditures are measures of the effort made and should be considered as guides for the administrator in terms of cost controls, methods of storing and distributing contra- ceptives to clinics, personnel requirements, etc. Also, to an extent, data on the numbers of acceptors and their characteristics give a measure of the effort by field personnel to encourage women to at- tend clinics and adopt a contraceptive method. Record keeping of new acceptors and revisits is necessary for estab- lishing a follow-up system through which women may be reminded of their next appointments and home visits may be scheduled by field workers for women failing to keep appointments. A simple form to keep records of new and returning patients should be designed by the evaluation unit. A summary sheet of daily activities, ke~eping new and returning patients separate, will be needed as a basis for prepar- ing a monthly report to the district or state headquarters. The monthly reports are usually designed to be filled with pre- coded information for punching. Tabulations may then be produced at the central level with mechanical equipment or computer. Prompt distribution of monthly reports (within three weeks) as feedback to districts and clinics as well as summaries for the whole country are important for effective use of the information. Detailed cross-tabula- tions at the state and national levels could be produced annually. A system along these lines, once in operation, will provide current in- formation about program performance and will also serve to establish a follow-up of patients. In addition, it will be used as a frame for the periodic sample surveys referred to below. Until now, in the absence of better information, data on new ac- ceptors have been generally used as a main indicator of "program success." It is recognized, however, that the important fact in a family planning program is not so much the acceptance of a method of contraception by a woman, but continued use, or retention in the case of the IUD. Evaluation of Intermediate Effect The extent to which women in reproductive ages accept contra- ception, and the rate at which they continue to practice it, may be determined from sample groups of acceptors, followed up in longi- 353 Annex 4 tudinal studies for relatively long periods, or from periodic retrospec- tive surveys among acceptors. An estimate of the relationship between extent of contraceptive use and pregnancy prevention may be derived from these follow-up studies of acceptors by matching, i.e., by including in the study a control group of other women outside the program, with socio-demographic characteristics similar to those of women in the family planning program. Controls in this case, as in medical investigations, pose sensitive problems to be tackled by the evaluator with great caution. The relationship between contraceptive practice and pregnancy prevention may also be assessed by comparing subsequent fertility of women in the program with an estimate of their "expected" fertility performance in the absence of contraceptive practice. Follow-up sample surveys among clinic acceptors will provide a measure of use- effectiveness of contraceptives; also, they will allow calculation of rates of discontinuance, determination of reasons for discontinuance and explanations for failures, and will help in correcting procedures and redirecting efforts for recruiting acceptors. A system of evaluation must be set up and conducted under the same principles ruling scientific research, with constant awareness that results are to be utilized by the administrator to increase effi- ciency in relation to the established goals. The ability of the adminis- trator to make purposeful use of these results is a key issue in the whole exercise of evaluation. Evaluation of Program Efficiency In the diagram of components of evaluation, the assessment of program efficiency is a separate activity ascribed to the evaluation unit. Its objective is to obtain a measure of the adequacy of the serv- ices. The proposed mechanism for this measurement is a special study to be made every two years to determine effectiveness of the family planning personnel in meeting the needs of the public and perform- ing the intended tasks. Time studies in which skillful individuals observe portions of tasks for short periods of time and take note of all movements and details of service performance, are appropriate for this assessment. As in the case of follow-up surveys, rules of scientific research apply to these studies. They should be based on a sampling scheme of clinics, personnel, and time intervals during the day, and should strive for representativeness and minimum bias. A time study of this type may require as much as a year for design, development of meth- odology, actual surveying and analysis. In the last few years, however, some universities in the United States have carried out studies of this 354 Annex 4 type in several developing countries, and a methodology has been developed which, with proper alterations for particular situations, may allow replication of the study in approximately three months. Thus, it might be feasible to perform one time study every two years as part of the general evaluation system. In a large-scale family planning program, there is pressure to oper- ate within a limited budget. Efficiency of the program, i.e., getting the job done well at the most reasonable cost, is important. Cost-effec- tiveness is measured by the ratio of input to output; inputs are given in monetary units and outputs in units of work in the program, e.g., number of acceptors for the clinics, number of visits made for the field workers, number of meetings held, number of persons attending meetings, etc. In order to allow for analysis and comparison of costs per unit, it is helpful if activities are classified in broad groups for accounting purposes. For instance, the groups may be: a) Administra- tion; b) Clinic and medical services; c) Field work and community organization work; d) Public information and mass communication; e) Training of personnel; and f) Research and evaluation. Some of the major categories above could be subdivided into a few sub-categories for principal types of expenditures, e.g., personnel (medical, non-medical), maintenance of buildings and equipment, contraceptive and medical supplies, and other supplies. This informa- tion from the accounting system would give the nominator for calcu- lation of the cost-effectiveness ratio, using as denominator data on number of "units" produced by the major activities of the program. Comparisons of cost-effectiveness ratios between clinics, districts (for field work), and training centers, would provide evidence to pinpoint operational problems. These data would also serve to calculate over- all cost-effectiveness ratios for the program. Evaluation of the Ultimate Effect-Fertility Decline Information about existing fertility levels and patterns in a country are obtained from the following sources: a) Censuses; b) Registered vital statistics; c) Sample surveys of vital events; and d) Sample surveys on knowledge, attitude and practice concerning contraceptives and fertility aspirations. Censuses provide valuable material for the calculation of fertility measures, through the application of special demographic tech- niques. In many developing countries with deficient vital registration systems, censuses taken around 1970 will be the main source of infor- mation for establishing changes in fertility and mortality that have taken place during the decade 1960-70. Detailed census tabulations needed for these analyses are delayed in many countries, however, . r_ r_ Annex 4 due to problems of enumeration, requiring manual verification of questionnaires, and to scarcity of computer programming ability. In many developing countries, the registration of vital statistics is incomplete or lacking entirely. In recent years, efforts have been made to remedy this deficiency by carrying out sample surveys which are also used as a basis for establishing or improving permanent reg- istration. Detailed tabulations on births by age, sex, parity, and other socio-economic characteristics are important to calculate refined measures of fertility, such as age-specific fertility rates, gross and net reproduction rates, total fertility rates, etc. Very few countries have good registration systems, however, or timely processing of data. Sample surveys for registration of vital events are the most expedient solution to this problem, but their design and implementation are costly and time consuming operations requiring trained personnel. Recent surveys carried out in developing countries with U.N. assistance are producing estimates of vital rates on the basis of area and household representative sampling and periodic interviews (one every three or four months) to register all vital events in the household during the interval. Usually, each house- hold in the sample is interviewed four or five times during a period of 12 to 15 months. Knowledge, Attitude, and Practice (KAP) Surveys In the last few years, several countries have conducted special sur- veys on the knowledge, attitudes, and practices of people concerning reproduction and contraception (KAP surveys). These surveys play an important role within the system of evaluation of a national family planning program. Data on the fertility, attitudes and contraceptive practices of the population in general are essential as measures of control and comparison with the results obtained from studies of acceptors. Therefore, KAP surveys should normally be an integral part of any evaluation system. Main groups of variables to investigate, common to KAP surveys and follow-up studies of acceptors of family planning programs, have been set up by an international committee of experts convened by the International Union for Scientific Study of Population and the United Nations1 (1966-70).Thesevariables maybe grouped as follows: * Data on fertility (complete marriage and reproductive histories). . Measures of the "intermediate variables" (according to the ana- lytic framework developed by K. Davis and J. Blake), that is, those through which any social variables should work to affect fertility. 'United Nations, Variables and Questionnaire for Comparative Fertilit Surveys, Depart- ment of Economic and Social Affairs, Population Stuidies No. 45, New York 1970. h1; Annex 4 These include, for instance, age at marriage (or at first sexual union), periods of separation, use of contraception, fecundity and its impair- ments, and natural or induced abortions. * The social norms about these intermediate variables, as well as the norms about family size and child spacing. * The elements of the social and economic organization which affect the norms, the intermediate variables and fertility itself (educa- tion, rural-urban background, women's economic activity, husbands' occupation, religion, ethnic language, extended versus nuclear family, income, etc.). Fertility Indicators Several indicators of population dynamics are derived from the data of censuses, vital statistics, and sample surveys of vital events. They serve to define the fertility of a population in terms of magni- tude, trends and differentials. The most common and elementary measure of fertility is the crude birth rate; it represents the number of live births in a population group during a given year divided by the population at mid-year (person-years of exposure). More refined measures of fertility are obtained by restricting the denominator of the rates to the population actually at risk (number of women in re- productive ages, or, even more specifically, women married or in consensual unions). The general fertility rate (number of live births over female popu- lation 15 to 44 or 49 years) represents an improvement over the birth rate as a measure of fertility. However, since the frequency of child- birth varies markedly with the age of women, age-specific fertility rates describe even better the fertility pattern of a population. A more synthetic measure, based on the age-specific fertility rates, is the total fertility rate, which results from adding the former and multiplying the results by the age interval used (usually five years). It represents the average number of children born to each woman during her re- productive life. The gross reproduction rate (GRR) is constructed in the same way as the total fertility rate, but is restricted to female births, thus being an estimate of the average number of daughters born to women during the whole reproductive period. The net repro- duction rate is based on the same principle as the GRR, but it takes into account the mortality of the female population. The indicators mentioned above are some of the conventional measures of the level of fertility. They permit making comparisons between geographic areas as well as observation of changes through time. For the latter analysis, a series of observations at successive dates is needed. Interpretation of fertility changes through time is 17 Annex 4 complicated by the fact that in response to especial conditioning, couples sometimes advance or delay their bearing of children. This may cause sharp movements of yearly rates, with little or no effect on the total size of family that they eventually attain. In other words, a change in annual rates may not be permanent, but reflect a shift in the scheduling of births. This is particularly so in societies where con- traception is practiced to a large extent. Analysis of the fertility of cohorts of women (age-specific birth rates derived from an actual group of women completing child- bearing) is being used increasingly in fertility studies of developed countries with good statistical information, dating back many years. For the less developed countries, however, data on reproductive his- tories are rare and comparisons through time will have to be based on periodic measures for a long time to come. Awareness of the use- fulness of cohort analysis, however, is important to stress improve- ment of registration of vital events, anticipating the need for more complex measures of fertility as contraception becomes an accepted practice. Impact of the Family Planning Program on Fertility Reference has been made to the measurement of program effect on the fertility of women attending clinics. A separate problem is the measurement of program impact on the level of fertility of the popu- lation in general. That is, if fertility indicators show a declining trend, the problem consists of explaining the fall in fertility in terms of sev- eral interrelated factors. A multivariate analysis with fertility as the dependent variable and program activity as one of the independent variables is appropriate in this case. Such analysis is, of course, con- ditioned to the availability of data. For this evaluation, a thorough understanding of the determinants of fertility is necessary in order to select pertinent variables, with explanatory power. One possibility would be to use geographical divisions in a country as units of analy- sis, with and without family planning programs or with measurable accomplishments in family planning. Information on vital events would also be necessary for each unit. Again, the problem of census data availability, with detailed tabu- lations by socio-economic characteristics of the population, is evident for evaluation purposes. For further refinement, a multivariate analy- sis could be performed, using as units of analysis individual couples in and out of the program. The sources of data for this analysis could be a KAP survey in the general population and a follow-up survey of clinic clients. A convenient way of dealing with the nominal character of client status and other unquantifiable variables is the use of dummy Annex 4 variables, applied often in sociological studies. Regression analysis offers the technical tool to judge the significance of the contribution of several variables (including clinic attendance) to the fertility level. Other less sophisticated ways to estimate what would have hap- pened in the absence of a family planning program may be suggested. One possibility is to compare actual present fertility trends with those estimated before the family planning program began, or with trends projected on the basis of experience in the past 10 years (which in some developed countries are showing slight declines). Still another way of assessing program impact would be to compare the socio-economic characteristics of clinic clients with those of the general population. Demographic evidence indicates that rising urbanization, literacy, income, etc., are major factors accounting for increasing contraception. Family planning programs have as their main objective promotion of the practice of contraception in spite of rural traditions, illiteracy, low income, etc. A test can be made by comparing groups practicing contraception outside and within the program. If clinic clients are disproportionately among the poorer classes, then the program can be considered effective in terms of goal achievement. This brief outline of alternative ways of learning the effect of a fam- ily planning program on the fertility of a population suggests the mass of information and technical knowledge required of the evaluation unit. The first step in tackling the problem of evaluation is recogni- tion of the magnitude of the task, and review of the possible means available. Administrative Aspects In most family planning programs now underway, evaluation is gradually being recognized as a necessity, but very few countries have adopted effective systems to provide it. In fact, due to the improvised fashion in which many of these programs originated (usually in the private sector, with undefined goals but great enthusiasm and faith in the cause of preventing pregnancies), the need for evaluation became clear only after governments began to develop national family plan- ning programs, and investment of resources in these programs had to be justified in terms of results related to goals. Even then, the concept has often been limited, at least in part, to the evaluation of effort and of progress toward the intermediate goal (most often, number of acceptors only). The main obstacles to establishing a system capable of covering all evaluational needs can be summarized: * Insufficient understanding or recognition on the part of program icn Annex 4 administrators and governing bodies of what evaluation really is, what is required to implement it, and how it can be used to improve the program. * Unavailability or scarcity of professional expertise. * Reluctance to accept technical assistance, which sometimes has been better designed to benefit the research programs of foreign institutions than to build effective evaluation systems for national purposes. * Lack of adequate demographic information from censuses and vital statistics registration. Once the obstacle of insufficient understanding is overcome and evaluation is accepted by administrators of the program as a useful tool for their activities, suitable expertise must be brought to bear on the problem. Often this can best be done by arranging for an expert team in demography, evaluation and social research to plan a system, with clear provisions in their contract concerning the national char- acter of the operation. If foreign experts are employed, it will still be necessary to arrange for a team of national professionals, to be trained in all aspects of evaluation and research in an appropriate institution. Where foreign technical assistance is used, the national team should of course take over the operation after training, with an overlap of the two teams working together (for perhaps a year) before the total transfer of responsibilities. The whole process, from the planning stage to the complete take-over by the national team, may take as long as five years. Given the availability of professional staff, a fundamental question is the place this evaluation team will occupy in the system. The fol- lowing possibilities may be considered: * An external body, outside the family planning program, partially devoted to family planning evaluation but also engaged in other activities (e.g., a department in a local university). * The evaluation unit might be part of the national family planning organization, partially utilizing the same personnel and facilities. * A separate unit, completely dedicated to evaluation and research in the field of population and family planning-a population center or institute, along the lines of the center in the Republic of China. The first possibility should be treated with reservation. It implies a situation of ambiguity, in which evaluation might be diverted to other research or become too remote to the needs of administration. The second alternative runs the risk of not being objective enough to exert effective criticism from within. A separate unit appears to be the most workable and effective solution, provided that a strong link is estab- lished with administration of the program. Annex 5 BILATERAL, MULTILATERAL AND PRIVATE AGENCIES Twelve years ago, bilateral, multilateral and private organizations together devoted a total of about $2.8 million to activities related to family planning; in 1971, the amount was about $225 million. The proportion of such resources going directly to developing countries has steadily increased, reaching about 70% in 1969, com- pared with some 45% three years before. The most important components supported with external finance have been training and technical assistance (including salaries, allow- ances, etc., of foreign advisers); together, they accounted in 1969 for about 80% of the external resources allocated to family planning programs in developing countries. A wide range of technical assistance is provided. It includes medi- cal, paramedical, educational and administrative personnel directly involved in the distribution of family planning services; assistance in undertaking social, technical, psychological and demographic studies necessary to determine the kind of program which will be best ac- cepted by the population; assistance in collecting information and in program evaluation; and assistance in training local personnel in all fields of family planning. Another major component has been the supply of small equipment and contraceptives. This represented about 18% of the total in 1969. The balance-about 2% in 1969-has been allocated to buildings, schools, vehicles and fixed equipment. In the same year, $18 million was spent on bio-medical and social research and $6 million in sup- port of university training programs in the population and family planning fields in developed countries. A summary of principal agencies and their programs follows: Bilateral Programs Canada. The government of Canada began assistance to family planning in developing countries after repeal, in 1969, of legislation against advertising or encouraging the use of contraceptives. In December 1970 the Government announced its first official support for family planning activities abroad. Assistance includes funding the work of international organizations (e.g. expanded medical research by UNFPA) and direct assistance in priority countries. Grants have been made to Barbados and Colombia, and to WHO for contracep- tive research. Annex 5 Denmark. Since the mid-1960s, the Danish Government has pro- vided multilateral family planning assistance through a number of agencies, cooperated in family planning training courses, and sup- ported contraceptive research. It has also provided bilateral assistance in family planning to India, Egypt, Thailand and Uganda. A total of $807,000 has been granted to the IPPF, both for special projects and for its general program. The UNFPA received $400,000 during the 1967-71 period. Family planning training courses are given for students from devel- oping countries in collaboration with the Danish Family Planning Association. In 1968, Denmark made a grant of $72,000 for activities of the Egyptian Family Planning Association and provided $135,500 worth of raw materials for production of oral contraceptives. Federal Republic of Germany. The Government allocated $1.5 mil- lion to the United Nations Fund for Population Activities in 1970 and 1971. Other financial support has gone to the United Nations Devel- opment Programme, which was granted $250,000 in 1969, and to the population group in the OECD Development Center. The Govern- ment supported a multi-functional training and research center in Tunis in 1971. Japan. The Japanese Government, whose assistance to family plan- ning in developing countries began only recently, expanded its efforts considerably in 1969. A grant of $100,000 was made to the Interna- tional Planned Parenthood Federation in 1969 and 1970; in 1971, the grant was increased to $500,000. In 1971 Japan pledged $2 million to the UNFPA. In October 1970, it sent a family planning mission to Indonesia. Following the mission's recommendations, the Govern- ment began receiving Indonesian trainees and sending advisers and materials to that country. The Family Planning Federation of Japan, Inc., established in April 1968, provides contraceptive materials, equipment and other com- modities to the Republic of China, Indonesia, Republic of Korea and Nepal. Since 1967, seminars on family planning have been held for doc- tors, nurses and government officials from southeast Asian countries. They have been conducted by the Government's Overseas Technical Cooperation Agency and the Family Planning Federation. The Netherlands. The Government of the Netherlands has contrib- uted to the United Nations Fund for Population Activities and is pro- viding about $1.7 million to assist family planning activities in Indonesia, Kenya, Pakistan, and Tunisia. In Pakistan, the Netherlands supports research to gather data on motivation of the rural population towards family planning and to Annex 5 help implement Pakistan's family planning program. Under a three year project begun in 1968, five Netherlands nationals did research in selected villages. A 1968-73 project in Kenya includes training of local personnel in contraceptive uses, and clinical research on the applicability of vari- ous family planning techniques and the causes and treatment of steril- ity. A two year project including clinical research and training of medical personnel in Tunisia will help to implement the Tunisian Government's family planning program. Two grants have been made to Indonesia; one was for the National Training Center, the other for the National Family Planning Institute and two Indonesian univer- sities for social and clinical research. Norway has given assistance to family planning activities since 1964. Since 1970, its aim has been to allocate approximately 10% of total aid appropriations to such activities in developing countries, divided about evenly between multilateral and bilateral programs. Bilateral aid, primarily financial, will go mainly to Norway's priority countries that request assistance. In 1971 Norway provided $1.5 million to the UNFPA. It has also contributed to the IPPF and UNICEF for family planning purposes. Under the bilateral program, $1.1 million was granted to the post- partum family planning program in India. Norway also contributed clinical equipment to Kenya for the establishment of 50 family plan- ning units in health centers. Sweden. Family planning has been assigned the highest priority in Sweden's development aid program. Aid is extended through the Swedish International Development Authority (SIDA), and has grown from a single project in Ceylon in 1958 to assistance in materials, finance, and personnel to numerous developing countries. Disburse- ments in 1969-70 amounted to approximately $5.3 million. They rose to about $6.4 million in 1970-71, and were expected to total some $11 million in 1971-72. The share of Sweden's bilateral aid for family planning has increased from a few percentage points in the early 1960s to about 10% today. SIDA now furnishes supplies and equipment to some 14 countries. In three of these (Ceylon, Kenya and Tunisia) it also provides expert personnel for government programs. SIDA also has a special arrange- ment permitting governments and organizations to buy contracep- tives at reduced prices, made possible by volume purchases. The Government provides substantial allocations to the International Planned Parenthood Federation and various United Nations organiza- tions. Assistance has also been given to WHO for research in human reproduction. Annex 5 The United Kingdom. The United Kingdom has recently increased its aid to population programs through both multilateral and bilateral channels. The allocation for 1971 was $4 million. Grants were made to the United Nations Fund for Population Ac- tivities and to the International Planned Parenthood Federation. At the meeting of the Consortium on aid to India in May 1970, the United Kingdom pledged an interest-free loan of $2.4 million for local costs of the Indian family planning program. Technical assistance is being given to some 10 or 12 countries. A Population Bureau was set up in 1968 by the Ministry of Over- seas Development. The Bureau encourages training and research and helps to provide operational and advisory personnel for overseas pro- grams. A fertility research unit was established at the London School of Hygiene and Tropical Medicine in January 1970. A graduate course in medical demography, at the same school, commenced in Septem- ber 1970. Both programs receive assistance from the Government. United States. The largest budget for population assistance is that of the United States, operating chiefly through its Agency for Inter- national Development (AID). AID's budget increased from $50 mil- lion in 1969 to $75 million in 1970 and $100 million in 1971, compared with $4 million in 1966 and 1967. AID has supported bilateral programs in 33 countries and helped finance activities in many others through organizations such as the International Planned Parenthood Federation (IPPF), the Pathfinder Fund, the Populaticn Council and the United Nations. Assistance covers all major disciplines concerned-information, education and communication, demography, social research, and research in fer- tility control methods. The Agency assists in manpower development, including leadership and specialized technical training in evaluation and analysis. It also promotes institutional development. In 1971, AID provided contraceptives for family planning programs in more than 70 countries and made available supplies and equipment to a number of institutions. In addition, research on population and family planning in devel- oping countries is carried out in the U.S. Department of Health, Edu- cation and Welfare, largely by agencies under the Public Health Service. Some 100 Peace Corps volunteers have worked in population programs overseas. Multilateral Programs The Food and Agriculture Organization of the United Nations (FAO) is becoming involved in the population field in two areas: 1t4 Annex 5 policy-oriented research into the implications of population trends for agricultural development (food consumption and nutrition and their effects on fertility and mortality), and integration of family plan- ning into home economics education programs. The International Labour Organization (ILO) has a mandate to pro- mote information and education activities on population and family planning and to conduct policy-oriented research on the demo- graphic aspects of social policy in such fields as employment promo- tion, social security, and enterprise-level medical services in family planning. The United Nations. On the basis of resolutions adopted over the last several years, all major United Nations agencies have authority to undertake action programs in population and family planning. In 1969 several of these agencies took the first steps to develop and pursue operational programs, with financial support from the United Nations Fund for Population Activities (UNFPA). This Fund was estab- lished by the Secretary-General in 1967 to finance an expanded United Nations program in population and family planning. During 1969, the Secretary-General turned over management of the UNFPA to the Administrator of the United Nations Development Program (UNDP), the central technical and development assistance agency for the United Nations. By the end of 1969, more than $5 million in contributions had been pledged to the UNFPA, of which $4 million was provided by the United States. Other contributors were Denmark, Finland, the Nether- lands, Norway, Pakistan, Sweden, Trinidad and Tobago, and the United Kingdom. In 1970 the Fund received pledges of $15.4 million, and in 1971 $28 million. By the end of 1971, about 60% of these funds had been received for disbursements. Most were obligated for projects involving support for demographic training activities, research on demographic and population questions, advisory missions and tech- nical services to member countries, national program support for non-conventional equipment and supplies, information and docu- mentation, and infrastructure costs within the United Nations system. The Population Division, within the United Nations Secretariat, continues its program of demographic research and projections, tech- nical information services, and support for conferences and technical meetings, funded from the regular budget of the United Nations. In addition, the Population Division executes for the UNFPA, as do other elements of the United Nations system, the provision of technical assistance to countries in areas within its competence. United Nations Educational, Scientific and Cultural Organization (Unesco). Population programs are a new area for Unesco, and the Annex 5 work program is developing rapidly. Unesco's population and family planning activities increased from less than $500,000 worth of UNFPA-financed projects in 1970 to an estimated $3.5 million in 1972. The Organization's mandate covers the development of curricula and teaching materials, teacher training, the use of communications and related training, and research in population-related education, motivation, evaluation, etc. Unesco has sent advisory missions to a number of countries, provided consultants and advisors for national education and communications programs, conducted regional meet- ings and training courses and, through its regional office in Bangkok, produced sample population education materials. Proposed projects include seminars and workshops and an expanded program of re- search, including the development of simulation studies on demo- graphic growth and educational opportunity. The United Nations Children's Fund (UNICEF) has provided assist- ance for family planning since 1967, as part of its maternal and child health programs. Its contribution has been in the form of vehicles, equipment and supplies, salaries for teaching staffs, and stipends for trainees. Under an Executive Board decision in 1970, UNICEF is also able to furnish contraceptive supplies to countries on request. The World Health Organization (WHO) has a key role in carrying out an effective United Nations family planning effort. WHO has a mandate to work in the health aspects of human reproduction, family planning, and population dynamics, and is expected to assist coun- tries in the development of family planning activities within the con- text of health services. Private Agencies The Ford Foundation. The Ford Foundation has contributed sub- stantial funding for population work. Since 1952, it has committed about $147 million for this purpose, and has been an important force in three areas concerned with problems of population: research and training in reproductive biology, establishment and expansion of uni- versity population study centers in the United States, and assistance to family planning programs in developing countries. Major emphasis has been on reproductive biology, with some $80 million in grants going primarily for fundamental research and train- ing programs in 93 institutions around the world. Grants have been made to a dozen university centers focusing on population problems. While two thirds of the Ford Foundation's population expenditures have gone to American institutions, their activities in the field are mainly directed toward population problems in developing countries. 366 Annex 5 Since the mid-1960s, resources devoted to population work in de- veloping countries have grown significantly. The Ford Foundation has financed family planning work in 26 developing countries. In Asia and Africa it is assisting family planning action programs as well as training and research. In Latin America, emphasis is placed on the study of population problems and reproductive biology. The International Planned Parenthood Federation (IPPF). The IPPF assists the formation and effective operation of family planning asso- ciations and institutional affiliates throughout the world. It encour- ages and supports the training of medical and paramedical workers, sponsors workshops and seminars, and promotes and organizes inter- national and regional meetings and conferences. The Federation also stimulates scientific research in the fields of biology, demography, sociology, methods of contraception, fertility and subfertility, sex education, and marriage counseling. Established in 1952, IPPF is an association of autonomous national family planning associations. One non-governmental family planning association in each country is eligible for full membership, provided it is a national organization. In 1971 there were 79 members, includ- ing associate members and two affiliates. Information and assistance have also been given to non-member associations. The IPPF has six regional offices, and representatives for Africa in Nairobi and Accra. IPPF is financed largely by foundations, individuals and grants by governments. It has also received assistance from the UNFPA. Indica- tive of the growth of the Federation in recent years was the increase in its annual budget from $325,000 in 1962 to nearly $20 million in 1971 and its estimated program of $25 million in 1972. The Population Council. The Council was established in 1952 and is financed primarily by the Rockefeller and Ford Foundations. It pro- motes research, training and technical assistance in the social and bio-medical sciences, and serves as a center for the collection and exchange of information on significant ideas and developments in the field. Its activities include an extensive publication program. The Council initially confined its activities chiefly to fellowships and small demographic and bio-medical research grants. In the early 1960s, however, it began to give technical assistance to family plan- ning and population projects in developing countries. Its budget in 1970 totaled $17.2 million. It had a staff of about 230, of whom 30 served in 18 foreign countries. Through its Technical Assistance Divi- sion, the Council provides support to family planning programs in 17 countries. It also seeks to evaluate the effectiveness of programs. The Council's Demographic Division has assisted the United Na- tions Demographic Centers in Bombay, Cairo and Santiago, and 367 Annex 5 national centers in 15 countries. Grants have been made to numerous university departments and study centers in developing countries. The Demographic Division also conducts research on fertility, estimation of rates of population growth, population policy, and related topics. The Bio-Medical Division has focused much of its resources on specific research leads in various methods of contraception. In 1967 it began a project to establish fertility regulation by continuous pro- gestin therapy. The progestin-in this instance megestrol acetate-is being field-tested in 10 countries. Research on a number of other contraceptive leads is being done by an international panel of experts in six countries. Basic research on reproductive biology and other aspects of fertility regulation is also being conducted. The Council started its International Postpartum Family Planning Program in 1966 in 25 hospitals in 14 nations. The program has grown to include 150 hospitals and clinics operating in 14 countries in Asia, Africa, Latin America and the Middle East. The Postpartum Program is an international demonstration effort to provide family planning information and services immediately after childbirth to urban women of low socio-economic status, in settings where delivery is institutionalized in public hospitals and clinics. A world-wide follow-up survey has been completed. The Rockefeller Foundation. Although the Rockefeller Foundation gave support to bio-medical research in fertility control in the early 1930s, it was not until the late 1950s and early 1960s that the Founda- tion made major commitments to help solve population problems. Since 1963 it has provided more than $45 million, including more than $15 million in 1970. The Rockefeller Foundation concentrates increasingly on the inter- action of social, medical and biological sciences. It finances research, training and experimental programs in a broad range of fields rele- vant to population and, like the Ford Foundation, is making a major effort to stimulate basic research in reproductive biology. At the University of North Carolina, a grant was made to finance research on applying techniques of modern cellular and molecular biology to problems of fertility control. A new building to house the Harvard Laboratory of Human Reproduction and Reproductive Biol- ogy was made possible by a Rockefeller Foundation grant of $2 mil- lion to equip the Laboratory and provide staff over the next 10 years. Grants were made to Tulane, California, Columbia and Georgetown Universities. The Foundation has also supported research at the Lon- don School of Economics and in universities in Chile, Colombia, Indonesia, Mexico, Thailand and Turkey. Annex 6 RESEARCH Research in population is as varied as the disciplines concerned. It can be classified into four groups: * Demographic research. Research related to the basic demo- graphic variables-fertility, mortality, migration, and implications for the size, structure, and rate of growth of populations. A growing but specialized aspect of this research concerns the demographic assess- ment of the effects of fertility control programs. * Economic research. Economic implications of population trends, including the economic effects of programs, and the economic analy- sis of programs (cost-effectiveness, for example), etc. * Bio-medical research. This covers the entire field of contraceptive technology and reproductive biology, including related clinical trials. It involves basic physiological research, research on products and agents and their action, and also applied field trials for clinical testing, evaluation of side effects, etc. * Operational research. This involves research directed to opera- tional problems in selected disciplines, e.g., management, evaluation, communications, educational methods, and media. Research in all these areas has increased as interest in population matters has grown and more money has become available. The spec- trum of interests is very wide, not only in its interdisciplinary scope, but also in its range from purely theoretical topics to narrowly focused projects designed to solve day-to-day applied problems. Much of the planned and ongoing research is not directly relevant to operational needs. However, in a rapidly expanding field where research has been neglected in the past, it is not always possible to assess the longer term significance of research topics that may appear, in the short run, to be of no direct operational relevance. There is a clear need to expand the boundaries of knowledge about population mat- ters in all directions, while recognizing that only a limited range of topics is likely to fall within the immediate interests of the Bank. It is useful to distinguish three kinds of research interests for the Bank. There is a wide range of population matters which do not fall within the competence of the Bank, and are not a direct concern of day-to-day operations, but in which the Bank has an interest in seeing that knowledge and understanding are expanded through research. This can be done by making the Bank's interests known and encour- aging other, more appropriate, institutions to plan and finance the research. The most important group of subjects in this category covers the fields of basic research in human biology and reproduction, con- qr1 Annex 6 traceptive development, and other medical and health topics. It will be necessary to follow the progress of this kind of research and to consider, at an appropriate time, whether the Bank can make a con- tribution in a coordinating role. Initiatives being taken by WHO and the Population Council through its International Committee on Con- traceptive Research, in addition to those of the industry, represent an expanded effort in this field. Second, there are research topics that the Bank has an interest in pursuing but where the work cannot be done easily within the Bank, either because of staff commitments or because the work is not feasible with facilities available. As in other areas of activity, such research can be contracted out under staff supervision. In this respect some aspects of population research would be no different, in prin- ciple, from other areas in which the Bank has undertaken work through consultancy arrangements. An additional reason for such arrangements, however, may be the need to encourage the develop- ment of research institutions and capacity in the developing countries. This may be as important in the population field as obtaining actual research results. The third area of research activity covers those studies which can and should be done by Bank staff. This group is not always sharply distinguished from the kind of topic that might be handled on a con- sultancy basis. It will normally be identified by the criteria that it can be best pursued within the context of project, sector and country economic work, and will focus on problems of operational interest to the Bank. The objective of most Bank-sponsored research will be to provide information which will ultimately lead to the improvement of Bank operations. This objective suggests that research projects which are to be useful in this sense must be linked closely to population pro- grams and relate to program needs, including the generation of appropriate data for this purpose. It will often be the case that the most effective research will be done through local talent, which is likely to be most sensitive to local needs and priorities. This would in no way preclude the use of expatriate assistance in the planning and undertaking of many aspects of research. The essential need is for a partnership rather than a paternalistic relationship. Experience with family planning programs so far strongly suggests that a mini- mum of three to five years may be necessary before research can lead to meaningful conclusions. The planning, collection and analysis of the data required set the minimum period before research results can be used to improve program performance and measure results. 27n Annex 6 One of the Bank's main objects in the field of research will be to build into projects it finances a strong management evaluation com- ponent, linked in the most appropriate way to program administra- tion. The usefulness of data generated in such projects will be enhanced if it is collected in such a way as to make possible mean- ingful comparative studies. This is particularly important if progress is to be made in understanding the relationship between costs and results of family planning services in terms of the number of people served and possible effects in reducing fertility. This information does not now exist on any basis which makes possible sound generaliza- tions about the scale of expenditures needed to produce a given reduction in fertility. Part of the research program of the Bank, therefore, both internal and as contracted through consultancy arrangements, will focus on key aspects of the management of family planning programs. In par- ticular, it is proposed to study the methodological aspects of family planning programs including the cost-effectiveness, evaluation and cost-benefit aspects. Of equal importance is the need to study the cost patterns arising from different variations in family planning pro- grams, including the resultant budgetary implications of program development. A third area of research of interest for program opera- tion covers the role of the commercial sector in the provision of family planning services. These topics are those most closely related to the operational needs of the Bank, insofar as lending for family planning is concerned. How- ever, there are other areas of population research equally relevant for operational purposes but not directly connected with project activity. These include studies to support general economic analysis of the development prospects of member countries, and related sector analysis. They include the demographic and macro-economic aspects of development. In choosing topics to be pursued through consultancy arrange- ments, special attention will be paid to proposals for collaborative research with institutions in member countries. An important role in country project development will be to support related social re- search through national institutions; this could range from the provi- sion of positions, supportto programs, and establishmentof centersfor this type of supportive research. The objective of these proposals is to provide support for strengthening research capabilities in the devel- oping countries and, at the same time, ensure that research of impor- tance to both the country and the Bank is carried out by local talent. I 71 TOURISM Review of the Sector .......... ............................... 375 Recent Growth Trends of International Tourism .... .......... 375 Factors Affecting Vacation Travel ...... ..................... 376 Likely Future Tourist Flows to Developing Countries .... ...... 378 Patterns of Tourism Expenditure ...... ...................... 379 "Seasonality" .. ........................................... 381 Tourism Facilities .......... .............................. 382 Financial Aspects of Investment in Accommodation .... ....... 383 Planning for the Tourism Sector ..... ............. ......... 384 Foreign Exchange and Employment Effects of Tourism in Developing Countries ................ .................. 384 Some Social Effects of Tourism Development ..... ............ 386 World Bank Group Operations in Tourism . ....................... 387 Financing ..................... ......................... 387 Preinvestment Activities ......... ..... ................... 388 Lessons Drawn from Past Operations ......... .. ............. 388 The Economic Evaluation of Piojects ......... .. ............. 391 Other External Financial and Technical Assistance ..... ............ 393 Future Bank Group Operations ............... .................. 395 Financing Program .................. ..................... 395 Sector Work ..................... ....................... 395 Directions of Bank Group Financing ......... ...... ......... 396 Investment Incentives ................. ................... 397 Research ...................... ......................... 399 Annex Tables 1. Comparative Data on International Tourism in Selected Developing Countries, 1970 ..... ................ 400 2. Tourism Financing by the International Finance Corporation . 402 3. Commitments for Tourism Projects under World Bank Loans to Development Finance Companies ..... ........... 403 4. Direct Bank Loans and IDA Credits for Tourism Projects, FY1971-1972 ................ ......... 403 5. Projected Bank Group New Commitments for Tourism, FY1972-1976 ........... ..................... ......... 403 373 TOURISM SECTOR WORKING PAPER * This paper describes the growth of tourism throughout the world, the various factors which have affected this growth, and the prospects for developing countries to benefit more fully from tourism. Focus is primarily on vacation travel since business travel is far less responsive to an expansion of tourism facilities and is, therefore, of less interest for Bank Group operations. The paper recounts the role of the World Bank Group in encouraging the establishment and extension of the tour- ist industry in developing countries and outlines its plans for the future. The term World Bank Group includes the International Bank for Re- construction and Development, the International Development Asso- ciation (IDA) and the International Finance Corporation (IFC). Money amounts are expressed in U.S. dollars. The Group's fiscal year (FY) ends June 30. REVIEW OF THE SECTOR Recent Growth Trends of International Tourism Since the early 1950s international tourism' has grown rapidly, particularly after the liberalization of foreign exchange and travel restrictions which characterized the years following World War II. In the early 1960s international tourism, which in Europe had been mainly limited to neighboring countries, spread more widely and developing countries began to benefit increasingly from its growth. From 1950 to 1970 international visitor arrivals (including vaca- tioners2, business and other visitors) in all countries grew from 25 million to 168 million, an average yearly growth rate of 10%, while international tourism receipts rose from $2.1 billion to $17.4 billion, 11% per year. In some of the developing countries much higher growth rates have been recorded in recent years (see Annex Table 1), as these more distant destinations have become increasingly acces- sible with the expansion of air transport. There are three major tourist generating areas: North America (United States and Canada), Western Europe and Japan. About three- 'An international tourist, as defined by the United Nations and the International Union of Official Travel Organisation, is a person who visits a country other than that of permanent residence for at least 24 hours, whatever his motive for travel. 'A vacationer, whether international or domestic, is defined by major European tourist services as staying away from home for at least four nights on any one trip and may in some instances include tourists traveling for a combination of business and pleas- ure. An excursionist is a person who stays in any one destination for less than 24 hours. 375 quarters of all international visitor arrivals, including the great ma- jority of arrivals in developing countries, are accounted for by 12 countries-United States, Canada, United Kingdom, France, Germany, Sweden, Denmark, Belgium, Netherlands, Switzerland, Austria and Italy. Until 1967, foreign travel from Japan was severely limited but with the easing of restrictions in subsequent years, foreign holiday travel particularly has grown very fast. Factors Affecting Vacation Travel Only a few countries have tried to analyze the factors affecting the various types of tourist demand and this only in recent years. The studies, usually based on sample surveys, indicate that levels of in- come, income distribution, educational levels, social structure, chang- ing vacation habits, the degree of urbanization and geographical location, are the main determinants of the growth of vacation travel. The following table shows the proportion of the population in the more developed countries which travel on vacation, both at home and abroad. The table shows that the correlation between per capita Gross National Product (GNP) and vacation travel is not very close. In the U.S., with the highest per capita GNP, the proportion of the popu- lation traveling on vacation is considerably lower than in some Euro- pean countries, and countries with similar per capita GNPs within the same region vary considerably in their travel patterns. In some highly industrialized countries like Great Britain and the nations of Scandinavia the proportion of the population traveling on vacation has reached 60% and more of the adult population and this pro- portion is unlikely to rise much further. In many others, however, including such major tourism generating countries as the U.S., Ger- many and France, it is only around 40%. There appears to be a large potential for further growth in vacation travel. The extent of foreign travel depends in part on income levels, levels of education, etc., but it is also heavily dependent on the size of the country of origin and its geographic location. A European vaca- tioner is much more likely to cross an international frontier than a North American vacationer though the latter may travel a longer distance. Even in Europe, however, the bulk of vacation travel is domestic. Although statistical data are incomplete, the broad pattern of pres- ent and prospective vacation travel is fairly clear. The bulk of vacation travel is over relatively short distances within and between the de- veloped countries. Over somewhat longer distances there are sub- stantial numbers of vacationers from the U.S. and Canada visiting 376 Table 1: Vacation Travel in Major Tourist Generating Countries, 19671 Proportion Proportion Traveling PrprinProportion to Foreign Countries Itr of Population Traveling Inter- GNP per Traveling only within the inter- regional Population Capita on Vacation at Home region2 regional Travelers Countries million $ % % % % thousand Sweden 7.87 2,500 77.0 52.0 22.2 2.8 172 Switzerland 6.07 2,310 57.0 28.5 28.5 n.a. n.a. France 49.55 1,950 41.0 35.0 5.1 0.9 267 Denmark 4.84 1,950 66.0 40.0 22.6 3.4 123 Norway 3.78 1,860 78.0 67.0 10.2 0.8 23 Germany, Fed. 57.70 1,750 39.0 20.0 18.4 0.6 277 Rep. of Belgium 9.58 1,740 34.3 16.5 16.9 0.9 65 United Kingdom 54.98 1,700 60.0 50.0 9.3 0.7 295 Netherlands 12.60 1,520 54.0 38.0 14.7 1.3 92 Austria 7.32 1,210 40.0 26.0 14.0 n.a. n.a. Italy 52.35 1,120 19.5 18.7 0.7 0.1 32 Yugoslavia 19.84 530 12.0 10.0 2.0 n.a. n.a. Western Europe 1,346 United States 198.63 3,670 36.0 33.8 0.8 1.4 2,240 Japan 99.87 1,000 n.a. n.a. 0.3 0.3 241 'Since most data are results of sample surveys they can only be considered as estimates. Data on vacation travel of persons 18 years of age and older. 2Regions are the following: North America, Europe and East Asia. Source: "Problems of Tourism for Selected Countries in the Mediterranean Area with Special Refer- ence to Local Conditions" 1969, Institute for Tourism, University of Munich. "1967 Census of Transportation", U.S. Department of Commerce, with own adjustments. "Statistical Bulletin of International Tourism", 1966-69, Japan National Tourist Organization. Mexico and the Caribbean and larger numbers from Western Europe visiting the Mediterranean. The figures of inter-regional vacation travelers shown in Table 1 are approximations, but they provide a reasonable estimate of orders of magnitude. As a proportion of all vacation travel, inter-regional travel is small, and within the absolute total of such travel the share of the U.S. is large (almost 60%). The majority (52%) of U.S. travelers leaving North America visit Western Europe. Analysis of the European travel market indicates that for the next decade an increase of some 40 million in the number of vacationers (approximately a 45% increase over present levels) can reasonably be expected. A similar relative increase, if not higher, is likely for the U.S. and Japan. Only a part of this growth will result in foreign travel and only a fraction in long-distance travel; it can be assumed that those already traveling will tend to go farther afield in the future while newcomers to the travel market may seek their first experience in domestic tourism. 377 The costs of transport are a key element in determining tourist flows. In almost all tourist generating countries the private car ranks first as a means of transport and is likely to gain further ground. Rail travel, once the most important mode of transport in tourism, is steadily declining; its share in European travel is hardly more than 30% and in the U.S. much less. With declining tariffs (at least in real terms) and better and faster service, air transportation, while still used by only 6% to 10% of all vacationers, shows the fastest rela- tive growth. The growth of air charters especially has made possible rapid increases in traffic from North America to Europe, from north- ern Europe to the Mediterranean-e.g. Majorca and Tunisia-and on a more limited scale from Europe to more distant destinations such as Kenya, Thailand, Ceylon and the Caribbean. Long-distance travel by air, which is virtually the only mode of travel to the more distant developing countries from the major tour- ist generating centers, will continue to account for a relatively small share of total tourism travel in the foreseeable future. Nevertheless, considering the likely growth of income in the main tourist generat- ing countries and the rising proportion of the population earning higher incomes and enjoying higher levels of education, the absolute number of vacationers traveling long distances by air should increase very considerably in the next decade. Likely Future Tourist Flows to Developing Countries The factors which influence demand point to the likely develop- ment of tourism by areas of destination. The bulk of vacation travel will continue to be within and between the developed countries but there will be large increases in flows of visitors to the developing countries in the Mediterranean Basin and to Mexico and the Carib- bean. These countries will include: Mediterranean Basin: Spain, Yugoslavia, Greece, Morocco, Al- geria, Tunisia, Egypt, Israel, Lebanon, Syria, Turkey and Cyprus. The growth of tourism in some of these countries (Egypt, Syria, Israel and Lebanon) has been hampered by the political situation. The Caribbean: most of the Caribbean islands; Central American countries and Mexico; as well as Colombia and Venezuela in South America. From studies made of the European market, and from data avail- able on the United States, rough orders of magnitude can be given of the likely potential increase in vacation visitor flows to these two regions over the next decade. For the Mediterranean this figure would be about 10 million additional visits, nearly doubling the 1968 total 1 7 Q and for the Caribbean and Mexico an additional three to four million, compared with about three million in recent years. For long-distance travel, the areas of greater interest among de- veloping countries are likely to be Eastern Africa (Kenya, Tanzania, Uganda and Ethiopia) and South East Asia (Iran, Afghanistan, India, Ceylon, Nepal, Thailand, Indonesia and Singapore). in these coun- tries multi-destination circuits have developed, for example through Eastern Africa, South and East Asia and the Pacific. With appropriate air fares and increased facilities such circuits might also be developed further in Western Africa and South America. The magnitude of the likely visitor flows to each of the more dis- tant destinations is difficult to predict. They are competing destina- tions and much will depend on the success of each in providing tourist facilities of the right type and at the right price. Overall, the growth in long-distance vacation travel is likely to be fast. but in absolute terms will probably not exceed two to three million addi- tional vacation travelers over the next decade-approximately dou- bling the levels of recent years. Regional tourism can also be expected to expand. The greatest growth is likely from Japan to nearby destinations such as Korea and the Republic of China, but more distant destinations-Thailand, Singa- pore and Indonesia-will also benefit from the fast growth of the Japanese market. Other areas of regional tourism will continue to be in South America, between Argentina and Brazil and from these two countries to Paraguay, Uruguay and Chile; in southern Africa, from South Africa to Lesotho, Swaziland, Malawi, Malagasy and Mauritius; and in the South Pacific, from Australia and New Zealand to Fiji, New Caledonia and Tahiti. A number of these countries will also be the destination of visitors from distant countries but in relatively small numbers. Patterns of Tourism Expenditure As a rule of thumb, vacationers are likely to spend an amount equivalent to one month's income per family on annual leave, al- though this varies according to the living standards, not only from country to country but also within each country. Generally speaking, tourists will respond to price incentives. They have demonstrated strong price consciousness not only about accommodation and other services offered but also in choosing modes of transport and travel distances. Changes in tariffs have quite often resulted in considerable redirection of tourist flows. Moreover, changes in relative air fares have meant increasing price competition even between regional des- tinations such as Western Europe and the Caribbean where price levels had until recently been relatively independent. This empirical evidence that demand for vacation travel, both domestic and foreign, is strongly price-elastic has been confirmed by the limited amount of serious research carried out to date. The studies also suggest that expenditures on long-distance foreign travel are highly income- elastic. A tourist's daily expenditure may vary from a very low amount (i.e. when camping and visiting friends and relatives) to large sums for luxury vacations. Tour operators, especially in Europe, exercise a certain price leadership in their market which indicates at what prices supply can be offered. For example, European tour operators at the present time regard a daily expenditure of about $9.00 for full board as a ceiling price, which will be exceeded only in exceptional cases. The individual tourist very often orients himself by comparing prices of total holiday packages (including transportation) offered by tour operators and therefore will not easily be prepared to spend more. With long-distance travel, not only do travel expenses increase but daily expenditure also tends to go up, since most vacationers prepared to travel long distances are likely to demand relatively high standards of accommodation. This, of course, limits the number of potential tourists. The data on average daily expenditure by foreign tourists in Table 1 of the Annex reflect the differences in the types of market for different developing countries. Mediterranean destinations such as Spain, Yugoslavia, Greece and Tunisia appeal to a broad spectrum of middle- income tourists. In these markets the role of tour operators organizing package tours and budget holidays is very significant. This is also increasingly true of such distant destinations as Kenya, Ceylon and Thailand. On the other hand, travel on the part of relatively well- to-do vacationers or business travelers and those who combine business and vacation travel predominates in most other distant destinations. Vacation travel to most Caribbean islands (other than Puerto Rico and the U.S. Virgin Islands) has until recently been domi- nated by relatively high-income North Americans-most individual vacationers-at least during the main winter season. Excluding travel costs, normally from 60% to 75% of tourists' expenditures goes on food and accommodation. The balance is spent on excursions, entertainment and shopping. In some tourist destina- tions shopping accounts for a much larger part of expenditures- notably in Hong Kong and Singapore, both internationally famous duty-free ports-but in virtually all destinations tourists buy souvenirs and local crafts, thus providing an addition to exports which might not otherwise occur. Travel costs as part of the total cost of the holi- day vary greatly according to distance and mode of transport. For long-distance destinations by air, travel costs can amount to 50% to 60% of total costs. In the light of the patterns of tourism expenditure which have been described, it is clear that in a number of countries a really sig- nificant expansion in demand can be achieved only by increasing the supply of moderately priced accommodations, particularly if suitable arrangements can be reached with tour operators and air- lines to reduce the costs of air travel. Some countries, distant from the main centers of tourism generation, might be better advised to concentrate on the high-income part of the tourist market. "Seasonality" School holiday patterns, industrial vacation customs and climatic and geographic factors in both tourist generating and receiving coun- tries cause most vacation tourism to be highly seasonal. There is, however, a growing trend to two or more holidays in some of the most highly developed countries, particularly where winters are very severe such as the Scandinavian countries, Canada and the northern United States. This mitigates to some extent the seasonal peaks in some tourist destinations such as Majorca, Morocco, Mexico and some Caribbean and Pacific islands. Business travel has little relation to seasons and to a limited extent offsets variations in vacation travel. Other travelers, whose motivation might be for personal reasons, may travel at any time. The possibilities of offsetting fluctuations in business and holiday travel are quite limited. Tourism facilities for vacation travel are often not suitable for business travel. City hotels usually show their lowest utilization during the peak holiday season and holiday resorts expe- rience rather slack business during the rest of the year. Apart from the attempts to stagger school holidays, which have not been notably successful, there have been other attempts to lessen seasonality. The most promising seem to be in the creation of new demand, such as convention business and in price policies that induce tourism through preferential tariffs in the off-peak season. However, there are limits to price elasticity and also to the number of conventions to be ar- ranged. Seasonality can never be fully balanced out. This is most important in planning tourism facilities since a project must be de- signed for specific occupancy rates. The following yearly bed occupancy rates can reasonably be expected: 201 (a) Business Travel-60% to 90%; there is a tendency for declin- ing utilization of business accommodation with the shortening of the work week. (b) Vacation Travel-20% to 60% and higher, if the climate is very favorable in the tourist receiving country. Mediterranean resort hotels usually have occupancy levels between 30% and 45%; in a few exceptional cases this may go as high as 60%. In some of the Caribbean and Mexican resort hotels as well as some East African and South Sea resorts which cater mainly to vacation travelers, occu- pancy rates may go up as high as 80% and 90%. This, however, is only possible with a very select clientele whose holidays are not tied to any particular time of the year. Tourism Facilities Tourism facilities do not consist only of accommodation like hotels and boarding houses. They include also recreational and sports facili- ties of great variety and, of course, all the necessary infrastructure like transportation and utilities. In countries of cultural interest, historical monuments may also be regarded as part of tourism infrastructure, and the preservation, restoration, or betteraccessibilityof these monu- ments may be critical for the future of the sector in these countries. The composition of superstructure, mainly accommodation, has undergone considerable changes in the last 20 years. New types of accommodation, particularly holiday villages suitable for family-type tourists, condominiums and apartment houses, private villas and camping facilities, have proved successful and to some degree are replacing traditional hotels and boarding houses. These changes reflect changes in demand with new, often younger, groups entering the international travel market and also new approaches to the prob- lem of providing facilities in a very competitive industry with a highly seasonal demand. It is typical of the tourism industry that individual units are of rather moderate size. Hotels in general range from 150 to 500 beds, while holiday villages may offer as many as 1,000 beds. Considering an average investment per bed to be in the range of $5,000 to $10,000 depending on quality and location, the total investment required for each unit could be in the order of less than $1 million up to $5 million and more. There are, of course, exceptions whenever the implemen- tation of a whole complex is being considered; this type of develop- ment provides potential economies in construction and in marketing which are likely to be particularly significant in some less developed countries. In almost any developing country a need exists for some basic tour- ist facilities to accommodate business travelers. In these cases indi- vidual investments will be of relatively moderate size. By comparison, the supply of accommodation needed to meet demand for vacation tourism will call for relatively large investments in the regions likely to attract the largest tourist flows (i.e. the Mediterranean Basin and the Caribbean area) and for rather sizable investments even in those not close to the major tourist generating areas (e.g. Bali). Financial Aspects of Investment in Accommodation Although international experience indicates that over the longer term well-conceived investments in tourism accommodation facilities show satisfactory financial rates of return, the supply of accommoda- tion has often failed to increase in line with the growth of demand. In many countries starting the development of tourism, it has been dif- ficult to interest the private sector to invest in hotels and other forms of accommodation. Apart from lack of experience in the sector, two features of hotel investment in particular seem to account for the reluctance of investors: (a) the fact that relatively large amounts of capital have to be tied up in fixed assets over a long period (up to 20-25 years) with all the risks of changes in market conditions, the political and social framework, etc., involved; and (b) the strongly seasonal demand which is particularly pronounced in resort areas. This gives rise to concern that it will not prove feasible to achieve sufficiently high occupancy rates to make satisfactory profits. All the more so since, contrary to a widely held impression, only a small pro- portion of hotel operating costs-the equivalent of 15% to 25% of total sales-are variable in the short term. With fixed costs relatively high, there is little flexibility to adjust to low seasonal demand. Many governments in both developed and developing countries have attempted to overcome this reluctance of private investors to finance hotels and other forms of accommodation through a variety of incentive schemes. These include providing cheap or free land, equipment grants, exemption from import duties, liberal depreciation allowances and tax [olidays, loan guarantees, long-term government loans and interest rate subsidies. The different incentives are being provided in varying combinations. In some cases they are made to apply to investments in all new accommodation, in others they are used selectively in an attempt to direct investment to certain areas or for accommodation serving a specific market. In the Mediterranean Basin equipment grants and interest rate subsidies are the most com- mon subsidy schemes in effect at the present time, while in other parts 2.1Q 2 of the world, notably the Caribbean, tax holidays and tax exemptions are the most widely used incentives. The question of incentives is discussed further on pages397 - 398. Planning for the Tourism Sector In the majority of developing countries tourism has only recently attracted the attention of government as a sector warranting special economic policies and institutions. Too frequently, expertise in this sector is not presently available in government. Often and despite a sizable tourist flow, the private sector as well as the public lacks the know-how that would maximize benefits from tourism. Although training programs are increasingly available from different institutions and the United Nations Development Programme (UNDP), the Inter- national Labour Organisation (ILO) and various bilateral programs have already made a significant contribution, the need for training and technical assistance in tourism will remain a constraint on the growth of the sector in developing countries for some time to come. The need for an improvement in planning techniques for the sector is being increasingly recognized. Lack of expertise in market analysis, in construction techniques for hotels and other accommodation and tourist facilities, in appropriate financing for the sector, in tourism infrastructure requirements, in promotion techniques and in means to reduce the adverse effects of seasonality, can lead either to over- or under-investment in supply of accommodation, as well as to in- appropriate construction in relation to the market. Furthermore, the tourism sector can develop adequately only when given an appropri- ate government and semi-government institutional framework. In many countries such a framework does not exist and tourism is quite inadequately represented in high policy-making bodies where deci- sions are made on such topics as aviation policy and exchange rate policy which are critical for the industry. A major task of governments that have an important or potentially important tourism sector is to design and implement an integrated development plan for that sec- tor. Many governments require outside assistance in formulating such a plan. Foreign Exchange and Employment Effects of Tourism in Developing Countries The data in Table 1 of the Annex show that for a number of devel- oping countries tourism has become a leading foreign exchange earner. In many cases gross tourism receipts amount to more than 20% of the total value of merchandise exports and in some-Spain, Lebanon, Mexico and many Caribbean islands-tourism has become the most important export of goods and services. Between 1960 and 1968, while exports from developing countries (other than oil exports) rose by 7.6% a year, receipts from tourism increased at an annual rate of 11%. In view of the dubious world market prospects of many primary products and the uncertainty about the extent to which the industrialized countries will permit increased imports of manufac- tured goods from developing countries, this trend may continue and the dependence on tourism by many of these countries may well tend to increase in the years to come. In any event, it provides for many a useful element in diversifying their sources of foreign exchange earn- ings and, for some, one of the very few export opportunities available. The net, as a percentage of gross, foreign exchange earnings of tourism by developing countries vary considerably but tend to be high relative to many other exports. The countries fall roughly into three groups. Tourist facilities in such major tourist destinations as Mexico, Yugoslavia and Spain are constructed, equipped and supplied largely from local resources and staffed by local labor. Many of them are locally owned and operated. The net foreign exchange receipts are in excess of 85% of the gross foreign exchange earnings of these countries. At the other extreme are some of the islands in the Carib- bean and the Pacific and some of the relatively undeveloped countries of Africa. In many of them, operating supplies such as many foods and beverages come from abroad; specialized management, equip- ment and a major part of construction materials have to be imported; and ownership of tourist facilities is often foreign. Even for these coun- tries the net foreign exchange earnings are estimated to be seldom less than 45% of gross receipts. In most other developing countries which are less heavily dependent on imports, net foreign exchange receipts from tourism range from 60% to 80% of gross earnings. Available data indicate that even for many developing countries where tourism has become a leading foreign exchange earner, the sector's output constitutes a relatively small portion of the GNP and employs directly only a small part of the labor force. It is often claimed that tourism is relatively labor-intensive but the available evidence is not conclusive on this point. Studies in Mexico and Kenya suggest that the relative amount of employment generated for each unit of capital invested has tended to be higher in tourism than in most other private sector activities. However, similar studies in Yugoslavia and Israel, carried out by the United Nations Conference on Trade and Develoment (UNCTAD) Secretariat, were far less conclusive. IRQ Since tourism can often be developed in the less developed regions of a country, it may become a significant factor in redressing regional imbalances in employment and income. The extent to which this may be so will vary greatly from country to country. In some of the least developed countries the rapid growth of tourism may even lead to the creation of acute economic and social disparities between the areas where tourism has its most direct impact on money incomes and employment and the rest of the economy. Some Social Effects of Tourism Development The rapid development of international tourism in some of the developing countries has given rise to social problems which must be of concern to governments and outside bodies. One such problem is the attitude of the local population to tourists and their reaction to the tourists' requirements for accommodation and service, which by local standards are luxurious. Another problem may arise from for- eign ownership and management of tourist facilities and the feeling that indigenous people perform only menial tasks. Tourism may be regarded as a threat to the indigenous culture and mores, and there is a real possibility of a serious deterioration in standards of local arts and crafts as efforts are made to expand output to meet the tourists' demands. Not infrequently, resort development has resulted in local people being denied access to their own beaches. All these factors can give rise to serious problems in the reception of tourists and to demands for limitations on the flow of visitors. A further problem may be the demonstration effect of foreign visitor standards on the con- sumption expenditure patterns of the local population, and the dis- satisfaction which can result if new wants cannot be satisfied. While these possible negative effects of international tourism can- not be ignored, positive effects may be the increase in international contacts and in cultural exchanges which occur. The demonstration effect may also provide a stimulus to effort on the part of the popu- lation of the receiving country, while the growth of tourist traffic may open up needed economic opportunities. The balance between the negative and the positive effects of inter- national tourism varies in different countries. The effects are complex and little work has so far been done to try to assess the full impact of tourism development in particular destinations. By careful planning and regulation, the negative effects can be mitigated and it is impor- tant that the responsible public authorities should take this into ac- countinformulatingdevelopmentpoliciesandprogramsforthesector. WORLD BANK GROUP OPERATIONS IN TOURISM Financing Recognizing the important role of the private sector in tourism development, the main forms of World Bank Group assistance to tour- ism were, until mid-1971, directed to the investing and lending opera- tions of the International Finance Corporation, and the financing of hotels through development finance companies. A number of Bank/ IDA financed infrastructure projects, such as the Adriatic Highway in Yugoslavia, had also made a significant contribution to tourism devel- opment. As is evident from the data presented in Annex Table 2, IFC operations in the tourism sector, which started in 1967, have thus far consisted of nine operations for a total of $24 million. Of the invest- ments held by IFC in tourism projects at the end of 1971, $4 million was in the form of equity participation and $16.5 million in loans. These investments and loans have contributed to total investments appproximating $70 million. Two of the IFC operations deserve special mention. One was a recent loan in East Africa which is designed to promote regional tour- ism circuits in Kenya and Uganda. This loan is to be used both to finance hotels in the two countries and also ancillary facilities such as vehicles for the associated touring company. The second, by far IFC's largest operation in tourism to date, was assistance to a privately owned tourism promotion and financing company in Tunisia- Compagnie Financiere et Touristique, S.A. This company with total funds of $40 million linked important private Tunisian investors with foreign private interests involved in tourism. The company is engaged in the planning of new tourist areas and is financing besides hotel projects, ancillary facilities such as golf courses, marinas and camp- ing sites. Through its loans to intermediary institutions, the Bank has pro- vided $32 million for 79 tourism projects in five countries. Until August 1970 these loans were made through development finance companies whose main business was lending for industrial projects and two of them-the Banque Nationale de Developpement Eco- nomique in Morocco and the Societe Nationale d'lnvestissement in Tunisia-accounted for the overwhelming share of such financing. In August 1970, a loan of $10 million was approved for hotel financing through the Credit Immobilier et Hotelier of Morocco, which in the last few years has been mainly engaged in financing tourism projects. The first direct Bank loans for tourism projects were made only in mid-1971 with two loans for large integrated projects, both in Yugo- Q7 slavia. In December 1971 a third loan was made to support the provi- sion of infrastructure for a new resort to be developed on Mexico's Pacific coast, and in March 1972 an IDA credit for two moderate-sized hotels in Nepal was granted. Direct financing for tourism by the Bank and IDA has thus far amounted to $56.2 million. Negotiations have been completed for the financing of a project which would provide infrastructure to six principal tourist zones in Tunisia. Preinvestment Activities While the amount of Bank/IDA lending for tourism has been lim- ited so far, preinvestment studies have been made in a considerable number of member countries. Comprehensive surveys of the tourism sector have been carried out in 10 countries. Less comprehensive reconnaissance surveys have been made in another 21 countries. The Bank has been the Executing Agency for a UNDP financed pre- investment study for tourism development in Bali (Indonesia) and is currently Executing Agency for studies in Afghanistan, the Dominican Republic, Fiji and Venezuela. The studies in Bali, Fiji, Dominican Re- public and Venezuela are designed to provide the planning frame- work for the development of new integrated tourist beach resort areas. In Afghanistan the purpose of the study is to recommend the location, scale and type of tourist accommodation needed over the next five years. The Bank is providing technical assistance for tourism studies under way or planned for Colombia, Korea, Kenya, Yugo- slavia and Mexico and for regional tourism development in West- ern Africa. Lessons Drawn from Past Operations During the relatively short time that the Bank Group has been actively involved in tourism development, a number of important issues have emerged. Among these are the appropriate strategies of tourism development in particular countries with special reference to market possibilities and to the types and location of tourist facilities; the appropriate air transport policies; the environmental, ecological, cultural and social impact of tourism development in particular areas; and the ownership and use of land for tourism purposes. In its evaluation of tourism projects to date, the Bank has paid par- ticular attention to the assessment of market demand, still a difficult task where the necessary data are fragmentary and where rapid changes in the organization of the travel trade and in the development of air travel can have a major impact on traffic to particular destina- tions. In view of the many uncertainties, countries should avoid com- mitting themselves too firmly to catering for one type of traffic. Experience has shown that developments in air transport (e.g. air charter flights) have brought quite distant destinations within the reach of much broader market segments, and this can provide the basis for an advantageous diversification and broadening of the national tourism industry. Within the framework of market possibilities and the tourist attrac- tions of a particular country, tourist facilities should be developed with due regard to the costs both of the hotels themselves and of the necessary supporting services and infrastructure. The development of well-planned resort areas at priority sites to serve relatively large concentrations of visitors is likely to be much more economical than more scattered development which would require infrastructure to be provided over wide areas. Moreover, the former type of develop- ment is likely to be more in accord with market preferences as the tourist can enjoy a wider variety of activities and services. These con- siderations have weighed heavily in the Bank's projects in Mexico and Yugoslavia and also in the consideration of tourism development plans for Bali, the Dominican Republic and Fiji now being formulated. The issues in air transport policy are highly complex, and no full treatment of these issues is possible in this paper. Insofar as tourism is concerned, however, one issue is preeminent. Tourist traffic to a number of countries has been severely restricted because air charter operations have been limited in order to safeguard the position of national airlines. This is not necessarily in the country's best interests, and it is essential that the potential advantage to the national econ- omy from the development of tourism should be fully considered in devising rational air transport policies. The importance of the environmental and cultural impact of tour- ism development looms especially large in some countries, for exam- ple the island of Bali, where one of the principal tourist attractions is the rich cultural life of the population. Here it is critically important that tourist accommodation and other facilities should be located appropriately and tourist flows regulated so as to minimize the dis- ruptive impact on the daily life of the people. The preservation and improvement of the physical environment either in new or established resorts requires the provision of adequate infrastructure and the enforcement of appropriate land-use patterns, zoning and building regulations. In the evaluation of the Mexican and Yugoslav projects, particular importance was attached to these aspects. In the proposed Tunisian project, the preservation of the environment and the safeguarding of existing assets and new invest- ments by the provision of adequate infrastructure, particularly sewer- age, is a principal objective. In many areas of the world tourism development has produced great disparities in the standards of amenities provided for the visitor and for the local population. This cannot be entirely avoided, but in the longer run the improvement of standards for the local population is probably a condition of successful tourism development. The up- grading of amenities to be provided for the local population is an integral part of the Mexican project. The development of a new resort by the provision of infrastructure costing perhaps millions of dollars has a great impact on land values in the area affected. Were these benefits to accrue largely to a few private landowners, the Bank would have difficulty in financing such a project. The acquisition of land by government or a public agency prior to development, as in the case of the Mexican project, eases the problems of planning on an integrated basis, and of enforcing proper zoning and building regulations. It also helps to ensure that windfall gains do not accrue solely to private land owners and provides the opportunity for the public agency to offer serviced sites to hotel developers at prices which can include if necessary an incentive ele- ment. There is a very close relationship between land prices and building densities, and by appropriate land pricing the development authority can help to ensure that acceptable physical planning and density standards are maintained. While projects consisting of hotels alone are relatively straight- forward, though they of course involve a thorough examination of market, architectural and engineering, managerial, staffing and train- ing, financial, legal and economic aspects, tourism infrastructure or integrated resort projects have proved much more complex. Such projects normally includeseveral kindsof infrastructure-roads,water, sewerage, electric power, telecommunications, and sometimes air- ports-and they also involve questions of area and urban planning and architecture, complex ecological questions and problems of land tenure, as well as most aspects of hotel projects though in some respects in less detail. The preparation of the project in the borrowing country is, therefore, often very complicated and requires the co- operation of manygovernment agencies aswell as of private investors. Difficult problems of coordination arise which can sometimes be resolved only at the highest level. Due to the complexity of the projects, close involvement by the Bank in the project preparation stage is essential. Project feasibility studies submitted by potential borrowers thus far have sometimes sqo lacked a realistic analysis of tourism demand. Plans for area develop- ment and accommodation facilities have tended to over-stress physi- cal factors with insufficient regard for economic and financial viabil- ity. Some hotel projects have been poorly designed from a functional point of view. Consulting firms are still largely inexperienced in this field though gradually a body of expertise is being built up, and an improvement in the quality of consultants' work is already evident. The Economic Evaluation of Projects The Bank attempts to evaluate projects in the context of the tourism sector, assessing the range of the country's tourism assets and the value to the economy of the sector as a whole. In this last respect, tourism's particular value may be as an earner of foreign exchange or in some cases as a creator of employment. However, foreign exchange earnings or employment creation are not in themselves a sufficient reason for investing in tourism facilities. The economic benefits of such investments must be related to costs as in any other sector, and the Bank employs broadly the same criteria in evaluating a tourism project as in evaluating a project in, for example, agriculture, mining or manufacturing. A tourism project is considered appropriate for Bank financing when the economic rate of return is at least equal to the opportunity cost of capital in the country in which the project is located. Revenues (benefits) are derived from all expenditures by tourists resulting from the project, and costs are the value of all the inputs needed to produce these revenues. Expenditures made outside the project area (e.g. on travel costs to the country) are normally ex- cluded. In cases, however, where the project would have a significant and measurable impact on the utilization of capacity elsewhere in the economy (e.g. of the national airlines) these benefits should be taken into account. Conceptually, the costs and benefits properly attribu- table to the project may be clear, but there are often great practical difficulties in obtaining even an approximation of the relevant data. Whether the project consists of hotels or infrastructure or a combi- nation of these, its economic evaluation is made in terms of the investment in the project itself together with essential supporting investments in local transport, entertainment, shopping facilities, and other services for the tourist. Some of the facilities to be provided would serve and be paid for directly by the tourist, while others would be compensated for only indirectly. Accordingly, great care has to be exercised in estimating the costs and the benefits, so as to avoid any double counting. Cost and benefits may be evaluated in appropriate Rq1 cases by the use of shadow prices, particularly for foreign exchange, labor and land. Although overall a tourism project may be economically justified, appropriate shares of the benefits may not be captured by the differ- ent investing agencies. For example a new airport serving a tourism development may not be self-supporting for many years, and charges for some other basic utilities may be below economic levels. This might in some exceptional cases be justified in the broader project framework, but care must be taken to ensure that charges for public utilities are normally not subsidized for the benefit of the private investor. Reasonably precise estimates of expenditures by tourists in hotels can be made on the basis of comparable past experience. Similarly, operating costs of hotels can be estimated on the basis of past data for comparable hotels in similar locations. Data on expenditures by tourists outside the hotels are less easily obtained. Such data can only be established by periodic sample surveys, which in many countries are not available. Even if national estimates of different categories of tourist expenditure are available, the net benefit to the economy of these expenditures is still difficult to establish. Bank practice so far has been to assess as accurately as possible the supplementary invest- ments required and to apply the typical gross operating profit of each relevant economic activity to each category of tourist expenditure. Besides tourist expenditures on accommodation, food and enter- tainment, internal transport, shopping and other services, a project will also yield net benefits to the national economy from payments for visas, insurance, port fees, landing rights, etc. made by tourists di- rectly or indirectly in their country of origin to the country in which the project is situated. However, these benefits are generally excluded altogether from the calculation, since it is impossible to determine the share attributable to any one project rather than to another lo- cated elsewhere on the tourist's travel route. Payments of commis- sions of travel agents in the tourist originating country are of course also excluded. In the case of a new resort to be developed in more or less virgin territory, provision has to be made for the permanent population which will be attracted to the area. Basic services such as roads, water, sewerage, health and education, must be provided. To the extent that economic charges are made for such services, no charge need be made against the tourism project as such. To the extent that similar services (e.g. health and education) would have to be provided else- where for the same population in the absence of the project, only the additional costs of providing the services in the particular resort area are conceptually properly chargeable to the project. In practice the determination of such additional costs is likely to be very difficult. The provision of greater amenities for the permanent population will produce benefits which should also be taken into account. To the extent that these benefits are reflected in the prices paid for the serv- ices, some measure of value is available, but there is likely to be a residual of services for which no direct charge may be levied. Here also, the practical difficulties of including these benefits in the cost/ benefit calculation are very great. OTHER EXTERNAL FINANCIAL AND TECHNICAL ASSISTANCE Data on tourism financing from external sources other than the World Bank Group are rather fragmentary since most financing has been done privately, often in the form of suppliers' credits, com- mercial bank loans and investments from international hotel com- panies. Financing of tourism by regional development banks has been relatively limited thus far. The Inter-American Development Bank has reserved portions of industrial lines of credit to intermediate credit institutions in a number of countries for the financing of small and medium-sized hotel and motel projects and has made a loan to finance site development and infrastructure for a new beach area on the Caribbean coast of Mexico. It has also sponsored a number of tourism preinvestment and feasibility studies which it is financing by both loans and grants. The Asian Development Bank has financed some tourism studies but has not yet made loans directly in the sector although some of its loans for infrastructure, e.g., improvement of the airport at Kathmandu, Nepal, will be very helpful for tourism development. The African Development Bank has sponsored a useful survey of tourism in 15 countries of Western Africa. The Bank for Central American Economic Integration has begun to finance a num- ber of tourism projects in Central America on the basis of a special line of credit from the U.S. Agency for International Development. Finally, the European Investment Bank is becoming increasingly in- terested in financing tourism projects in states associated with the European Economic Community. To tourism projects in associated states it provides loans on particularly easy terms. Bilateral official assistance for tourism development has taken the form of technical assistance, sector surveys, preinvestment studies and training programs financed with grant funds, as well as loans and guarantees from official credit institutions. Belgium, France, Germany, Italy, Netherlands, the Scandinavian countries, United Kingdom, South Africa, United States, and recently Japan, have furnished an increasing amount of such assistance. The French and German official lending institutions have been particularly active in this field. Among agencies concerned with preserving open spaces for recreation the programs carried out by the United States and Canadian National Park Services deserve special mention. While there has been some technical assistance by the United Nations and the Organization of American States in the tourism sec- tor over the years, the financing of specific preinvestment and feasi- bility studies with UNDP funds is a relatively recent development. The physical development plan for the South Adriatic region of Yugo- slavia financed by UNDP was the single most important preinvest- ment study with strong emphasis on tourism carried out prior to the study of tourism development in Bali, Indonesia, for which the Bank was Executing Agency, and the other preinvestment studies, presently under way, to which reference has been made. The International LabourOrganisation is acting as ExecutingAgency in education and training programs in the tourism sector. Largely as a result of ILO's initiative, a number of comprehensive education and training programs have been started in countries such as Cyprus, Iran, Lebanon, Singapore and Tunisia, largely financed with UNDP funds. The U.N. Educational, Scientific and Cultural Organization (Unesco) has in recent years financed a number of studies of historic monu- ments and sites and their potential tourism aspects, including plans for their restoration and preservation. Unesco has also given advice and financial assistance to member governments in their efforts to restore and preserve historical monuments. The United Nations and UNCTAD Secretariats, as well as the Orga- nisation for Economic Cooperation and Development (OECD), have produced valuable studies of worldwide tourism trends and tourism developments in individual member countries. In the area of data collection, the key role is played by the International Union of Official Travel Organizations (IUOTO), whose legal status is being changed to that of an international inter-governmental organization under the name of World Tourism Organization. Composed of 109 countries and 96 national and international organizations, its operation is based on six Regional Commissions. The Bank has been working closely with this organization, including providing some financial assistance for a study sponsored by IUOTO. 5nA FUTURE BANK GROUP OPERATIONS Financing Program The Bank Group's projected financing program for the period 1972-76 contemplates total investing and lending of about $404 mil- lion for 44 tourism projects (including loans to development finance companies lending significant amounts for tourism projects).' This does not,include the tourism element of projects in the fields of air transport, highway construction, water and sewerage and vocational training. Sector Work This sizable investment program will require increasing the Bank's knowledge of the tourism sector in those countries and areas in which the bulk of operations is scheduled. Major sector reviews are sched- uled for Yugoslavia, Morocco, Mexico, Brazil, India, Tunisia and Turkey. Less comprehensive sector surveys will be made in many other countries, among them Kenya, Indonesia, Thailand, Malaysia, and countries in Western Africa and the Caribbean. The following table summarizes the Bank Group's actual financing of tourism projects and that projected through FY1976. It also shows the number of major sector studies already mounted or planned. Table 2: Summary of World Bank Group Tourism Activities by Fiscal Years, with Projections Through 1976 Actual Program Actual' Program 1969 1970 1971 1972 1973 1964-68 1969-73 1972-76 Sector Studies 3 2 3 1 3 1 12 7 Commitment ($ million)14 5 42 55 59 3 175 404 % of Total Bank Group 1.0 0.2 1.6 2.0 2.1 0.1 1.4 2.4 Number of Countries 4 2 6 6 8 1 19 22 Number of Financing Operations 4 2 7 6 8 1 27 44 Projects under Supervision (End FY) 5 7 12 18 25 1 14' 42' 'Including scheduled for 1972-73 'Average for five years 'Cf. Annex Table 5. Directions of Bank Group Financing While the Bank Group still has limited experience in financing tourism projects, conditions in different developing countries point to the need for flexibility in considering the types of projects it may finance. Until now, consideration has been given mainly to financing hotels or infrastructure, or a combination of the two, and the main emphasis has been on international tourism. Essential tourism infra- structure may include the restoration and better presentation of cul- tural attractions and such work may be included in a project provided the economic return on such investments is adequate. In many of the developing countries there are reservations about the desirability of developing tourism on the basis of private foreign ownership of hotels. If such foreign ownership is linked with inter- ests promoting and organizing the flow of visitors from the main market countries, it can contribute significantly to the success of a project. The Bank Group might be able to promote various forms of joint ventures whereby the developing countries would retain major- ity ownership of the facilities while the foreign investors would have a large enough commitment to ensure their interest in the success of the venture. In the richer of the developing countries, domestic tourism is ex- panding quite rapidly. In the Yugoslav and Mexican projects domes- tic visitors make up part of the projected traffic flows. Domestic tourism can contribute significantly to the success of a project by supplying a demand in the off-peak seasons and securing better aver- age occupancy rates. Provided that economic returns are satisfactory, there appears to be no prima facie case for the Bank to exclude from consideration projects catering to a significant extent for domestic tou rists. A further area in which the Bank Group may expand its operations is in financing related projects in two or more countries. In several tourist destination areas (e.g., Eastern Africa, Western Africa, East Asia) the attractions of one country alone may be insufficient to draw visitors from distant markets, and tourist circuits through several countries have been developed. The IFC is already participating in a project which includes hotels and lodges in both Kenya and Uganda. The development of tourist circuits may significantly expand the tourist flow to Western Africa and a regional project has already been submitted for consideration by the Bank Group. Increasing attention will be given to such proposals in the future. The large increase in tourist flows to the developing countries has for the most part occurred in the past 10 to 15 years. The industry is still young. Traditions of visitor reception and hotel management are only now developing. Skilled personnel at all levels are in short sup- ply. With the large prospective increases in visitor flows, the problem of providing skilled manpower in the numbers required will be seri- ous. In these circumstances the Bank Group will need to pay greater attention to training and, where appropriate, include funds for train- ing in its lending for tourism projects. A start in this direction was made in the case of the Yugoslav and Mexican projects. In its lending operations in tourism the Bank Group aims, as in other sectors, to combine financing with technical assistance. Besides advice on the technical aspects of the project itself, emphasis will be given to broader questions of planning, including the organiza- tion of the industry, the structure of planning in the public sector, regional and area planning, and planning the preservation of the environment. For effective action in developing tourism, which im- pinges on many economic sectors and areas of responsibility, a strong organization either in government or elsewhere in the public sector is essential, and the effective coordination of all public sector agen- cies responsible for the implementation of a tourism development program usually requires attention at the highest level of government. Tourism cannot develop under a policy of benign neglect. In re- gional and area planning, also, the active support of top government is needed, for the implementation of soundly conceived regional plans with adequate zoning and building regulations may meet strong opposition from powerful private interests. Here the support of the Bank Group may be of considerable value in bringing the type of development which does not destroy the physical environment but preserves irreplaceable assets for future generations. Investment Incentives The successful development of tourism requires the cooperation of government and private investors. Conditions must be created to attract the private investor. Thevarious kinds of investment incentives which many governments offer have been noted on page 383. A case can be made for incentives to hotel investors in certain circumstances. The hotel is only one, though a key part, of the total "package" bought by the tourist. Because a substantial part of tourist expendi- tures is made outside the hotel, the hotel investor may be unable to capture all of the benefits flowing from his investment. Hence the prospective financial returns on the hotel may be unattractive. In these circumstances, assistance to the hotel investor may be justified, but the case for incentives of wide applicability is not proven. Each system of investment incentives has its specific advantages and disadvantages. For example, the size of the once-and-for-all de- mand on public sector resources made by the capital grant incentive system is one of the major disadvantages of this type of incentive for governments. The tax rebate system has been an important in- ducement to invest, but has been difficult to suspend. Tax holidays are normally offered for part of the life of the project only, so that the project's cash position can change unfavorably once they are terminated. Moreover, it is very difficult to estimate the costs to governments. Subsidization of interest rates has appealed to governments as it can be easily administered. A major advantage is that the cost of an interest rate subsidy is calculable compared with a tax holiday. A dis- advantage is the stimulus provided to investors to increase the pro- portion of debt in the project's capitalization beyond a prudent level. The major disadvantage of all types of incentives is that they may lead to over-investment and a waste of scarce capital. If assistance to the tourism sector through investment incentives is provided, experience to date points to the desirability of limiting such assistance in time and providing it selectively. The danger exists that in some countries such selectivity may be poorly and corruptly administered. The need for special assistance may, however, be avoided or greatly reduced if capital for tourism projects, particularly hotels, can be provided on extended terms related to the economic life of the assets and with adequate grace periods. The absence of such terms has frequently caused cash squeezes which have resulted in poor service and maintenance and inadequate replacement ex- penditures. Because loan funds on suitable terms are frequently not available in developing countries, the contribution which financing from the Bank Group can make may be of particular value. In evaluating tourism projects, the Bank Group will continue to pay close attention to the overall impact and justification for investment incentive systems. There are formidable conceptual and practical problems in assessing such systems and research here will be needed to provide a basis for discussion with member countries where in- centive systems are in effect. It must be recognized, however, that even if a case against such incentives in a member country is estab- lished, any change in the system may be very difficult to bring about or may not be justified if investment incentives are offered in com- peting tourist destinations. 3Q9 Research Reference has been made to the need for research into the scope and impact of investment incentives. It is proposed that attention should be focused first on the Mediterranean area and the study would begin with a comprehensive review of the legislation and practical operation of the incentive schemes operating in Morocco, Tunisia, Spain, France, Italy, Yugoslavia and Greece. This would pro- vide the basis for formulating guidelines for Bank Group operations in countries where important incentives for tourism investments are provided. To provide fuller information on hotel operations in different countries, the Bank will continue to finance the work of a German research institute which collects comparative hotel operating statis- tics for about 479 hotels in 28 countries. This research will be ex- panded, the objective being to include additional hotels in the developing countries. A third area of study will be the Caribbean. The major markets for all Caribbean destinations are the same, and the development of capacity in one destination will affect others. A study is proposed of the tourism development programs of all Caribbean destinations, including Venezuela, Colombia and Central America, in relation to the prospective market. It is hoped that this study can be carried out in cooperation with the Caribbean Development Bank. More basic research into income and price elasticities of demand for vacation travel, into the likely trends of air transport costs, and into patterns of tourist expenditures in developing countries will be carefully considered. It is hoped that some of this work could be carried out in cooperation with selected developing countries which have already become important tourist destinations. Annex Table 1 Comparative Data on International Tourism in Selected Developing Countries, 1970 Growth of Average Gross Average Gross Tour- Estimated Foreign Length Foreign Daily ism Receipts Foreign Visitor of Stay of Tourism Foreign Compared to Visitor Arrivals Foreign Receipts Tourist Merchandise Arrivals Index 1969 Visitors ($ Expend. Exports (000 (1965=100; (Nights) million) ($) (%) East Africa Ethiopia 53 - 4.0 6 28.0 4.9 Kenya 344 - 8.8 52 16.8 24.0 Tanzania 63 9.0 14 24.8 5.6 Uganda 80 - 9.7 19 24.0 7.7 West Africa Ivory Coast 42 - 4.0 5 30.0 1.5 Senegal 40 - 3.5 4 25.0 2.6 East Asia & Pacific China, Rep. of 472 352 4.9 89 38.8 6.2 Hong Kong 927 228 3.8 293 83.2 11.7 Philippines 144 171 7.2 32 30.5 3.0 Thailand 629 280 4.8 104 30.0 14.9 Indonesia 129 - 5.0 16 25.0 2.0 Fiji 110 275 7.0 28 29.4 27.5 Korea 173 524 4.0 19 27.0 5.6 Europe, Middle East and North Africa Cyprus 127 385 5.2 20 30.4 18.3 Greece 1,408 166 10.9 194 13.0 30.2 Israel 437 168 8.3 104 28.7 13.4 Lebanon 822 137 7.4 132 14.1 76.7 Malta 171 356 14.2 29 12.0 74.4 Morocco 747 497 9.0 136 15.0 27.9 Portugal 3,343 221 3.9 222 16.8 23.5 Spain 24,105 197 7.0 1,681 15.7 71.7 Tunisia 411 248 9.3 55 14.6 30.2 Turkey 484 154 7.7 51 13.7 8.8 Yugoslavia 4,748 179 4.8 275 12.2 16.4 South Asia Afghanistan 100 909 9.1 6 7.8 7.1 Ceylon 46 230 10.6 4 8.2 1.2 India 281 190 22.7 51 8.0 2.5 I ran 299 220 7.0 42 20.1 1.8 Nepal 46 511 4.0 2 13.0 9.1 Pakistan 122 165 8.9 10 9.3 1.4 Central America and Caribbean Bahamas 916 212 5.3 221 45.5 248.3 Jamaica 309 163 8.1 95 40.0 28.2 Mexico Tourists 2,246 187 13.6 575 21.0 41.0 Border Travel - - - 879 - *Colombia 143 180 7.0 30 30.0 4.9 *Venezuela 108 200 13.0 46 30.0 1.8 AtflfO Annex Table 1 continued Growth of Average Gross Average Gross Tour- Estimated Foreign Length Foreign Daily ism Receipts Foreign Visitor of Stay of Tourism Foreign Compared to Visitor Arrivals Foreign Receipts Tourist Merchandise Arrivals Index 1969 Visitors ($ Expend. Exports (0OO (1965=1001 (Nights) million) (S( South America Argentina 695 233 6.8 77 16.2 4.3 *Brazil 173 168 n.a. 28 n.a. 1.2 Paraguay 119 476 4.0 14 29.0 21.9 *Uruguay 620 103 7.5 45 10.0 18.0 *1969 data. NOTE: The above statistics are mainly from official sources, supplemented by semi-official estimates. In view of the many different definitions and methods of collection used, the above figures should be taken as indicative only, reference being made where possible to the most recent official sources available. Sources: IUOTO, International Travel Statistics OECD, International Tourism and Tourism Policy in OECD Member Countries IMF, Balance of Payments Yearbook IMF, International Financial Statistics World Bank, Various Reports A £1- Tourism Financing by the International Finance Corporation Investment held by IFC Original Amount FY of as of December 31, 1971 of I FC Commitment Original (S thousand) Country Project or Approval Commitment Comments (S thousand) or Approval Equity Loan Total Kenya Hotel Properties 3,204 1967, 1968 561 1,550 2,111 Part-financing of 200-room hotel in capital city, some game lodges and 100-room beach hotel. Jamaica Pegasus, Hotels of 2,913 1969 679 1,280 1,959 Convention hotel in capital city. Jamaica El Salvador Hotel Miramonte 933 1969 333 600 933 224-room first class hotel in capital city. Tunisia Cie. Financiere et 9,905 1969 1,905 6,891 8,796 Tourism development and hold- Touristique ing company. Colombia Hoturismo 6 1969 6 - 6 Participation in hotel develop- ment company. Colombia Pro-Hoteles, S.A. 1,045 1970 238 800 1,038 225-room business and tourism hotel in provincial capital. Mauritius Dinarobin Inns & 600 1971 - 600 600 Financing of two beach hotels Motels Ltd. comprising 360 beds. Panama Corp. de Desarrollo 1,473 1971 267 1,206 1,473 256-room hotel in capital city. Hotelero, S.A. Kenya & Tourism Promotion 2,420 1971 3,600 3,600 Financing of 6 hotels and lodges Uganda Services Ltd. 1,180 comprising 950 beds and 138 vehicle touring service. Total 23,679 3,989 16,527 20,516 e *i -I Annex Table 3 Commitments for Tourism Projects Under World Bank Loans to Development Finance Companies Bank Funds Committed by DFCs No. of as of Dec. 31, 1971 Country Name of Company Projects ($ thousand) Malaysia Malayan Industrial Development Finance Limited 1 325 Tunisia Societ6 Nationale d'lnvestissement 32 14,510 Morocco Credit Immobilier et Hotelier 22 8,131 Morocco Banque Nationale de D6veloppement Economique 22 7,871 Thailand Industrial Finance Corporation of Thailand 1 260 India Industrial Credit and Investment Corporation of India 1 972 79 32,069 Annex Table 4 Direct Bank Loans and IDA Credits for Tourism Projects, FY1971-72 Date of Country Type of Project Agreement $ Million Yugoslavia (Slovenia) Integrated Resort Complex (Bernardin) 6/18/71 10.0 Yugoslavia (Croatia) Integrated Resort Complex (Babin Kuk) 7/21/71 20.0 Mexico Tourism Infrastructure 1/18/72 22.0 Nepal Hotels 3/22/72 4.2 Total 56.2 Annex Table 5 Projected Bank Group New Commitments for Tourism (') FY 1972-76 Direct Financing $ Million % Bank/IDA 294 73 IFC 37 9 331 82 Development Finance Company Financing 73 18 404 100 M) Excluding loans to industrial development banks which do only minor tourism financing. AC52 URBANIZATION Introduction and Summary ......... ........................... 407 Dimensions of the Urbanization Program ...... .................. 412 The Growth of Urban Centers ....... ...................... 412 Urban Employment, Income Distribution and Living Conditions . . 418 Resources for Urban Development ...... ................... 422 The National/Regional Setting ........ ......................... 425 The Problems of Urban Centers ........ ........................ 433 Urban Transport and Urban Patterns ...... .................. 435 Land Use Guidance .......... ............................ 442 Urban Housing ............. ............................. 445 Levels of Urban Services ........ .......................... 450 Institutional, Planning and Pricing Problems ..... ............ 452 The Bank's Role . ............................................. 457 The Program of Urban Activities ...... ..................... 462 Formulation of National Urbanization Policies ..... ........... 464 Technical Assistance in Urban Programming and Institution Building .......... ............................ 465 Urban Transport ............ ............................. 465 Housing and Urban Works Projects ...... ................... 468 Urban Region Projects ......... ........................... 470 The Design of Urban Projects ....... ...................... 471 Research .......... ..................................... 472 Annexes 1. Background Data on Urbanization ...... .................. 475 2. Urban Land Policy ........... .......................... 488 3. International Assistance in Urban Development ..... ........ 503 4. Bank Lending for Urban Purposes ...... .................. 513 URBANIZATION SECTOR WORKING PAPER INTRODUCTION AND SUMMARY Problems of urban poverty and unemployment, of inadequacy of housing and urban infrastructure, have been recorded throughout history. What most distinguishes the current urban problems of the developing countries is their scale and intensity. The severity of the problems reflects primarily the rapidity of overall population growth and the acute shortage of resources with which to equip the additions to urban population. The proliferation of squatter settlements and slums, and the rising backlog in urban services, have been accompanied by growing recog- nition that "development" implies much more than just expansion of output. Yet with few exceptions, measures so far undertaken have signally failed to reverse these trends or produce more efficient patterns of urban growth. In recognition of the importance of these issues, the World Bank' recently decided to supplement its activities in individual sectors with a more direct focus on problems of urbanization. As this is a new field for the Bank, the discussion and policy proposals contained in this paper are more tentative than for fields in which the Bank has many years' experience. The first section outlines the broad dimensions of the problem. The combination of a 5% urban population growth with historically rapid expansion in rural population appears unprecedented. About half of the additions to urban population are migrants from rural areas. How- ever, the migrants represent only a fraction of the natural increase in rural population. Existing trends provide no expectation of an early alleviation in total or urban population growth rates. Even if the totality of the net savings of the poorer countries were devoted to this purpose, only $5002 to $800 per head would be avail- able for equipping the additions to their population. Much the same 'All references to the Bank include the International Development Association (IDA) unless otherwise specified. References to the World Bank Group include the International Finance Corporation (IFC) as well as the Bank and IDA. 2Amounts are expressed in U.S. dollar equivalents. 4n7 level applies to the urban centers where higher savings are offset by more rapid population growth. In practice, a large slice of the savings will inevitably go to other purposes. Municipal revenues in relation to population growth are substantially lower and are largely pre- empted by current expenditures. By contrast, capital costs per head for minimum conventional hous- ing and for education, water and sewerage, typically total over $500. Capital costs for providing employment vary from perhaps $400 per worker for "traditional" to upwards of $1,500 in "modern" occupa- tions. Many other investments, in transport and electricity for in- stance, are also required to provide infrastructures for the urban population growth. In the poorer countries at least, the resources available are quite evidently insufficient to equip the additions to urban population at customary standards. The next section considers the possibilities of relieving urban pres- sures by measures to restrain migration from the countryside. The arguments for greater emphasis on rural development as against urban development, and on smaller towns as against major cities, are inconclusive. Cases undoubtedly exist where urban biases should be corrected. Of much greater potential importance, however, are poli- cies to promote better use of national resources, particularly labor, by reducing factor price and other distortions. Such policies will af- fect the composition of investment in both town and country and may lead to either greater or less priority for urban development according to the differing conditions existing in individual countries. Greater emphasis on development in rural areas or the smaller towns is, in any case, unlikely to cause significant reduction in the rapidity of growth in urban population or the problems of the larger cities over the next two decades. Rural development in the circum- stances indicated in the text may even accentuate migration to the towns, and from the smaller towns to the larger. The third section accordingly turns to consideration of the prob- lems within urban areas and the associated opportunities to economize in use of resources and increase urban efficiency. Such opportunities arise from the extensive interrelations between urban sectors inadequately taken into account in decision making, from customary standards of urban services not appropriate to prevailing conditions and, closely connected, from inappropriate pricing and taxation policies. Nowhere are these factors more evident than in the trilogy of urban transport, land use and housing which together play a leading role in shaping urban growth. The population and area of many of the cities will triple within 20 years, necessitating heavy investment in roads, other transport in- AnUQ frastructure and equipment for the mass transit of the majority of the population who cannot afford automobiles. Private automobile use is rising rapidly. The resultant congestion greatly complicates provi- sion of mass transit. A strong case exists for bringing automobile user charges more into line with the high marginal costs of providing for additional vehicles in congested urban areas, and for other steps to limit their growth in numbers. No less important are the possibilities provided by future exten- sions to the urban areas which in much less than 20 years will be larger than the existing areas. These urban extensions offer opportu- nities for promoting more economical urban patterns than would be achieved by continued laissez-faire growth. While transport facilities play a large part in determining urban growth patterns, transport requirements can be greatly reduced by appropriate siting of employ- ment and residence. Strong interrelations also exist between different transport modes, necessitating their consideration in the context of the urban transport system as a whole. The shortcomings of existing land use regulations, and methods of public acquisition of land present a serious obstacle to more rational urban development. Shortage of resources for extending urban serv- ice networks restricts the availability of "urbanized land" and adds to the already strong forces of population growth and increased income in pushing up land values. The heavy costs of acquiring land in turn impinge on municipal financial resources. Public acquisition of land and other measures can greatly facilitate a more integrated urban development. Reform of urban land taxation can play a crucial role in strengthening municipal finances. Housing needs are being met to a large degree, though inade- quately, by the rapid growth of squatter settlements which in many cities already accommodate over a third of the population. At present income levels, it is impossible for most urban inhabitants to afford even minimum standards of conventional permanent housing. Given the limited resources available, there is no prospect of adequate pro- vision of subsidized housing for the new additions to population, at least in the poorer developing countries. In these circumstances, more emphasis on "site and services" type schemes providing urban- ized land for self-help housing is indicated. Indeed, for most devel- oping countries, the harnessing of the self-help and savings potential provides the only realistic possibility for substantial alleviation of housing conditions. "Trade-offs" between levels of urban services and corresponding costs have not yet been adequately analyzed in relation to resource availability. Possibilities of significant economies may well exist. A AOO 10% reduction in capital costs of electric power generation and dis- tribution, if attainable, would provide a saving of over $1 billion a year by the 1980s. A choice may often exist between services at conventional standards for a limited proportion of the population or for a larger proportion at somewhat lower levels but lower unit costs. Early attainment of comprehensive, and at the same time effectively policy-oriented, urban plans and models, is precluded by the unsatis- factory state of methodology and the rapidity of change, not only in population but also in social patterns and economic functions. In these conditions, pricing policies and taxation incidence acquire a critical importance in harnessing market forces to produce more rational patterns of urban growth. Better charging of marginal costs of peak loads and of social costs of private investment could result in substantial economies. The multiplicity of local governments and autonomous agencies greatly complicates the problem. In assessing advantages of greater centralization and the closer relation of plan- ning with budgeting and political decision making, the importance of promoting local initiative and self-help should not be neglected. It would be unrealistic to expect the degree of coordination, and of planning and executing capacity, needed to take into account fully all the complex urban interlinkages. Rather a sub-optimization ap- proach seems indicated based on rough alternative patterns of urban growth, and initially taking into account only major interlinkages. The framework for evaluation of projects on such an approach will vary according to the type of project. Calculation of economic costs/ benefits should be supported by an assessment of other aspects such as impact on income distribution and employment. A similar step-by- step approach is indicated towards institution building. It is difficult to perceive a better practical alternative to the present ad hoc deci- sion making. The final section considers the role of the World Bank. Although the Bank has only recently undertaken explicit consideration of prob- lems of urbanization, it is already involved as a result of the high proportion of its lending that is related to urban areas. In extending its activities in the urban field, several types of problems have become evident, ranging from institutional difficulties where several agencies, often relatively weak, are involved, to development of new method- ologies and pricing policies. Such difficulties make preparation of projects specifically directed to the underlying problems of urbani- zation particularly time-consuming and arduous. Experience in this field both in the Bank and elsewhere is scarce. In view of the large number of developing countries which are members of the Bank and the far greater number of their urban cen- Al 'I ters, a policy of selective concentration both as regards countries and types of urban project is indicated. Advantages will stem from both continuity and complementarity between projects. Some new types of urban projects addressed to particular urban problems, such as provision of site and services to low-income groups, should be under- taken in a wider range of urban centers. A program is outlined of urban projects for FY1972-761 and of other activities to complement the more traditional work of the Bank in such fields as public utilities. The program encompassing some 40 projects and a roughly estimated $700 million of lending, is neces- sarily more tentative than in fields where the Bank has longer experi- ence and extended connections with borrowing agencies. To increase the impact of the program, collaboration will be sought with other agencies with experience in the field. The Bank's system of economic missions, economic reports and country programs will be progressively extended to help national au- thorities provide a focus for consideration of urbanization problems. Technical assistance on programming and institution building is fore- seen in connection with the urban projects. Improvement of urban management is particularly important. A number of projects for mass transit are included in the program with particular emphasis on bus systems. Interlinkages both between modes of transport and with the pattern of urban growth will be an important aspect of evaluation, including policies towards restraint of private automobile use. Similar considerations apply to the urban road projects. Wholesale market projects, which can make a large contribution to urban efficiency, are also included in the project program. It is recognized that the Bank's contribution in terms of resources to solution of the housing problems of urban centers can only be small. The several projects in this field included in the program are designed accordingly for maximum demonstration impact and insti- tution building. For several reasons, direct lending for house con- struction appears of lower priority than for site and services type projects which can mobilize self-help, reach the poorer levels of population and stimulate savings and employment. The project pro- gram is therefore concentrated on this type of scheme. Other projects in this field include improvement of low-income settlements and in- tegrated urban extension areas that can both produce economies within the areas selected and promote more economical patterns of urban growth. 'The Bank Group's fiscal year (FY) ends June 30. A1 d While direct lending for house construction is not contemplated at this time, the Bank Group is prepared to consider lending for seed capital to develop housing finance institutions. The emphasis here is on the leverage effect in promoting savings and developing capital markets in a form which will lead to amelioration of the overall hous- ing and employment situation. Particular attention will be paid to the income groups for which housing is proposed, the mobilization of small savings and the possibilities for supporting improvements to substandard housing. More generally, throughout the field of projects in urban areas, the consideration already being given to ensuring project designs appro- priate to the local resource availabilities will be further extended. In this connection, the research program, which will be undertaken in collaboration with other agencies, will contain some specific studies of relations between costs and levels of services. Municipal finance, and urban transport interdependencies, will also be prominent in the research program. It should be recognized that this is a preliminary stage of the Bank's efforts to relate its work more closely to the problems of urbaniza- tion, unemployment, poverty and congestion. The changes may ap- pear cautious in comparison to the mounting urban crisis. But from the review of activities in the field of urbanization has come an acute awareness of the paucity of knowledge and methodology on how best to cope with the problems. A policy of learning while doing and readiness to adapt policies with experience is therefore strongly indicated. DIMENSIONS OF THE URBANIZATION PROBLEM The Growth of Urban Centers To assess the implications of the rapid growth in urban centers in developing countries, it is necessary to distinguish between the rise in the proportion of urban to total population and the growth in ab- solute numbers of urban population. The rise in the urban proportion does not appear exceptional. The urban proportion in the industrial- ized countries rose at much the same rate during the initial periods of their technological revolution and when their urban proportion was similar.' As the following table and the chart in Annex 1 indicate, a 'Kingsley Davis estimates the rate of growth in the urban proportion for 171 develop- ing countries between 1950 and 1960 at 2.29% per annum. For 18 developed coun- tries during their decades of fastest urban growth, the average was 3.09% per annum, but the rate was somewhat lower during initial periods of accelerated growth from the new technologies. strong correlation exists between degree of urbanization and level of income per capita. Table 1: Urban Population in Developing Member Nations of the World Bank by GNP Per Capita (1970) Under $100 $200 $350 $575 to Over GNP Per Capita $100 to $199 to $349 to $574 $1,000 $1,000 Percent of Total Population in Urban Areas 9.5 17.7 39.5 45.3 53.6 69.0 Percent of Popu- lation in Commu- nities of More Than 100,000 4.6 10.1 19.2 22.2 27.2 31.7 What is exceptional is the rate of increase in absolute numbers of urban population and, even more, the accompanying growth in rural population. The underlying cause is obvious-much higher natural rates of total population growth than ever reached by today's devel- oped countries. The urban population is currently growing about 5%5 a year, or doubling in about 15 years, a rate only occasionally reached by the developed countries for short periods at a much later stage of economic development. For such growth in urban population to occur accompanied by a rapid increase in rural population appears unprecedented. Rural pop- ulation grew by over one-third between 1950 and 1970. Compared with the first half of the century, the growth rate accelerated faster than did that of the urban population. At comparable periods in the developed countries, rural population, with' few exceptions, showed very slow growth or absolute decline. As much as half of the urban population expansion is accounted for simply by natural increase-rather more than half in South and South- east Asia and probably somewhat less in Africa. Not only are birth rates considerably higher than in, say Europe in the early nineteenth century, but life expectancy is now much longer.2 '1950-60 estimate for urban agglomerations of over 20,000 inhabitants, "Growth of the World's Urban and Rural Population, 1920-2000," United Nations, 1969, Table 11. Unless otherwise indicated, "urban" data in this paper follow the UN definition of "over 20,000 inhabitants." 'Despite the squalid conditions of the city slums and generally far greater poverty, medical progress has ensured a much lower death rate particularly during infancy than in the cities of Europe at comparable periods. Even for countries with income per capita below $200 per annum, expectation of life at birth in cities of over 100,000 habitants is now on average appreciably higher than in London as late as the mid- nineteenth century. No early alleviation of this component of urban expansion can be foreseen. As incomes rise, large families are more likely to become an economic liability to their parents than in rural areas. Together with the expansion of birth control programs, this influence can be ex- pected to reduce urban fertility. But such trends are slow to develop. In the short term, their effect may well be offset by other influences, including further declines in mortality rates. Migration from rural areas accounts for the other half of urban population expansion.1 The many studies on the causes of this migra- tion have shown so many "push-pull" aspects and such variety in its composition, not only between countries but also between regions of the same country and over time, as to make generalizations hazard- ous. Differences in social norms, including the degree of "emancipa- tion" which the towns offer, in existing links with urban families, in prevailing conditions in the towns and the countryside, particularly the dynamism of urban economic growth and the availability of addi- tional agricultural land, all influence the flow. That greater opportunities are seen in the towns primarily in terms of income seems clear enough. There is considerable evidence that the migrants, or at least those who stay, are generally successful in this respect, even though there may be a period of waiting during which support is received from friends and relatives. Unemployment among migrants of a few years' residence is generally noticeably lower than urban averages. The attraction of amenities as such seems of less immediate signi- ficance. Use of amenities generally requires payments and is therefore conditional on earning a sufficient income. Conversely, attempts to restrain migration by deliberate policies of restricting the growth of urban services appear to have been unsuccessful. Migrants generally seem to regard living conditions even in squatter settlements as no worse than those they have left. And certainly the life-style is gener- ally considered more attractive. Surprising to the town dweller the neon lights and billboards may also appear more beautiful to the migrants than the countryside. As a very tentative generalization, the migratory movement has slowed significantly only when the absorptive capacity of the towns in terms of economic growth remains stagnant and urban unemploy- ment is persistently high. There are, however, isolated cases such as Taiwan where prolonged and successful rural development of a 'This broad discussion abstracts from consideration of urban population growth re- corded as a result of villages growing to exceed the lower urban limit of 20,000 habitants. This statistical factor accounts for around 10% of the apparent urban population growth. 4lA labor-intensive nature has been accompanied by significantly lower than average migration. The marginal nature of rural to town migration deserves emphasis. Important as the migration is in relation to existing urban population, it is nevertheless equivalent to only a fraction even of the increase in rural population. If the population of the rural areas is six times the population of the urban areas (as in the most populous of the devel- oping countries), and the rate of natural increase the same as in the towns, then migration equivalent to only one-sixth of the additions to rural population will double the urban population growth rate.' These considerations throw some light on the variations in urban growth rates between countries and regions of the developing world. Urban population growth is most rapid where total population growth is particularly high as in South America, or where, as in Africa, a somewhat more modest population growth is accompanied by a particularly high proportion of rural to total population. The unex- pectedly slow rate of urbanization in India in the 1960s appears to be associated with very slow economic growth and very heavy unem- ployment in some of the major towns, such as Calcutta. Additional information on urban growth in individual countries is given in Table 3 of Annex 1. Projections of future urbanization levels depend on the assump- tions made as to changes in population and urbanization trends as the process of development continues, and as to the speed of the rise in income levels. The estimates given below are from United Nations sources. Comparisons with other estimates do not indicate important divergencies over the next two decades due to the strong influence of the existing age structure and the limited potential for changes in birth rates over this period. Even for the year 2000, the differences can be considered marginal in relation to the size of the problem- similar levels of urban population would be reached a few years ear- lier or later. Thereafter divergencies become much more marked. The projected increase in the urban population of the developing countries in the 40 years from 1960 to 2000 is over one billion, more than four times the increase in the previous 40 years and about three times the total urban population of the developed world in 1960. If this appears scarcely plausible, the projected increase in rural popu- lation appears equally so. The one and a half billion increase foreseen is three times that of the previous 40 years and almost as large as the 'In countries such as most of those in Africa with still lower proportionate urban population, even a relatively small migration of rural population can thus have a dramatic effect on the rate of growth of urban population. In the more highly urban- ized Latin American countries, migration of a similar proportion of the rural popu- lation will have much less impact. 415 Table 2: Urban/Rural Population of Developing Countries 1920 1940 1960 1980 2000 Urban (above 20,000) Popu- lation (Millions) 69 (6%) 128 (9%) 310 (15%) 693 (22%) 1,436 (31%) Rural and Small Town Population (Millions) 1,118 (94%) 1,346 (91%) 1,705 (85%) 2,431 (78%) 3,235 (69%) Annual rate of increase -in urban population 3.1% 4.5% 4.1% 3.7% -in rural population 0.9% 1.2 % 1.8% 1.4% Source: Table 32 in "Growth of the World's Urban and Rural Population, 1920-2000." UN 1969. Esti- mates are an average of four alternative series based on existing trends and weighted for each region according to the assumptions judged most relevant. total rural population in 1960. An increase of at least 50% in the ratio of rural population to agricultural land may be implied for many countries. Yet there is no indication that total population will increase less than assumed so as to be able to reduce either rural or urban estimate without increasing the other.1 Perhaps the most striking feature of the urbanization process in the developing countries has been the rapidity of growth of large cities. Many are more than doubling in population and perhaps tripling in area within a decade. This phenomenon is not confined to any region. Abidjan grew from 69,000 in 1950 to well over 500,000 today; Lagos from under 250,000 to 1,500,000; Bangkok from less than 1,000,000 to 3,000,000; and Bogota from 650,000 to over 2,500,000. No close relationship has been established between city-size distri- bution and either relative economic development or degree of urbanization. It is perhaps hardly surprising that the developing countries show considerable variation in the growth pattern of differ- ent urban types and sizes. They contain many of the world's most ancient cities and urban cultures as well as youngest. In some coun- tries such as India intermediate size cities appear to be growing faster than others. In other countries, both the smallest and largest urban centers are growing appreciably faster than those in the middle of the 'Such evidence as exists suggests that both total and urban population in 1970 were somewhat higher than the projections. For the developing world as a whole, the increase in urban population between 1960 and 1980 may equal the total urban population of the developed world in 1960. It may be noted that the projected rural population of the developing countries would still approximate about 70% at the end of the century and the 30% urban population would correlate with a national income per capita average of only a little over $200. range. In others the smaller are growing the fastest. In some, seaports are declining in relative importance, in others growing. Dominance of one primate city is, however, much more evident than in the developed countries, the capital city often being several times the size of the next largest. Such primacy, with significant ex- ceptions, tends to be associated with such factors as low income, small densely populated countries, high population growth, relatively strong agricultural exports and colonial history. Not surprisingly, in view of transport and communications constraints, very big countries such as India have several large centers; but physical features, such as the separated valleys of Colombia, may also lead to several large cities in smaller countries. Notwithstanding these differences, there is considerable evidence that after a certain stage of development a tendency exists towards a less top-heavy distribution of urban size. The largest cities continue to grow, but less fast than intermediate ones. The slowly downward trend in the growth rate for total urban population, indicated in Table 2, also leads to some expectation of a gradual decline in growth rates of the large cities. Such a trend seems likely, however, to be a com- pound of slower growth rates for large urban centers of the richer developing countries as urbanization approaches or passes the 50% mark, and still rising growth rates of big cities in countries at early stages of development, particularly those where total population growth rates have not yet peaked. It is not, however, necessary for the purposes of this paper to attempt sophisticated projections of large city growth.1 Foreseeable difference in growth rates among urban categories would generally not bring about major changes in the urban patterns of individual countries over the next one or two decades. This is particularly true of primate cities. Even significantly higher rates of growth in smaller towns would have little effect in the medium term on the size of the primate city because of the low level from which the secondary cities start.2 Changes in relative rates of growth can be very important in the 'To illustrate magnitudes, if (in rough conformity with past observations) one-half of the projected urban population increase accrues to big cities, the developing coun- tries are likely to have well over 300 million people in cities of over 500,000 by 1980 compared with 130 million in 1960. "In Thailand, for example, if Bangkok's growth is reduced by the equivalent of an extra 2% on the growth rate of the nine next largest cities, the effect will be only to reduce Bangkok's population in 1990 from 7.6 million to 7.1 million (the present population is 3.4 million)-and the difference would be made up by 1993. If the nine cities achieve 3% additional growth, the difference for Bangkok in 1990 would be about one million or some five years further growth. A1 7 long run and hence for policy decisions now. But for most developing countries they are likely to have only marginal significance in the context of the overall urban problems of the next two decades. Urban Employment, Income Distribution and Living Conditions The malaise of urban centers in the developing countries is only too evident in the squalor of the rapidly growing slums and unau- thorized settlements, the deterioration in many public services, the extreme shortage of housing, and the congestion in the streets. Less immediately evident, but certainly no less important, are the growth of unemployment and the worsening of income distribution. Data on urban employment are sadly inadequate partly because the wage-paid element of the labor force is generally very low and the conceptual difficulties in defining "unemployment" and measur- ing "income" for the remainder are considerable. Scattered evidence suggests that unemployment in urban areas, as variously defined, is much more often above 10% than below, and is over 20% in some countries and, more frequently, for the 15 to 24 age group. There are again, however, a few significant exceptions both among countries with a strong record of rural development, such as Taiwan, and among those with particularly high urban and industrial growth, for example, Korea. For large cities, the position appears worse than for the smaller urban centers. In some major African cities estimated unemployment exceeds 30%. The urban unemployment situation appears generally to have worsened over the last two decades, though it is by no means clear whether unemployment plus underemployment is worse than in the countryside.1 The limitations of modern industry as a means of ab- sorbing the available manpower have become increasingly evident. The growth rate in industrial employment tends to be only about half that in industrial output. Nor does this take into account the reper- cussions that expansion of modern industrial output may have in reducing employment in small-scale handicrafts-though this may in some cases be more than offset by employment directly created in associated service industries. Estimates bythe International Labour Office (ILO) indicate thateven after allowing for a decline in participation rates, the total labor force 'it is to be noted that the data usually underestimate the seriousness of the problem, particularly the degree of underemployment and involuntary withdrawal from the labor force as a result of poor employment possibilities, of activities in brief well below the economic potential. 41P; of developing countries will expand by at least 25% between 1970 and 1980.1 Such a rate is unprecedented even in countries which had large-scale immigration, such as the United States. This surge in the labor force derives from the lagged effect of the earlier surge in life expectancy at birth. With few exceptions a continuance of this trend in labor force at least until the middle 1980s is implied by the pattern of birth and mortality rates over the last two decades. Growth in urban labor force is likely to be particularly rapid. The natural increase will be high as a result of past immigration which has tended to raise the average age in the countryside and lower it in the towns.2 Moreover, in many developing countries, and in large re- gions within others, the availability of suitable land for agriculture is already very restricted and the median size of farms small. In such cases, any acceleration in the growth of rural labor force can be ex- pected sharply to increase pressures to migrate, further intensifying urban employment problems.3 The wide divergence between urban employment patterns in de- veloping countries and those in developed countries deserves em- phasis. "Dualism" in the countryside between modern and traditional agriculture is paralleled in the cities by a dualism between the small modern wage-paid sector and the much larger traditionaJ or "bazaar" sector. A large part of this latter group is covered in conventional terminology under the "service sector"; yet its functions are much wider than the nomenclature indicates. A multitude of small enter- prises generally exists in shantytowns as in older areas of the cities. Casual labor and petty trading are highly important not only as a source of employment but also for the economic functioning of the cities and the economizing of scarce resources, in re-use of products as well as in reduced needs for equipment and buildings. Though productivity is low compared to the modern sector, it is far superior to unemployment and in all probability exceeds productivity in mar- ginal employment in agriculture. Unfortunately, concentration of attention on industry has resulted in little analysis of the much larger "service" sector of urban employ- ment. Little is known of its composition and functioning or its link- ages with the modern sector. Yet there is no evidence that it will 'As workers in urban centers enter the labor force later and retire earlier than in agriculture, the urbanization trend of itself tends to reduce participation in the labor force. 'Where migration has been predominantly of unaccompanied male workers, the impact on future labor force may not occur. Where, as in many African countries, additional cultivable land is relatively abun- dant and a large part of rural migration is to colonize new areas, the increased rural labor force may have less urban impact. A1Q decline in importance in the near to medium future. Even looking further into the future, employment in manufacturing seems most unlikely ever to attain the levels of 40% or more of the total reached in the developed countries. The reason is simply that advances in technology are progressively reducing the proportion of labor in the manufacturing process. Machines are more efficient than they were and power is more readily available. The functions of the city and composition of employment are thus likely to remain very different from the present or earlier situation in developed countries. There is no inevitable technical connection between industrialization and urbanization. Solutions based on his- toric Western typology are most unlikely to be appropriate for the urban employment problems of the developing countries. In the developing countries, the service industry predominates over manu- facturing industry in the majority of urban centers and is likely to continue to do so. Closely associated with the employment problem is that of income distribution. It is widely recognized that a large gap exists between incomes in the towns and those in the countryside. Average Gross Domestic Product (GDP) in the major urban centers of the develop- ing countries may be three to five times that of rural areas, with a somewhat lower differential for the smaller towns. Even allowing for higher prices in the towns, the gap is wide and, on somewhat scant evidence, appears to be widening. A similar situation of polarization of incomes has developed within the urban centers with a small and increasingly wealthy group separated socially and often physicially from the poorer mass of the population. In large part, this situation reflects the impact of the technological revolution, an impact much sharper than it was in today's developed countries. The differences between productivity in the modern and the traditional sector are much greater than they used to be-for production of non-agricultural goods, for agriculture, and for serv- ices. Not until modern technologies have spread much more widely in both town and countryside will the underlying causes abate. While the choice of technology alternatives will play a large part in the speed of diffusion of impact, the time required must be measured in decades for most developing countries. Present trends and the pressures of population and labor force thus indicate a further worsening of income distribution over the near future at least. Sharp gains in the small modern sectors of both urban and rural areas will be paralleled by more rapidy rising num- bers in the more traditional sectors. Not only are considerations of equity and social unrest involved. A more even income distribution would also create the basis for mass markets for manufactured goods of lower import content and provide a stimulus to growth and urban employment. With a few notable exceptions, the high natural growth of urban population, the inflow of poor rural migrants, and the increasing un- employment and imbalance in income distribution, have been re- flected in a deterioration in urban living conditions.The shortcomings of present living conditions are, it is true, generally less for the mi- grants than those they have left in rural areas. Parallels can, more- over, readily be found in many industrial cities of the nineteenth century, in the inadequacy of sewerage for instance. What is most disturbing is rather the scale of the problem and the prospect of urban populations increasingly outdistancing the urban services and housing available. Many big cities are experiencing shantytown population growth in excess of 20% a year and a doubling of slum and shantytown population within the next four to six years is now in prospect. The shantytown population already accounts for a third or more of the total population in many cities. Such settlements are therefore very quickly becoming, if they have not already become, the major part of the city in terms of living conditions and urban pattern. Though varying widely in type, social structure and the degree of poverty, they are in large part characterized by absence of even minimal stand- ards of housing, water supply, sewerage, streets, and social facilities. Housing conditions are generally poor in the extreme. The case of Calcutta where more than two-thirds of families are reported as liv- ing in one room is only slightly more drastic than for many other cities of Asia, Africa and Latin America. Housing construction is typi- cally running at a quarter or less of estimated minimum require- ments.' The World Health Organization (WHO) estimates that only a quarter of the urban populations currently receive public water supply in house or courtyard and only a further quarter is supplied from public standpipes.' The situation for sewerage is certainly worse with the backlog of requirements rapidly growing. The position with regard to other public services is more varied. A distinction is necessary between the proportion of population served and level of service provided. On both counts, electricity supply probably has the highest rating. Even in squatter areas without water and sewerage, some system of electricity supply is usually obtained. Though often illicit, make-shift, and far below prescribed 'UN Documnent A/8037-21 August 1970. 'WHO Document A23/P&B/5, 10 April 1970. 4)1 standards, the systems function. Public transport presents a more mixed picture with marked variations in adequacy of service and probably, overall, an appreciably deteriorating trend. Although school enrollment continues to increase, average education stand- ards in many urban areas may now be declining, judged by the pro- portion of children of relevant age reaching various grade levels. Facilities for health have greatly improved in the treatment of illness and, for many illnesses, in prevention. Traffic congestion has greatly increased in the last decade and is itself a major factor in the deteriorating level of public transport services. In many cities, congestion is also seriously affecting the transport of goods. The rapidity of the increase in private car owner- ship presages an intensification of the trend in the future. Resources for Urban Development The acuteness of the shortage of resources in relation to urban growth has probably not yet been fully grasped. On the one hand lie the basic limitations of total resources available for development, on the other the heavy costs of providing urban services, particularly long-life infrastructure to support the rapid increase in urban popu- lation. For countries with low income levels and high population growth, the squeeze between these constraints has become intense. A level of net savings of 10 to 15%, fairly typical of the developing countries, implies net savings per head below $25 a year for the ma- jority of them, and below $15 for many. Even if the totality of these savings could be mobilized for the benefit of the additional popula- tion, the amount per head of the additional population, assuming a 3% annual growth rate, would average from $500 to not much more than $800. By contrast, net national savings per head of population growth in the richer countries of Latin America may exceed $4,000. Obviously, the problems for such countries though serious enough are of a quite different order. Such illustrativefiguresmay appear undulypessimistic inthecontext of urbanization problems since urban incomes and savings are well above the national average. However, though savings in the major towns are typically two or three times the national average, so also is their rate of population increase. Net savings in the towns per head of urban population increase are accordingly likely to be of the same order of magnitude as the national average. Relative rates of popula- tion growth between town and country in other words tend roughly to parallel relative levels of saving. 422 Comparative information on costs of urban infrastructure is scarce. Available data indicate incremental costs of water supply averaging around $100 per head and sewerage about the same. There are considerable variations, but these appear to depend much more on physical conditions than income level. Conventional "low income" housing costs range from a minimum rarely below $1,000 per family unit for the house construction alone to double this figure or more; $200 per head is perhaps a typical minimum. Primary school capital costs for projects with which the Bank has been associated range around $450 per student place, or $90 per head assuming one-fifth of the population in this age bracket. Capital costs for employment vary widely from perhaps around $400 per worker in the more tradi- tional urban sectors to over $1,500 in the modern sector and several times this amount in the more capital intensive advanced technology occupations. With a labor force of one in four of the population, which may be low for many of the towns on which migration is con- centrated, this would give a range of $100 to upwards of $400 per head for employment facilities. Large investments are also needed for transport infrastructure and equipment, health services, electricity, po- lice and fire protection, garbage collection and other urban services. It is thus evident that the total capital requirement for services at existing standards greatly exceeds resource availability for the poorer developing countries, even were the totality of net savings to be de- voted to providing such infrastructure for the additional population. In practice, of course, the resources effectively available to equip the additional population with basic economic and social infrastructure are, even after making allowance for foreign aid, much more limited. Strong pressures for higher living standards by many groups, par- ticularly those which the modernization process favors, and improved replacements for existing equipment and structures, make heavy calls on the nation's savings. Taxation measures sufficient significantly to offset such pressures have yet to be devised. In most developing countries income redistribution by taxation is quite limited. The municipalities, in particular, find themselves increasingly short of revenue to meet both capital and current expenditures required to maintain even existing levels of service, let alone to reduce the backlog. Municipal revenues for all purposes amounted to no more than $84 million in Bombay in 1970, or roughly $330 per head of the increase in population. Even in the much richer Caracas, municipal revenues of $120 million, amount to little more than $1,000 per head of the population increase. In practice, a major part of these revenues is committed to the continuation of existing services to the present population. 423 The information available is inadequate to make any generaliza- tion on the extent to which the municipalities receive financial sub- sidies from, or provide subsidies to, the rest of the country. While transfers are customarily made from the national budget to muni- cipalities for various services, what is not clear is how far these offset, or more than offset, the national taxes raised on the urban economy.' What does seem clear is that the taxation base of the municipalities is generally less responsive to growth in income than that of the na- tion, implying a greater relative stringency at the municipal level as economic growth and urbanization continue. Several conclusions follow from these considerations. The first is the obvious importance of concentrating on ways of economizing in resource use, including maximizing use of existing urban service capacities. Closely related is the question of standards. If only a frac- tion of the urban population can be supplied with basic infrastructure at conventional standards, the choice has to be faced of leaving the rest without basic services or of designing lower, less costly levels of service for a larger part of the population. It is difficult, however, to avoid the conclusion that even with strenuous efforts to economize in use of resources, to modify design standards, and to improve municipal finance, some continuing de- terioration in average living and working equipment for city popu- lation in the poorer developing countries may be in prospect until incomes rise appreciably above present levels and population growth slows. The importance of sufficient information to enable the choice of priorities to be made against a clear understanding of the practical alternatives for the urban population as a whole is correspondingly increased. In these conditions, it is also obvious that considerable emphasis should be placed on "self-help" projects. The personal initiative and work stimulated by such schemes adds both to output and sav- ings in a sector where resource limitations are so evidently crucial. A further deduction is that much more attention should be paid to shorter life structures with provision for future amelioration. In large measure, urbanization problems derive from the long-life nature of housing and other infrastructure of high initial cost. Whereas a 3% population growth will increase food consumption by roughly 3% compared with a stable population, it will increase house con- struction needs by roughly four times if housing has an average life of a hundred years (i.e. 1% requiring replacement with a stable popu- lation). With continued growth in income, the resources per head 'Some Indian studies purport to show a considerable net subsidy from the major towns to the rest of the state in which they are located. 424 available to future generations for such purposes will be greater than at present, the more so as population growth is curbed. The relative social costs of providing now rather than later for future generations are correspondingly high. Finally, the wide variations in income levels among the developing countries involving very considerable, almost quantum, differences in relative resource availabilities, have important policy implications. The proportionate difference between say Pakistan and Chile is much greater than between Chile and the United Kingdom. Costs of pro- viding comparable urban services do not vary to anything like the same degree. Indeed construction costs may at times be higher in the lower income country. The appropriate mix of urban investment and level of services supplied are, for both reasons, likely to vary widely according to the income level of the country. THE NATIONAL/REGIONAL SETTING It is clear that urbanization problems have reached a magnitude and importance necessitating explicit consideration in overall na- tional development policies. It is no less clear that considerations of social policy, equity and regional balance-in large part involving value judgments-as well as more narrowly defined economic con- siderations, must play a considerable role in determination of na- tional urbanization policies. One school of thought holds that much greater emphasis should be placed on creating jobs and improving amenities in the countryside. Urban pressures would thereby be lessened through a reduction in migration. Income disparities between town and country and be- tween different regions would also be narrowed, and help provided to the rural poor, who have as yet received little or no benefit from economic development. And if, unfortunately, it has to be accepted that large-scale unemployment is inevitable for at least several years to come, "Planners should ask themselves: Where is it more con- venient for the unemployed to be, in the country or in the city? In the country, of course."1 Both infrastructure and welfare require- ments will be less in rural areas than in the cities not only because costs may be lower, but also because lower standards are accepted. The foreign exchange element of the infrastructure and of the spend- ing from rural incomes will also be lower. 'Professor Edmundo Flores, "Economic Growth and Urbanization," Rehovot Confer- ence paper, 1971. Those who hold this view contend that the existing urbanization trend, particularly to big cities, is biased by a number of influences. Costs of relocation of migrants are not sufficiently taken into account in considering urban employment, nor in assessing the benefits of labor-saving investments in agriculture. Failure to charge industrial- ists with the full social costs of their operations, including pollution and congestion, biases location and investment towards the big city. Above all, the concentration of political power produces a very fa- vored position for big cities in the allocation of investment resources. Such influences operate through the allocation of development funds and favored treatment for various types of permits. Perhaps more importantly, policies such as those towards exchange and interest rates swing the terms of trade between town and country in favor of the former. On this view, in short, attention and investment have been unduly concentrated on urban areas and on accompanying capital intensive projects which, while they may provide the greatest returns on out- lay, do not maximize the use of the nation's total resources. Or more extremely, the urban population is increasingly a parasitic growth, its privileged position based on exploitation of the rural surplus. What is generally implied is that a given total of local resources and foreign exchange can, with appropriate policies, achieve more in the countryside than in the cities in creating employment, raising living standards, and also possibly in output. By reducing migration, cities will be relieved of a corresponding burden of expenditure. The opposing view sees the solution of rural poverty in accelerated migration to the towns. Urbanization is expected to continue to offer the most important opportunities for increasing employment, output and savings, and thus also the possibilities for increased future invest- ment in both town and country on any but a short view. Concentra- tion of activities in towns provides the basis for specialization, and for increased productivity in manufacturing and supporting services. Large, flexible labor markets with diversified skills are needed to match changing patterns of production. Cities provide concentrated marketsfor the output and are the link with the outside world through which technological know-how flows. The concentration of popu- lation also permits standards of education and health not possible in the countryside if only on account of distances. The informal spread of knowledge of modern methods can be as important as the more formal training possibilities. Urbanization is considered a necessary condition for increasing productivity in rural areas by providing markets for agricultural prod- ucts and, most importantly, stimulating specialization of agriculture between regions. Urban activities provide agricultural inputs and incentive goods, and indirectly promote use of modern techniques. Only by rapid migration can the population of rural areas be brought into balance with rural resources and productivity, incomes raised, and the poverty of the lowest income rural groups, the land- less and marginal farmers, reduced. The dualism between "modern" and "traditional" sectors existing in the countryside will inevitably increase as commercialized agriculture and new techniques spread until the countryside as a whole enters the modern sector. This situ- ation is a long way away and will be retarded by anti-migration policies. That migration of rural labor to higher incomes in the towns may increase disparities between regional average income levels is not denied. But the relevance of such statistical measures of income bal- ance is questioned if the persons who move are better off while those left in the rural areas are at least marginally better off. Reduced den- sities as a result of migration should, in most circumstances, result in higher average productivity per man in the rural areas since there will be less fragmentation and fewer inhabitants in the less produc- tive areas. The national, as opposed to regional, income distribution should be improved.' Higher incomes of the remaining rural popula- tion will, moreover, provide additional markets for urban products. On this view, in brief, the real problem is rural overpopulation not urban overpopulation. The question which should be posed is ac- cordingly not how many should be in towns, but how few are needed in agriculture. As for the basic problem of harmful population growth, the greater the urbanization, the sooner a declining rate can be expected. The relative concentration of national investment on the towns is therefore not considered surprising. The towns provide probably about half of total GNP from a fifth of the population, and higher levels of infrastructure investment are appropriate in the context of more rapidly growing populations, now and in the future, which relieve rural areas of a burden. By contrast, agriculture provides con- siderably less than half of GNP in most developing countries and will continue to grow more slowly and with lower productivity than other sectors. The urban development protagonists accordingly feel that 'For greater inequality on a national basis to be produced by migration, it would be necessary (i) for rural production to increase more than proportionately with in- creases in population and (ii) the returns to population growth to be lower in urban than in rural areas. Such a combination appears generally remote. Income distribu- tion within regions may also be improved by migration providing an outlet for the rural unemployed and underemployed. 4997 within limits of ensuring adequate national food supply, agriculture deserves lower priority. Attempts to reverse or delay the process of urbanization will only lead to dissipation of the very scarce resources. This dilemma between rural and urban priorities has led to increas- ing attention to the possibility of accelerating the development of small- and medium-size towns or creating new urban growth centers. At least prima facie, such an approach appears to combine many advantages. It would relieve population pressures both in the major cities and the countryside, increase the modernization spin-off which urban centers provide to surrounding rural areas, provide less con- gested and polluted urban living at an infrastructure cost no greater and possibly less than in the main cities. It should also reduce re- gional imbalances and capture to a greater extent the potential ini- tiative from regional decentralization. Such generalized approaches to the balance of rural and urban development do not, however, appear particularly productive. No satisfactory basis of facts, experience or theory exists, or may ever exist, for a generalized solution to the opposing arguments. Not only are the interrelationships highly complex. Physical conditions vary greatly, as do existing patterns of urbanization which greatly influ- ence the benefits of alternative future policies. Models to simulate existing patterns have not yet proved possible. Construction of dy- namic models for future growth, which would allow alternative poli- cies for changing existing spatial trends to be adequately tested, does not appear attainable in the near future even in developed countries. For the more rapidly changing social and functional structures of developing countries, the problem is much more complex. The un- derlying springs of development, of entrepreneurship and organiza- tion, or inter-occupational impacts, are still at best imperfectly understood. It is evident, nevertheless, that the three approaches each contain important elements warranting consideration in a national urbaniza- tion strategy. Attention generally has been centered more on urban investment opportunities than on the rural development potential. Cases of very extensive and expensive investment programs in rural areas are, however, not lacking. In some countries, political pressures certainly bias the allocation of resources towards urban centers and distortions in factor prices may seriously compromise rural develop- ment. The social costs of individual units of production in urban centers may often significantly exceed the private cost on which decisions are made. It is less clear than these "diseconomies" offset socio-economic advantages of agglomeration not fully reflected in the benefits to decision makers. A )O The impact of rural development in reducing migration can easily be exaggerated. Policies which provide incentives for agricultural output, for example through correction of inappropriate terms of trade between agricultural and urban products, can certainly often provide a marked stimulus to rural production. The effect on rural employment and in narrowing the gap between rural and urban wages may however be marginal. Much depends on local conditions including the unit value of crops grown, land tenure arrangements and the stimulus given to adoption of new techniques. Evidence of the impact of programs promoting green revolution techniques is somewhat contradictory. In areas of relatively good land, which benefit most from the new technologies, higher produc- tivity per man and per acre has been accompanied by a somewhat higher total of persons employed, particularly where double crop- ping is introduced. In some areas, however, the introduction of farm machinery, even if limited to the minimum required to exploit the new technologies, may reduce labor requirements. Machines re- quired to prepare land in the short period between crops or for diffi- cult soil conditions are, rationally, used for other labor-saving purposes. The position of marginal farmers and landless laborers has often been depressed. In some countries, land reform can provide importantopportunities for increasing agricultural outputwith greater employment. But the process is typically slow due to inherent ad- ministrative as well as political difficulties. With rare exceptions, the impact on overall employment and migration appears likely to be small. In brief, accelerated expansion of agriculture, though obviously important for national development, seems unlikely to go far in re- ducing migration. Indeed, for many countries, the impact of the foreseeable growth in agricultural production seems unlikely to offset acceleration in growth of labor force. Even if rural employment grows by 2% a year, an increase in urban employment of 41/2 % a year will still be needed to provide sufficient jobs for the additions to the na- tional labor force, assuming growth rates of 20% in the urban popu- lation and 21/2% in labor force; a 7% urban employment increase is needed if the labor force grows by 3%.1 The influence on migration of providing greater amenities to the rural population is also problematic. To have a significant impact on the marginal fraction of population that migrates to the town would require very large and prolonged efforts to improve conditions 'The absolute number of jobs required in the rural areas with a 2% expansion rate, will of course still be much larger than that required in the towns with a 41/2% or 7% rate due to the much smaller starting base in the towns. throughout the rural areas. Provision of housing and many utilities at a level judged acceptable in rural areas is likely to be appreciably cheaper in the countryside than conventional urban standards.' How- ever, except perhaps for some richer developing countries, the total number of units needed to obtain a general improvement in condi- tions is many times the number required in the towns to accommo- date the small fraction which migrates. Administrative as well as resource constraints set limits to any feasible policy, however justified such policies may appear. Greater social amenities in rural areas may even promote migra- tion, at least initially. So far as can be judged from limited studies, extension of education, improved transport facilities, the spread of national information media, and increased contact with urban-type amenities tend to increase rather than decrease the impetus to mi- grate to urban centers and from the smaller centers to the larger. A faster growth of small- to medium-size urban centers relative to the largest has obvious attractions. A rapid growth of market and servicing centers can be expected to accompany modernization of agriculture. Marketing requirements are highly geared to the margin of production above subsistence and to the expanded use of fertil- izers and other inputs. Specific promotion of new manufacturing centers is another mat- ter. For the poorercountries in particular, concentration of their sparse physical and entrepreneural resources in one, or at most two or three, large cities may have considerable advantages during initial stages of development. Economies from scale of operation and specializa- tion, contact with foreign capital and dissemination of technology may thereby be maximized. The primate city at the hub of a radial transport network may also at this stage result in economies in the national transport system. Whatever the historic reasons for location, it would be unwise to assume without explicit evidence that the con- tinued growth of primate cities is not well based on considerations of efficiency. Though social considerations may favor decentraliza- tion, it is important that the probable economic cost is realized. Experience in a variety of developing countries, ranging from India and Pakistan to Venezuela and Brazil, shows that promotion of as- sumed "growth centers" can be very expensive both in terms of finan- cial and administration resources. In practice, great difficulties have been encountered in choosing appropriate new "natural" growth centers. It is generally much easier to chose an appropriate location 'Electricity is an important exception; education and health may be. For the same standards, more people can generally be served per investment and operating dollar in urban than in rural areas. for an individual industry than to identify a group of industries that can be attracted to, or generated in, a given area. Dangers of invest- ing in infrastructure which proves to be underutilized are easily demonstrable. There are several reasons for the difficulties encountered. Most small urban centers in developing countries lack the basic infrastruc- ture of transport and services which in developed countries permits manufacturing units readily to locate away from major centers. Man- agement and professional staff are unwilling to move from the major cities. There is, moreover, no clear evidence that costs of urban serv- ices for the same quality of service are less in the smaller centers.' Even if lower urban service costs in the smaller centers were proved, it would still be necessary to show that the differences exceed the net advantages that the larger cities provide from concentration of population and economic activities. Completely new towns seem particularly expensive. A fairly large "minimum mass" is needed for the town to operate as an efficient largely self-contained unit.2 In the process, considerable underutilized capacity has to be created. In existing towns such "lumpy" invest- ments are staggered as capacity limits in each utility are reached. Some "over-capacity" utilization both of services and housing customarily occurs before new investments are undertaken. Lacking the hard-to- define but nonetheless important organization "infrastructure" of existing centers, the administrative requirements of new towns are particularly heavy. Such problems appear noticeably less for centers where above aver- age growth is already apparent. Removal of specific and evident con- straints can in such cases lead to accelerated growth at relatively small expenditure. Towns within a narrow radius of large cities, where in- termediary services of the metropolis can be utilized and public utili- ties more readily linked, can also offer favorable opportunities. 'Even looking at the growth of individual urban centers (perhaps a clearer indication of scale economies since towns of a similar population vary so greatly in density, urban layout and physical constraints), there is little indication in most cases of sig- nificantly higher incremental costs with growth. Many urban services can be dupli- cated without increase in costs. For others, such as electricity and various types of health and education facilities, economies of scale continue up to very large popula- tions. For a minority of services including water (if more distant supplies have to be tapped) and passenger transport, marginal costs rise after a certain size. However, the greatest constraints on economic efficiency as the cities grow appear to lie in the field of the growing complexity of administration. 2Widely differing views are held on the size of the minimum mass. In the developed countries, the minimum size for a viable new town has been progressively increased and 300,000 is now often held as the lower limit. The relevance of Western experi- ence is, however, dubious. In the circumstances described, a pragmatic approach to a national urban strategy, based on general policies to reduce distortions and a deeper evaluation of identifiable practical alternatives for both rural and urban development, would appear the most productive. Such an approach would seek to analyze and reduce underlying biases to- wards urbanization in such fields as taxation and price policy, and in the inadequate charging of social disutilities to private and public decision makers. It would place greater emphasis on investigation of the development potential in rural areas. It would seek to identify smaller or medium size towns already demonstrating above average growth potential where removal of specific constraints should lead to acceleration of growth without heavy prior infrastructure investment. Even more important appears the need for a new perspective to- wards employment, both urban and rural. The availability of surplus labor should be regarded as an opportunity for undertaking relatively labor-intensive investments rather than simply as a problem. At a later stage of development, as employment and wages rise, such invest- ments are likely to represent a much higher "opportunity cost". In general terms, the need is to promote a better combination of the scarce resources of capital, land, skills and foreign exchange with the increasingly abundant supply of labor. There appears, indeed, little prospect of reducing unemployment unless the pattern of demand is changed by policies towards exchange and interest rates, taxation and subsidies, and investment programs, in the direction of goods, serv- ices and production processes in which the relative share of unskilled labor is higher. Only by such an approach does it appear possible to make economic growth compatible with social balance in the existing conditions of most developing countries. How such a reformulation of policy will affect the balance of rural and urban development cannot be decided in the abstract. The widely varying conditions will produce different answers for different coun- tries. Neither increased emphasis on rural development, nor on urban development will, however, by itself produce a panacea. Farms, rural villages, market towns, intermediate centers, and major cities form a continuum in which complementary action is required to make the best use of national resources. The dichotomy between urban and rural development is to this extent false. Difficult reconciliations between higher returns on projects as conventionally evaluated and greater equity appear inevitable. The unfortunate reality is that the best economic returns for further in- vestment tend to be where conditions are already relatively favorable and prosperous-within sectors such as farming as also between re- gions or between urban centers. This does not however mean that increasing GNP is at variance with reducing unemployment. On the contrary, the evidence suggests that maximization of use of total re- sources, including labor, rather than maximizing returns (as narrowly defined) on individual projects can often provide the best prospect for growth in total output. In not a few developing countries, particularly where growth of agricultural output is already adequate and rising faster than popula- tion, the policy review suggested may indicate the desirability of greater concentration of investment in the towns-though, as the next section indicates, by no means necessarily of the present pattern. The costs of providing additional urban employment compared to result- ing output, evaluated from both economic and social aspects, may in such circumstances prove more attractive than for rural projects. The arguments against very capital-intensive industrialization should not be identified with urban occupations as a whole, since only a small percentage fall into this category. Even, however, where the approach suggested leads to stronger emphasis on rural investments, it should be recognized that the prob- lems of urbanization are unlikely to be significantly reduced in the shorter term. Indeed, it should be accepted that success in stimulating overall national development, through green revolutions in agricul- ture for example, may lead, even with more appropriate technology, to an intensification of urban pressures. THE PROBLEMS OF URBAN CENTERS The despondency surrounding the task of ameliorating urban con- ditions in the developing countries arises primarily from the speed of urban growth and shortage of resources, human as well as finan- cial. There are further contributing factors which, though also ap- parent in developed countries, operate with much greater force and are less amenable to solution due to the shortage of resources. For instance, experience has shown that the provision of more and better urban highways provides by itself no more than temporary allevia- tion and does not solve traffic congestion. Metros or urban railways tend to have their major impact in diverting passengers from buses. Investment in transport facilities accordingly can appear of dubious permanent value. Housing also appears as a "bottomless pit", individual projects being too marginal to have any real impact and the costs of major pro- All grams beyond the realm of possibility. In most developing countries, "low-cost housing" tends inexorably to provide shelter for the me- dium rich rather than the more needy poor whose payment of in- direct taxes often subsidizes it. The "trickle down" of space freed by better-off inhabitants moving to new houses proves to be negli- gible. Water supply and sewerage projects often appear to be fight- ing a losing battle, the backlog growing faster than it can be filled. Soaring urban land prices make urban projects increasingly expen- sive. Comprehensive urban plans have tended to be unrealistic in terms of resources and implementation capacity, and out-of-date by the time they are completed. They have had little influence on decision making. Urban administration is woefully lacking in capacity to deal with the problems and highly fragmented. The list could be readily extended. Yet within less than 20 years, the present populations and areas of urban centers will account for less than a third of the total; the same is true of employment and of housing and urban services. Inevitably many new roads and transport facilities, dwellings and utilities, em- ployment sites, markets, and social facilities, will have to be provided. Equally obvious, mLuch can be done to promote more efficient use of public and private resources and to generate additional resources. Here, the urban centers of the developing countries have some ad- vantages. The rapidity of their growth, and the still small number of industrial plants and private automobiles, offer the opportunity to develop a more rational urban pattern and avoid the necessity of very costly reconstruction which now confronts most major cities of the industrialized world. The opportunities for improvement derive principally from three sets of considerations. The first set concerns the important and ex- tensive interactions between investment decisions usually undertaken in an unrelated manner either by individuals or by various public entities. Because of these interactions, private benefits, or benefits of individual public enterprises, may vary considerably from the so- cial benefits. The possible "externalities" or social "disutilities" of industrial lo- cation have long been recognized in terms of pollution and other nuisance imposed on the community. Fortunately for the developing countries, these effects are generally much less than in the past and many more industries can now be closely integrated with residential areas. The impact of industrial location on transport and utility requirements, and the similar impact of business and residential location and densities, have received less attention. Even within resi- dential areas, private development of small lots can result in great 434 waste of space, particularly for roadways.' Similarly, uncoordinated investment in different transport modes can result in large unneces- sary expenditures. The second set of considerations concerns the extent to which levels of urban services and housing can be tailored to the resources available, both public and private. Poor housing in the urban centers is primarily a reflection of the basic poverty. Attempts, such as have often been made, to set housing standards out of line with the re- sources of the inhabitants and the community cannot succeed. Simi- larly, the scarcity of resources means that decisions on the level of public services provided are tantamount to decisions on the propor- tion of the urban population to be served. Closely connected to this question of standards is the highly important opportunity to stimulate self-help. The third set concerns pricing, including charging by way of taxa- tion. Considerable opportunities certainly exist for using pricing mechanisms better to harness market forces in producing a more efficient pattern of urban development, as well as to increase the total of resources available for public purposes. Since private investment customarily accounts for over two-thirds of urban investment, and detailed physical regulation by public authorities is at best weak, pricing policies must inevitably occupy a central role in measures for increasing urban efficiency. In the context of the concentration of population and economic activities in the towns, these considerations provide the main ra- tionale for treating the urbanization problems of developing coun- tries as a separate field from the more general problems of poverty and underemployment or of individual sectors. Nowhere are the problems and opportunities better illustrated than in urban transport, urban land use and housing, three fields which together play a lead- ing role in determining the patterns of urban growth. Urban Transport and Urban Patterns Despite many studies, no simple physical patterns have been es- tablished as optimal for the cities of developing countries. Existing facilities and topography as well as differences in functions, and in social values and organization, necessitate considerable variety. Some considerations do, however, appear to have general application. Up to a certain size, generally well below one million inhabitants, higher densities can result in considerable economies in the provision of 'The new town of Milton Keynes in England is illustrative of the potential. Even with 90% of residents owning cars, congestion will be avoided with only 11% of its area taken by transport and communications facilities compared with 30% and over in many American cities. transport and utilities.' For the larger cities, general urban spread based on concentrated employment in a single central district is clearly uneconomical in resource use. An urban pattern with several nuclei of relatively high densities connected by major transport ar- teries, and with considerable open space or low-density development preserved between them, can offer considerable advantages. Close location of residences and employment, which with city growth requires the establishment of new employment groupings away from existing centers, can secure considerable economies. Fur- ther opportunities exist in the development of new urban areas on a sufficiently large scale to enable economies of road and utility layout as well as better environment to be achieved. The appropriate em- phasis in developing the cities is in general likely to be on largely self- contained urban areas which substantially reduce transport needs with adequate mobility between them to reap the benefits of spe- cialized services. The importance of transport to the pattern of the urban centers and the efficiency of their functioning is such that it is difficult to discuss other urban problems meaningfully without becoming involved also in urban transport. The existing urban pattern, particularly of housing and employment, largely determines present transport needs; the provision of new transport facilities largely determines the future pat- tern of the city. Deficiencies in urban transport facilities add substan- tially to the costs and difficulty of moving both goods and workers within and through urban areas; transport access is a prerequisite of urban development. Transport constitutes moreover a major item of public capital expenditures in urban areas, often to an extent that significantly limits the availability of funds for other purposes. It is also a sector where large economies appear attainable, not least in utilizing more fully existing transport facilities. Similarities between traffic congestion in the business centers of the major cities, particularly the richer ones, and congestion in de- veloped country cities should not obscure some fundamental differ- ences. The much higher rate of population growth means that what- ever solutions are sought, a high level of investment in transportwill be required on grounds of population growth alone. Secondly, despite the rapid rise in the number of private automobiles, usually over 10% a year, in most developing countries the number of automobiles is still very small. Less than 10% of the world's automobiles are in de- 'The point at which congestion, not only of transport but also in providing utilities such as sewerage funnelled through narrow arteries, causes rising unit costs with increased density, obviously varies considerably according to existing layout, physical conditions and costs of reconstruction of properties. A RE; veloping countries. For many years to come, the great majority of the city populations will, on account of their income level, have no option but to rely on walking, bicycles or some form of public transport. The rise in private automobile use at this early stage of development greatly intensifies the problem because of the high ratio of road space per passenger involved and the generally low proportion of road space to urban area.' But even with a curb on private automobile use, investment costs of providing roads and other expensive long-life transport infrastructure and public vehicles on an adequate scale will remain high. Nevertheless, since roads represent the largest part of transport in- frastructure investment and private automobiles are the most obvious cause of congestion, it is here that attention is understandably cen- tered. The balance between personal and mass transport, between vehicles and roads, is not only at the heart of the urban transport problem. The decisions taken regarding this balance will fundamen- tally affectthe quality of urban life in the future. The developing coun- tries are still in a position to avoid exaggerated dependence on the private automobile from which many developed countries are now trying to escape. In considering the relation of private automobile to road expendi- tures, generalizations based on national totals are of little help. On highways between towns, additional private vehicles cause little con- gestion or road maintenance. The marginal costs incurred through vehicle use are likely to be below average costs. In city centers, where each road user imposes costs and delays on others, the reverse is true. Widening of existing city streets is very expensive, provision of urban highways even more so, typically costing well above a million dollars per lane-mile. For a 10-mile stretch, the cost per car is likely to be in excess of $2 per day.2 Additional urban parking space represents a further important investment, particularly if costed on an "alternative use" value. The growing use of private cars also slows down and increases the cost of public transit and goods traffic. Outside of peak hours, the marginal costs of private cars are much lower. However, the costs of additional roadways and public utilities incurred in "city sprawl" resulting from private automobile use also have to be taken into account. 'The typical Western European-North American city devotes 25% or more of urban land area to streets and highways. in developing country cities the proportion is often only half as much. 'Assuming that a new expressway lane can accommodate an additional 3,000 cars during a two-hour peak period then, on a conservative costing of $1 million per lane/mile and a 10% rate to cover interest, amortization and maintenance, the public cost of a ten-mile highway per automobile round-trip at peak periods is about $2.50. The conclusion appears inescapable that the urban motorist is gen- erally charged well below the social costs incurred. Considerations of national priorities and the use of foreign exchange reinforce this viewpoint. Expenditure on automotive equipment often equals that on road infrastructure. It has a high foreign exchange content and is con- centrated on the upper income group in the large urban centers. Though much more extensive analysis is required, there appears to be a considerable element of truth in the contention that a continua- tion of present trends in allocation of resources for the private auto- mobile may seriously undermine the possibilities for more general economic and social development. Congestion can be relieved by staggering working hours to reduce peak travel. Road capacity can also often be increased at relatively low expenditure by better traffic controls, particularly synchronized traffic signals, and by construction of short highway links to complete major circuits. With such obvious exceptions, investment in urban highways in heavily built-up areas is difficult to justify unless and until complementary steps are taken to restrain private automobile use. Traffic congestion is a method, albeit a poor one, of restricting the growth in private car use and of providing incentives to locate new employment away from existing central business districts. Insofar as steps are not taken to charge automobile use with the costs incurred, it appears better, given the sparse resources available for other critical needs, that restriction by congestion should occur at lower rather than higher levels of inner-city highway investment. The logic of the traffic congestion situation is to tax the urban mo- torist much more heavily, particularly at peak hours. Several schemes for peak hour charging are under consideration in developed coun- tries, but none has yet been introduced. In developing countries it is perhaps more realistic to attempt a progressive increase in private vehicle taxes, with a surcharge for congested cities, combined with the prohibition of private vehicles in central areas at peak hours, and much stricter parking enforcement with higher parking charges. But it should be recognized that-as for many other policies such as those to reduce births-the prospects for effective measures are greatly enhanced if underlying conditions can be improved to reduce the demand. Adequate provision of alternative transport and develop- ment of urban patterns that reduce trip requirements must be re- garded as a partner in policies to control private automobile use. The main problem is the provision of transport facilities for the greater majority of the population, rapidly increasing in numbers, who cannot in any case afford private cars. Better facilities for walking and bicycles are both important and usually neglected. However, mass transit must play a much greater role than in most cities in de- veloped countries. It is an essential component in linking workers with jobs. Several alternative and often complementary modes need to be considered-buses, jitneys, taxis, urban railways and metros. One general point, however, deserves note. Improved mass transit in the developing countries' cities is unlikely of itself, even if offered free, to wean a significant number of passengers away from private automobiles unless automobile user charges are raised considerably. Social and income differences between the two classes are generally too great. Buses are, and seem likely to remain, the most important type of mass transit. They have many advantages. Capital costs and road space requirements are low in relation to passenger capacity. Running costs can be readily reduced by taking buses off the roads in off-peak pe- riods. Routes are flexible and can be altered as cities grow and change. The frequency of stops to meet the needs of the large number of passengers and limited maneuverability, however, cause greater dis- location of traffic flows than the smaller "jitney-type" vehicles with a handful of passengers. Jitneys are also more effective on less popu- lated routes and provide greater convenience for individual origins and destinations. Both, the buses more than jitneys, suffer from idle capacity during off-peak periods though to a lesser extent than other modes of passenger transport except taxis, which are more expensive but can convert into jitney-type service during peak periods. Faced with both mounting requirement for mass transit and in- creasing congestion-which slows buses and streetcars even more than it does automobiles-attention is increasingly turning to systems of priorities for buses. Separate lanes for buses can provide great im- provement in service while lowering operating costs. Unfortunately, in heavily congested inner-city areas, the practical difficulties of ex- clusive surface rights-of-way are great. On existing roads, serious underutilization, difficulties at intersections and stops, as well as en- forcement difficulties, are likely to arise. New lanes for buses are very expensive. However, when bus traffic is sufficiently heavy to permit near capacity utilization of separated lanes, at least four times as many passengers may be accommodated as a lane of private automobiles. Table 9 in Annex 1 provides some estimates. Unfortunately, little expe- rience is available to provide reliable guidelines on practical systems. At the opposite extreme, capital and foreign exchange costs of metros and urban railways for infrastructure as well as rolling stock, are very heavy' and routing is inflexible. Capacity of rapid rail trans- 'Excavation costs for metros vary greatly but are typically above $15 million per route/mile; total costs may be upwards of $20 million per lane/mile. port is large at 30,000 to 40,000 passengers per hour per track. This represents 10 to 20 times the capacity of urban highway lanes used for private cars and at least four or five times an exclusive bus lane. The combination of high capacity and inflexibility, however, necessi- tates coordination with lesser capacity modes as feeders and dis- tributors and also more passenger interchanges. Unless fares are coordinated among transport modes, charges per passenger trip tend to be correspondingly higher. Generally speaking, due to the high capacity and expense, a multi- million population with relatively high density both of residence and employment centers appears a prerequisite for consideration of the metro alternative. In such cities, a metro may be viable when surface congestion reaches a point that necessitates further extensive high cost urban highway construction to provide for essential traffic move- ment even with feasible restriction of private automobiles. It will not remove congestion from the streets but it can, in appropriate circum- stances, provide the minimum peak load transport facilities required more cheaply than by additional highway construction and surface transport.1 Several cities in developing countries are considering metro construction ranging from Calcutta to Caracas, Singapore to Sao Paulo, and many more can be expected to do so with the future growth to multi-million populations. Such considerations emphasize the need to consider urban trans- port as a system covering all the various modes both in terms of econ- omizing total resources and in the closely connected aspect of pricing policy. Providing for peak load traffic by whatever mode is more expensive, as a result of unutilized off-peak capacity, than providing for traffic levels at other periods. The relief given at times of peak load compared with other alternatives for the urban transport system as a whole is, therefore, of central importance in evaluation of proj- ects such as metros. The feasibility of alternatives cannot, however, be adequately assessed without consideration of pricing policy throughout the sys- tem. Almost all rapid transit systems throughout the world lose money and require heavy financial subsidies by public authorities. Such finan- cial subsidization -whether or not justified on socio-economic grounds-together with the heavy initial costs, may exclude serious considerations of metros by the poorest cities unless they are very large and have high densities. Meeting peak load needs adequately with buses in heavily congested conditions with corresponding low 'This conclusion is generally supported on the basis of present road use. Some ex- perts believe, however, that metros cannot compete with the new road/exclusive bus lane alternative if private vehicles are strictly controlled. levels of service is also likely to result in deficits for bus companies. A general increase in fares may cause the familiar downward spiral of lower utilization, declining service and increased deficits. To raise fares for peak hours without a corresponding charge on other modes, particularly the private automobile, is likely to be politically difficult as well as discriminatory. "Ideal" conditions of full user charges on private automobile and coordination of public transport might well permit fare levels both politically acceptable and sufficient to cover costs. Private bus com- panies do make profits though often not carrying a significant part of the peak load. There are evident possibilities for economies from better coordination of routes. However, even with heavier taxation of private automobiles, some compromise on public transport pric- ing may be inevitable. Shortage of resources restricts possibilities of expanding the transport infrastructure. Historic patterns of services and pricing cannot quickly be radically changed. The multiplicity of operating agencies, both public and private, greatly complicates coordination. In practice, the question tends to become one of mini- mizing the financial burden while ensuring that the most economical pattern of transport modes in terms of overall resources is progres- sively attained. It should be recognized, however, that holding of urban transport charges below costs may not only increase congestion and day time densities in central business areas to levels where increasing marginal costs of facilities apply. It may also induce dispersed and much lower densities elsewhere, causing unnecessarily high roadway to popula- tion ratios and expensive utility connections. Unless attention is focused on this longer term context of interlinkages, attempts to solve the urban transport problem based on existing needs can easily aggravate the future situation. The greatest scope for economizing in urban transport and reliev- ing congestion in the longer term lies indeed in influencing the growth of the cities in ways which reduce the need for transport, particularly to and from the present central business and industrial districts with their already determined and usually insufficient and inefficient road network. The provision of a highway between two distant cities obviously does not affect the location of the two cities and to only a limited degree, at least in the short term, the location of population between them. An urban highway can vitally and quickly affect the pattern of a city, the location of its occupations and its dwellings, and its growth potential. Traffic generated by trans- port facilities is as important a consideration as the meeting of exist- ing demand. Too often consideration of existing transport demand, or that pro- jected on the basis of existing trends, precludes consideration of how the pattern of demand could be changed and reduced by better sit- ing and coordination of transport facilities and by better transport pricing. Conversely, haphazard regulation of urban land use can multiply travel requirements. The two are intimately connected. Similarly, scarcity of housing leads to immobility in residence change and unnecessary cross traffic, while the development of new housing areas is dependent on provision of transport facilities. Consideration of urban transport, accordingly, cannot be separated from that of land use guidance and housing. Land Use Guidance Measures to guide and direct land use are of obvious importance to the growth pattern of urban centers. Nowhere perhaps are social and individual interests likely to be at greater variance. Greater or lesser powers to control land use exist in almost every city in the world. They comprise specific measures such as zoning, physical building regulations and permits, controls on pollution and public purchase of land, as well as more general measures such as taxation of land, property and betterment values, and rent controls. The prob- lems to which these measures are addressed range from efficient zon- ing of different activities, particularly industry and housing, and consolidation and acquisition of land for specific purposes, to the reduction of land speculation and the raising of tax revenues. In the urban centers of developing countries, the problems of land use are particularly acute. Land values are in many cases mounting precipitously' to levels often in excess of similarly situated sites in cities of the developed world. Sites with, or readily accessible to, urban services remain unoccupied often to a much greater degree than in developed countries.' Effective regulation of location by user type is frequently minimal, and collection of land and property taxes only a fraction of a realistic assessment. Stocks of undeveloped public land suitably located are meanwhile often rapidly diminishing. Many attempts have been made by municipal authorities to im- prove their land use powers and programming. It cannot be said, unfortunately, that the progress to date has been great. The legal bases for acquisition of land, control of type of use, building permits, 'For example, in Taipei land values are reckoned to have increased roughly 1,000%/o in under 15 years. 'Unbuilt land within metropolitan areas frequently exceeds 40%, e.g., as shown by a survey of 103 cities and towns in India; also Bangkok, Buenos Aires. A A t taxation powers related to unexploited land and betterment values, are more often than not quite inadequate. Nor are those that are available energetically used. This is not merely a problem of national or local politics, though this aspect is often a major contributory fac- tor. In the dire straits in which the city authorities find themselves, it is difficult for them to turn down proposals for construction that yield immediate additional employment and increased tax revenues even if they cause considerable disutilities. The rapid increase in population, in income level, and correspond- ing advantages of central sites would in any case lead to increased urban land values, particularly in central areas. The trend is reinforced by investment and speculative advantages of urban land in the ab- sence of comparable investment avenues, a situation often exacer- bated by quasi-monopolistic land ownership. Perhaps even more important is the relation between population growth and supply of urbanized land with adequate services. Compared with the devel- oped countries, a considerable scarcity value attaches to urbanized land due to the financial difficulties of municipalities and utility agen- cies in extending their networks at a rate commensurate with popu- lation growth. When to such tendencies are added the inherent divergencies between private decisions and social benefits in urban land use, the very limited public financial resources, and the importance of land costs in public projects, it is abundantly clear that market values are usually not an efficient allocator of urban land. Also obviously, the capacity does not exist, either now or in the foreseeable future, for efficient detailed land planning by public authorities, and even less for effective implementation. In practice, a compromise has to be sought in which controls and public acquisition of land or devel- opment rights provide a framework which allows the price mecha- nism to play a still important role in detailed allocation. The policies should also seek to augment public financial resources by taxes on land profits exceeding levels necessary to provide the de- sired incentives. Charges for utilities higher than those necessary to cover costs may constitute a form of such taxation. Given the present exiguous levels of municipal budgets, more adequate land taxation is generally an essential element in programs for bringing about an overall amelioration of urban conditions. An initial review of the measures, ranging from simple regulatory controls to land nationalization, which various countries are using is given in Annex II. Land acquisition by public authorities and sub- sequent development by, or in cooperation with, private developers, may provide the most effective route to establish efficient urban patterns and also the easiest means to capture a large part of the value created by provision of urban services and by the growth in total population and income. Supplementary measures can improve the functioning of the land market. In this connection, development of limited new urban areas on an integrated basis appears to offer particular advantages, social as well as economic. Low-income housing, or "site and services" proj- ects can be combined with a variety of other land uses, including employment places and community service centers. Prices, or rents, obtained from land associated with the higher income, commercial and manufacturing uses, together with the economies realizable from marginal additions to utility capacities, can then permit much lower charges, compatible with income levels, for the land devoted to low- income housing. Segregation of the rich and poor can be reduced in a manner compatible to both. Physical controls and building regulations provide an important method of generalized control. However, in many cases, existing regulations are often inappropriate to prevailing conditions and can easily do much harm. Housing standards or zoning regulations based on Western concepts which take little account of prevailing customs of business/residence mix provide examples. Taxes on betterment and on unutilized urban land, more rigorously defined and imple- mented, can direct the growth of the urban centers in more rational directions as well as adding to financial resources. Selected provision of urban services can further influence urban growth with transport occupying a leading role. The denial of urban services to localities not in accordance with the general framework for urban develop- ment, though so far little used and politically difficult, could also be effective, particularly if adequate services are provided in localities chosen for development. Indeed, expansion in the supply of land with adequate urban serv- ices for generally defined purposes at appropriate new locations appears as a central objective to which individual land control and taxation measures need to be specifically related. The new urban extension areas require even more attention than the old which will soon constitute only a minor part of the total and are much more expensive to alter. The importance of the new areas lies not only in conserving resources by better urban patterns but also in relieving the pressure on the older sectors. At a later stage, with more wealth and municipal resources, and lower population growth, the remodel- ling of older areas can be more effectively tackled. AAA Urban Housing Food, clothing and housing constitute basic necessities for living. "Progress" accordingly appears questionable if it takes the form of higher production of less essential goods accompanied by acute and growing shortage of housing for the poorer mass of the population. Such considerations add further point to the importance of housing as customarily the largest component of urban investment. Housing typically represents well over 10% of total private and public national gross investment, with by far the largest part in urban centers'. It is probable that over 200 million new dwellings will be needed in the urban centers of developing countries between now and the end of the century simply to house the additions to urban population. This total is almost equal to all the urban dwellings of the world in 1960. Perhaps the most salient feature of the housing situation is the stark fact that typically well over half of the urban population cannot afford minimal "permanent construction" housing, even if financing arrangements are made available or limited subsidies given. An exam- ple, purposely on the high side, will illustrate the point. With a family income of $400 a year perhaps 15% may be available for housing (excluding services and maintenance) after allowing for food, cloth- ing, transport to work and other necessary expenditures. On this basis, the capital cost of the house which can be afforded with a 20-year loan at 12% is only $450, or at 7% $640, compared with minimum costs for conventional housing usually exceeding $1,0002. Nor, in prevailing conditions of acute shortage of housing and re- sources, can public subsidization provide a solution. Total savings of the countries and even more of the municipalities are quite inade- quate. For at least the poorer countries, encompassing the great majority of the people of the developing world, basic resource con- straints preclude public provision of new housing for the urban poor directly or indirectly. There is, moreover, ample evidence that hous- ing provided for a relatively small proportion of the poor families at standards appreciably above what they could normally afford quickly tends to be occupied by higher income groups. Either the selection 'For new towns, housing may exceed 40% of the total investment. 'A rule of thumb for housing that can be afforded is 2 to 21/2 times family annual in- come. This is probably excessive at prevailing interest rates and for large families, or where incomes are so low that food and clothing preempt a very high proportion. Inflation and high interest rates cause additional difficulties; in real terms, the pay- ments during initial years are much higher than in later years-but the high initial levels are likely to be prohibitive. 445 process is subverted, or informal arrangements are made by the nominal tenants'. A second major feature is the extent to which dwellings, albeit of "substandard" quality and subject to great overcrowding, have in fact been provided by the unaided self-help efforts of migrants. In many of the rapidly growing cities, over half the population has se- cured accommodation by such efforts and many others will reach this position within the next decade. Because the sites are often il- legally occupied and the construction supposedly "temporary," no note is generally taken of this output and investment in national ac- counts. Yet in total, such activity undoubtedly represents a major share of dwelling construction. The situation is less one where provi- sion of housing attracts migrants-of which there is little evidence- than one where the migrants provide the housing for themselves. The migrants appear at least as much to be providing a solution as to be creating a problem. The wide variety of shantytown conditions deserves attention. The densely crowded inner-city slums to which migrants, in many coun- tries at least, initially gravitate in their search for work, are usually quite distinct from the shantytowns of the urban periphery in the problems posed and possible solutions. The shantytowns themselves are widely dissimilar in density, income level, and type of location. Of particular note, many shantytowns of some years standing show evidence of self-generated upgrading. Dwellings of more permanent materials appear along with a proliferation of small enterprises. Com- mercial and social organization improves. Rudimentary urban facili- ties are improvised. The essential conditions for such dynamic ame- lioration are imperfectly understood; but certainly security of tenure and development of community organizations play essential roles. Populations of shantytowns often appear to place more emphasis among their wants on the provision of minimal utilities and services, particularly water supply and health facilities, than the provision of more adequatedwellings as such. For shantytowns on the peripheryof large cities, transport to areas of employment is of great importance. There are several instances of under-utilized public housing or "site and services," simply because the location has been too far from em- ployment centers and the transport facilities inadequate or too costly. Such considerations pose a series of dilemmas to the public authori- ties. Initial reaction to the shanty settlements has often been to tear 'There is also evidence of considerable difficulties of upkeep and rent collection even in subsidized public housing. Moreover, with housing shortage also existing at higher income levels, the "trickle-down" of vacated housing space to the large mass of poor families is negligible. 446 them down as both illegal and an urban blight. It is by now generally appreciated that such action provides no solution. New shanty set- tlements appear more quickly than the old are removed, and all that is usually achieved is destruction of useful capital investments. Much the same considerations apply to pulling down and reconstructing central slums. Unless the city is wealthy enough to provide in advance for relocation of the inhabitants at a cost they can afford, the imme- diate loss of dwelling space will compound the problem and tend to offset potential future advantages. For the poorer majority of devel- oping countries, large-scale reconstruction of central areas appears a luxury that cannot yet be afforded. Improvement of existing squatter settlements raises similar, if less intense, questions of priorities. If resources are concentrated on im- proving existing settlements, at the end of a few years new settlements duplicating the poor conditions of existing ones but of larger area will have sprung up. If measures are taken to restrict new settlements, the growth of population density in the existing settlements will be intensified, compounding the problems there. Settlement improve- ment, in brief, does not add to the stock of housing, the basic problem. Moreover, the difficult sites of existing settlements, often on moun- tain-sides or swampy land, their restricted access and low proportion of unbuilt area, tend to increase the cost of improvements. In such a situation, it may seem more logical to concentrate on providing services and achieving a better layout in new settlements specificallydesigned to allowfor future amelioration. Much, however, depends upon local conditions. Cases certainly exist where, with rela- tively small public expenditures combined with local self-help, major improvements can be achieved in living conditions without pre- empting much public resources otherwise available for creating new urbanized areas.' The benefits may then clearly justify the allocation while reducing pressures to demolish the settlements with conse- quent waste of existing capital. Over the last decade, considerable experience has been gained in "site and services" projects providing urbanized land on which low- income groups can construct their own dwellings often with or- ganized technical assistance. While the scope of such projects in relation to population growth is still very small, there is growing rec- ognition that this type of approach deserves greater emphasis. A con- siderable variety exists in space allocated per family, in provision of 'For example, the Kampong improvement program now under consideration in Dja- karta would entail costs of only about $15 per head. Difficulties may arise from substantial improvements producing a jump in land values if the occupiers of the land are not the owners. 447 utilities and social services, in the assistance given, in the type of tenure, in stipulated standards for construction of dwellings, and in the participation of private development enterprises. The schemes may or may not be subsidized, though any subsidy element is likely to be much less per family than for public low-cost housing. Considerable variation in the organization of site and services proj- ects is hardly surprising in view of the differences in income levels, social values and organization, climate and topographical features. The more conspicuous successes, however, tend to demonstrate the importance of securing involvement of the population concerned through community groups, not only for initial implementation but also in creating a momentum for further amelioration, in provision of communal facilities and in the collection of payments.1 Typical costs per urbanized family plot, including provision of water and sewerage, roadways and street lighting, and reservation of space for social facilities, commercial and cultural activities, may be of the order of $300 to $600.1 Where community labor is organized and given technical assistance, costs may be further reduced.' Econ- omy in costs of site and services and low-cost housing projects is however only one aspect of their contribution. Properly planned in relation to employment, social services and environment, they can lead to a marked improvement in general living conditions while re- ducing costs of supplying urban services. Placing the major stress on the "site and services" approach does not imply that possibilities of low-cost housing should be ignored where action in this field would be appropriate. In the richer Latin American cities, low-cost housing schemes can make a significant contribution. Even for their poorer income groups, over 20% of in- come may be devoted to housing needs against probably less than 10% in India. Singapore, with its rapid economic expansion, limited population growth, and well above average income for the region, provides a further, obviously successful example. 'Rent collection appears generally easier than for public housing. The amounts as- sessed on "site and services" schemes are much lower and the families, with their own capital involved, have a greater interest in maintenance. Community group organizations can more readily apply sanctions, or provide help in deserving cases, than can public authorities. 'Dakar "site and services" project now under appraisal by the Bank will result in costs of about $300 (excluding land costs) for the relatively large families concerned. The average cost per housirig unit for 16 public housing programs in the same city is over $4,000. 'in one El Salvador program, the overall costs per family in small planned units of development, including land and the self-help construction of houses with indi- vidual water supply and toilet, the materials being bought on a bulk basis, appears to have been brought down to below $800 for a 44 m' floor space house. AAR In this connection, a far greater effort in cheap building techniques using local materials appears called for. Little effective use seems to have been made of local building research centers. Though results with prefabricated dwellings have so far not been particularly en- couraging, scale economies for various components including on-site fabrication, appear to offer brighter prospects. Much more analysis is required of the relative merits of high-rise apartments in reducing land use and, if appropriately located, also transport needs, compared with lower density housing. There has been a tendency to minimize the contribution of housing to national output and welfare because of its high capital to output ratio. Such a simplistic approach, however, may considerably under- estimate the importance of housing partly because both "capital" and "output" are ambiguous in this context. In the developing countries, construction of housing and supporting utilities makes little call on capital equipment but has a relatively large input of labor, mainly unskilled. Shortages of skilled labor for certain tasks may occur; but the training required to overcome such shortages is relatively simple. For low-income group housing, individual or small entrepreneur ef- fort predominates. Indeed construction appears as one of the easiest occupations for surplus labor to enter. Building materials for low- income housing can generally be developed relatively easily within the countries from local raw materials', and here the overall capital to output ratios are generally favorable. Again considerable employ- ment is created. For other materials, such as timber, a contribution may also be made to development in poorer areas of the country.2 Investment per house is high in relation to output as valued by rent obtainable, though the high rents charged in overcrowded de- veloping country cities may produce a lower ratio than in the devel- oped countries. The ratio is certainly substantially lower for less permanent buildings. Nor, in terms of welfare, does rent appear to be an adequate measure of output. Perhaps more importantly, the savings required are largely gen- erated by the house construction. There is ample evidence that peo- ple building a house at all income levels save more, and thus reduce their other consumption calls on resources which are likely to be more capital and foreign exchange intensive. Scattered evidence from 'In some African countries, however, the proportion of building materials imported appears to be very high. 2It can be argued that for some countries at least, stimulation of building provides a better route to industrialization than specific promotion of import-substituting in- dustries that, on experience, are likely to be very costly or premature and, because of higher import content, are much more limited by foreign exchange scarcities. site and services projects indicates that the initial investment in ur- banizing the land is at least doubled by the efforts of the families in constructing dwellings on the land provided. Indeed, only a system which mobilizes this savings/investment potential is likely to succeed in substantially alleviating the housing shortage.1 The importance of stimulating savings by way of housing suggests the need for greater attention to the financial mechanisms for provid- ing housing credit. The extreme stringency of public resources, in- cluding foreign aid, does not indicate a high priority for subsidizing the housing of the relatively well-to-do either directly or through artificially low interest rates. But the success of certain countries, such as Brazil, in mobilizing savings through housing finance institutions with a modicum of seed capital and technical assistance demonstrates a wider potential. For the poorer income groups, the need may be more for credit for construction materials organized through com- munity groups. Looking at housing as a separate urban sector does not, however, provide a sufficient basis for determining an appropriate strategy. The opportunities for economies and increased efficiency are usually dependent on action in other urban sectors. The provision of new areas for low-income settlements for instance, raises problems of acquiring suitable sites, of segregation of rich and poor, of higher costs of transport with increased distance, of higher costs of some public utilities and reduced costs for others with lower densities, and the more general problem of the level of services to be provided. Levels of Urban Services The fundamental shortage of resources for urban infrastructure to equip the additions to population, let alone to ameliorate existing conditions, poses in acute form the quality of services to be provided. The possible extent and implications of lower levels of service as a "trade-off" to lower costs represent a critical element in determining the proportion of the population that can be provided with the serv- ices. Also of great importance is the relation between original capital cost and maintenance, and the degree to which variations in this rela- tionship affect the proportions of domestic labor and materials involved. 'The output from provision of dwellings stands in urgent need of a deeper analysis of direct and indirect impact on available national resources and foreign exchange. With the exception of one study in Japan which tends to confirm the much higher rating to be given house construction if the linkages are taken into account, and a more general analysis in the case of Colombia, the overall impact of housing does not appear to have been carefully investigated. Some consideration has already been given to such aspects in the case of housing. In the case of urban transport, restriction of private automobiles can be regarded as a choice between high levels of service for the few and adequate levels for the many. Levels of service in mass transit can also be widely varied including the degree of over- crowding and discomfort tolerated at peak hours. The choice of al- ternative transport modes besides influencing total costs also greatly influences the proportion to labor of capital costs and the foreign exchange component of both capital and operating expenses. For a single transport mode, for example a metro, there are trade-offs be- tween degrees of sophistication in equipment and speed, and be- tween machines and labor, for instance in fare collection. For buses, vehicle replacement and expenses of maintenance are directly re- lated. It is most unlikely that the best pattern for a developed country will be appropriate for a developing country, or that for a richer developing country appropriate for a poorer one with perhaps a fifth of the level of income. It should be possible to develop some general guidelines. The particular features of each individual city, especially the existing infrastructure, will nevertheless require significant varia- tions around any "norm". In other sectors, considerable opportunities may exist for lower cost services of minimum "utility" quality. Electricity distribution is often being designed at standards approximating those of developed countries. Simpler systems would increase the risks of loss of supply but could provide appreciable savings of costs and foreign exchange. In some non-industrial areas, the supply of electricity to more cus- tomers at lower cost with somewhat increased failure rate may well appear preferable to higher costs for a better quality of service to a smaller proportion of the inhabitants.1 Greater attention to measures to reduce peaks and to the relation between lower reserve capacity and the risks and costs involved also appears appropriate. Electric power customarily represents about one-sixth of total public invest- ment. A 10% saving, if attainable, would represent over half a billion dollars a year at projected rates of power investment in the develop- ing countries for 1971-75, or over a billion dollars a year by the 1980s. For water and sewerage, "trade-offs" of both convenience and health are involved. Undoubtedly reducing water pressure or filtering processes augments health hazards to the consumers affected; and 'See also Sector Working Paper on Electric Power. The comparison may in practice be between systems designed for lower service levels and costs, and systems de- signed for higher levels which because of overutilization (and poor maintenance) still incur high risks of outages. Out-of-date or otherwise inappropriate safety regula- tions can be an important factor in over-design. 451 in the former case may result in higher capital costs for individual storage tanks and deterioration of the transmission and distribution system. Poor intermittent service also creates difficulties in collecting or raising charges and in management. On the other hand, a larger proportion of the population supplied with lower quantities, rationed by restricted outlets or higher charges, may improve overall health and also reduce the risks of contagion spreading from previously un- serviced areas. In countries where expenditures on medications are a significant item in family budgets, charges at economic rates for good quality water supply may be lower than the savings on medicaments.' Sewerage presents other trade-offs. Depending on soil and topo- graphical features, lower densities of housing may permit cesspools for all or part of a development area. Evidence on economies through communal provision of water and sewerage facilities is contradictory. Individual supply appears generally to lead to much better main- tenance. But much depends on local community spirit and habits. Other possibilities of economizing resources arise from the fact that many water-using devices, such as toilet flush tanks, are not designed for high water economy, principally because water has traditionally been so cheap and charges inadequately related to volume used. Institutional, Planning and Pricing Problems The problems arising from the multiplicity of agencies dealing with urban growth, weak planning and management, inadequate policies on pricing, land regulations and taxes, and the scarcity of municipal finance are inextricably interwoven. As many as five levels of govern- ment-national, state, metropolitan, regional, municipal or local- may be involved with often little communication between them. More than 30 municipal authorities may partition a single metro- politan area as is the case in Calcutta. The boundaries of local govern- 'The wide range of possible combinations of quantity and quality of water is indi- cated in the range from a minimum of under 2 liters to over 200 liters per capita per day in usage. Investigations in East Africa indicate that communal standpipe supply typically works out at less than 20 liters per capita per day; for a single tap per family without indoor waste disposal at 30 to 40 liters; while for multiple tap and waste disposal the range mav be as wide as 60 to 600 liters with mean use of the order of 100 to 180 liters. costs of construction for piped water per capita may vary by a factor of 30. The variations in charges appear to be much less. Charges for carried water in areas without public facilities are generally much higher than for piped water. (They may absorb as much as 10% of family income compared with typically less than 2% for upper income classes with piped water.) A significant proportion of urban water may go to industrial and other nonconsumptive uses and could be re- cycled or returned for other uses if treated. In such conditions, simple projections of demand are likely to be an inadequate base for investment decisions. 2S ment jurisdictions are slow to respond to fundamental changes in urban configurations in all countries. The exceptional speed of large city growth in the developing countries poses particularly acute prob- lems of adaptation. The number of individual uncoordinated agencies under a single authority is no less a problem. Creation of autonomous agencies to ensure better management of individual services while improving some operations may, in the absence of a similar strength- ening of municipal administration, compound the problem of inte- grating policies and determining appropriate priorities of the urban center as a whole. Urban planning has a long history; but it is a history centered on physical and architectural planning. Only recently, even in the de- veloped countries, has the focus changed substantially to incorporate economic and social objectives and constraints, and inter-sector link- ages, in the context of dynamic growth. To date, no satisfactory com- prehensive models have emerged. Even the transport/spatial models, on which most work has been done, remain essentially static in con- cept with little in the way of feedbacks to incorporate the alternative impacts of the transport mode patterns tested. The more complex the models, the fewer the alternatives it has proved possible to con- sider. For various reasons, the alternatives have not been radically different in the sense of recognizing the potential for structural trans- formation, and consequently have produced remarkably uniform cost/benefit results. There is little evidence of their having had a sig- nificant influence on decision making. More generally, descriptive models based on observed relationships are of limited use for ana- lyzing changes in public policy designed to alter fundamental bases, such as price relationships in transportation and housing. In the developing countries, the lack of adequate attention to re- straints on resources and implementation capacity greatly reduces the utility of most of the "comprehensive" town plans. Lack of ex- plicit consideration of policies for pricing and land use, for example, and of specific short and intermediate run realistic programs, limit their effectiveness. There are additional reasons for the current dis- illusionment with comprehensive urban planning. Income redistri- bution effects are frequently mentioned but rarely studied in depth. Quantification of social indicators, moreover, does not solve the problem of choosing the parameters for defining the development sought. The planners have been largely isolated from the policy mak- ers whose decisions are involved. Moreover, inadequacy of data and the rapidity of change of the urban centers, and in their social struc- ture and economic functions, makes the determination of relation- ships, now and for the future, much more difficult than in developed AC2 countries. Already by the time comprehensive plans are produced, they tend to be sadly out-of-date. Similar difficulties are faced in the evaluation of individual urban projects. Methodology even in the developed countries is weak, par- ticularly in the evaluation of benefits. Because of the linkages with other sectors and the importance of social considerations, both the boundaries of relevant results and the weighting to be given them are difficult to decide. It is often difficult to determine unequivocally the relevant levels of demand when, as for many social services, the user may not be the purchaser in the sense of the taxpayer. In urban trans- port projects, differences in financial setup between municipal and various types of private or mixed public/private finance, makes com- parison of alternatives more difficult. More generally, techniques evolved for evaluating projects in the complex urban context tend to be weak in generating them. Nowhere are the shortcomings of existing urban development poli- cies and planning more evident than in the field of pricing in respect both to individual urban sectors and, perhaps even more, to their interactions. Failure to take into account such linkages often produces a conflict between the twin objectives of pricing policy-to secure a rational allocation of resources for urban development and to mo- bilize resources for this purpose. The appropriate pricing for one mode of urban transport is dependent on the pricing policy towards other modes. The price to be charged for providing a utility service to a specific locality, though it covers the costs, may not be appro- priate if it involves the costly provision of other facilities such as roads. More generally, a failure to take into account marginal costs rather than average costs in pricing policies is widely evident.1 As in the national context, there is an absence of clear guides to optimal allocation of resources and strong practical limitations to administrative regulation. The reduction of pricing anomalies so as to allow market mechanisms better to play a leading role in allo- cating resources accordingly assumes special importance. In this connection, taxation of benefits resulting from city improvement ex- penditures, and charges to obtain compensation for social disecono- mies attributable to private actions are of central concern in attaining a closer balance between private and social costs and benefits. Yet there has been remarkably little analysis of municipal taxation and its economic incidence. The great variety of tax and rate structures, 'Also of a general nature is the difficulty of establishing an appropriate "opportunity cost" for pricing capital for municipal undertakings when overall resources are so limited and when different types of projects have a widely differing impact on the tax base. 454 and the wide divergencies between theoretical incidence and actual collections, preclude a simple formulation of policy measures re- quired.' Analysis is needed on an individual city basis. If one field more than others should be singled out, it is probably the moderniz- ing of property taxes and their implementation to capture more of the gains resulting from public action and population growth. The need for greater coordination of urban development program- ming to take into account the intersector linkages and permit better overall establishment of priorities is now widely acknowledged. The prevalent ad hoc decisions at the time of municipal or agency budgets quite evidently do not lead to preferable growth patterns. The deci- sions can be demonstrably counterproductive, accentuating longer term problems and making their solution more difficult. There is much less agreement on the form or degree of centralization which should be attempted. A concensus seems apparent on the need to relate urban programming and planning much more closely with municipal budgets and the decision makers. General objectives, priorities which are necessarily politically determined, and financial allocations can then condition from the start the programming effort and make it more realistic. At the same time, however, there is increasing recog- nition of the vital need, if the urban problems are to be countered, of stimulating local effort and self-help which require rather small organizational units. It appears of little practical value to contend that the extensive interlinkages between sectors require a highly complex system of analysis and models projecting detailed future consequences; or that a high degree of coordination between all the agencies involved is a requisite for preparation and implementation of projects and pro- grams. To make achievement of such a situation a condition for in- troducing improvements would indefinitely delay effective action towards solving the mounting problems. The immediate need seems to be formulation of more general ob- jectives and priorities, based on a cross-sectoral examination of broad alternatives of future patterns of urban growth, resource limitations and feasible implementation. Within these broad limits, alternative packages of definite projects and policy proposals can be evaluated, using a restricted analysis encompassing major indirect as well as direct benefits and costs. Aspects that can be quantified in economic cost/benefit terms should be separated from others, such as the im- 'It is in the context of overall taxation incidence that remedial action on the details needs to be considered. Dependence on indirect taxation resulting in a regressive general tax structure may be partially offset by subsidies, for example to mass trans- port. At present such offsets are more likely to be unintentional than deliberate. 11 r; r; pact on income distribution. Both groups of considerations are rele- vant to decision taking. But it is a fallacy to imagine that they can be adequately combined in a single measure, at least at this time. Rough calculations of orders of magnitude of key variables can provide suf- ficient quantitative insight to narrow substantially the field of practical alternatives for more intensive study. The framework required for reasonable evaluation varies according to the type of project or program. Leading dynamic sectors, such as major urban transport projects, deserve particular attention and re- quire greater depth of analysis than others. Detailed consideration of overall urban growth patterns and social and economic structure is not, however, necessary for taking a reasonable decision on installa- tion of traffic lights. The framework needed for evaluating a new wholesale market may be quite limited if, as usually the case, the transport network indicates only a few possible focal points. A single "site and services" project, depending on site and relation to other areas, may not need to wait for an overall urban plan. The develop- ment of a limited area of urban extension, while requiring considera- tion of overall city growth patterns, may in many cases be sufficiently obvious as to warrant a fairly narrow framework of analysis if new basic decisions on alternative strategies of overall city development are not involved. In other words, it would seem that within such frameworks the emphasis should be primarily on an incremental approach, based on relatively short-run programs explicitly linked to individual projects, budgets and the political decision process, particular attention being given to the sectors primarily shaping the pattern of future urban growth. That some imbalances will result requiring correction is in- evitable. But it is difficult to perceive a better practical alternative to the present ad hoc urban decision making. Some imbalance is, in any case, a natural concomitant to dynamic growth and can provide the spur to further action. The experience gained should however pro- duce progressive refinement of techniques and also the flexibility in adjusting objectives necessary to cope with the rapidly changing con- ditions of the urban centers. A similar approach to institution building appears to be called for, centered on the links between political decision-making, municipal budgets and programming/project preparation activities. The trilogy of urban transport, housing, and land policy appear to merit particular attention due to their role in determining patterns of urban growth. The problem of coordination of the wide variety of public urban activities is one with which all major urban centers of the world are contending. It would be unrealistic to assume that it can be easily 4S6, and quickly solved in the developing countries. But an incremental approach based on a definite program for the more dynamic activities seems to offer the greatest hope of progress. This implies that the pattern and degree of coordination must be expected to differ con- siderably from one city to another; local traditions and norms, and the need to stimulate local initiative and self-help, should be taken carefully into account in evaluating feasible organizational improve- ments. THE BANK'S ROLE A large part of Bank lending has been urban-oriented. Most of the lending for power, water and telephones, as well as the lending for port facilities, urban expressways and industry has been for urban purposes. The major part of transport and telecommunications lend- ing has been to link urban areas. Bank lending in these various cate- gories which has been directly urban-related totals roughly $13 billion, or over three-quarters of total lending to the developing countries.' On a more restricted definition, excluding inter-urban transport and telecommunications, the proportion is still nearly half. Much of this lending has been concentrated on major cities. Over $2 billion, or more than 10% of all Bank lending, has been absorbed by 13 cities, each receiving $100 million or more. Projections for the next five years indicate no diminution in this effort. In its early years, limited resources, emphasis on projects of high priority in terms of output, and on strong borrowing organizations (strong both administratively and financially), restricted the types of the Bank's urban-oriented projects. Electricity supply has represented over a third of the total and inter-urban transport nearly 40%. In more recent years, the pattern has been changing. For example, for the 13 major cities, some 70% of the lending has been for electric power. This proportion declines to 40% for the projections for 1972- 76. The financing of water supply operations began in the early 1960s and though still relatively small is beginning to represent a substantial portion of Bank lending to a number of cities such as Bogota, Singa- pore, Caracas and the Moroccan coastal towns. In the mid-sixties, the first loans were made for road projects within metropolitan areas. These, however, have remained a small proportion of the total. The nature of past lending for urban purposes has had two conse- quences. The borrower has generally been an existing or newly set up 'Current borrowing countries, excluding Europe. Loans for services to both rural and urban areas have been split in proportion to the urban use. Annex 4 gives some details of the Bank's urban-oriented lending. A!7 national or autonomous agency, rather than a municipality or tradi- tional agency. Secondly, the limited scope of the Bank's contribution made it natural to concentrate on the particular field of activity of the project concerned. While it was recognized that the projects repre- sented only part of a wider complex of activities, it could generally be assumed that necessary complementary action would be under- taken by other parties. To a large extent this approach implied a fairly rigid and constrained view of the "externalities," or wider repurcus- sions, of the projects. Rate-setting and cash flow analysis could be considered largely in isolation from overall considerations of munici- pal finances and welfare. Experience in lending to local agencies indicates that this is gen- erally much more difficult than for national agencies. Local officials tend to be less qualified and local organizations less efficient and per- haps more often subject to political pressure. In such conditions, con- siderable staff time has been required to ensure that the funds lent are effectively used, that construction is carried out and subsequent oper- ations are conducted with a reasonable degree of efficiency. A large measure of institutional reform has been both a requirement and a result of some of the Bank's lending for municipal water systems. In some cases, suitable organizations simply do not exist and must be created before any operation is possible.' Rate problems, varying from one utility to another, have been involved in placing urban services on a financially viable basis. Resist- ance to rate increases in water supply appears greater than for telephones and power. Urban mass transit rates appear still more difficult to adjust. A subsidy to the poorer class of subscribers is com- mon in the case of power and water.2 Differentiation of this sort while maintaining an overall financially viable system seems likely to be much more difficult in mass transit. More generally, experience indi- cates that concentration on overall financial viability may not lead to sufficient attention to the structure of rates in the context of either social considerations or marginal costs. A limited number of cases have arisen where it has been considered acceptable practice to combine the earnings of several utility services 'This has been the case where the rational organization of a project involves a num- ber of independent jurisdictions, particularly where the interest of the parties in- volved are not identical. In the case of the water and sewerage system in Calcutta, for example, the Bank recommended formation of a Metropolitan Water Authority as necessary in order to devise and execute a rational plan for the area. 2However, the definition of subsidy may be based on an arbitrary allocation of com- mon costs, and charges may be little connected with volume, particularly at peak periods. The marginal costs of adding low income users may in such cases be below the charges made. 458; to achieve a satisfactory yield-when, if each were taken singly, all would not give a satisfactory yield. For a few projects, it has been con- sidered acceptable for a particular class of consumers to be subsidized out of general revenues paid by the state. Rural electrification and urban water supply to slum areas provide examples. The physical standards of Bank-financed projects have also raised problems. A change from local standards and local ways of doing things is generally involved, for example regarding road standards, use of equipment in construction, or use of foreign consultants. While the authorities of developing countries are sometimes more aware than foreign consultants of the need to tailor standards to local con- ditions, the reverse situation of local agencies favoring the highest standards in use in developed countries appears more general. The successful operation of projects outside of urban areas have occasionally been jeopardized, or at least limited in their benefits, by failure to develop complementary activity within the urban area. A few cases have arisen of city distribution systems not capable of handling the increased generation from a power project. Inter-urban highways or ports may sometimes have contributed to increased congestion in urban areas. Although the Bank's lending covers only a small fraction of total national and urban investment, experience indicates that it can have an influence out of proportion to its size. The progressive extension of Bank activities has accordingly been accompanied by more com- prehensive approaches to lending than in the Bank's early days. Work on country programming and individual economic sectors has already been considerably extended. With problems of urbanization becom- ing increasingly severe, attention is now being devoted to how the Bank's operations can be more consciously and effectively related to improving the efficiency of the urban centers both for production and for living. The preceding discussion has focussed on the problems involved in ameliorating urban conditions in the developing countries. The situation is far too complex, variations between countries and be- tween urban centers too great, information and methodology too poor, and institutional framework too weak to permit any simple de- termination of appropriate urbanization policies. To these limitations must be added a recognition that policy changes concerning major investment programs in developing, as in developed countries, are likely to be incremental in nature. That solutions will be difficult to achieve is obvious. On the other hand, the very weaknesses and often chaotic conditions of organization of urban growth, and the wide field for new types of policies, programming and projects, A n offer prospects of achieving substantial improvements even by meas- ures appreciably short of optimal. Many opportunities for Bank assistance are evident. To at least as great an extent as in other fields, the Bank can supplement the direct contribution of its lending to the total of resources by focussing at- tention on major issues and introducing new concepts, by suggesting ways to economize in resource use, by stimulating policies to reduce price distortions or improve controls and by assisting in the building of institutions. Before turning to consideration of future Bank activities in this field, some limitations with which the Bank is confronted deserve mention. The most obvious is shortage of experience and expertise on the urbanization problems of developing countries both within and outside the Bank. Only relatively recently have the acuteness and magnitude of the problems, deriving as they do from the post-war upsurge in populations and economic development, come to be rec- ognized. Other aid providers, like the Bank, have concentrated pri- marily on individual projects or sectors rather than a cross-sectoral approach combining both socio-economic and physical aspects. Few of the developing countries themselves have yet come to grips with the problems of formulating urbanization policies. Consultants experienced in the field are rare. When set against the number of the Bank's developing country members, and the number of their urban centers-more than 500 have populations over 100,000-it is obvious that the shortage of expertise available involves difficult choices in country and city selection. Shortage of expertise relates not only to formulation of general urban strategies but also to individual projects. The possible range of projects contributing to urban efficiency is very large. However for many types, including such important sectors as mass transport, the expertise available is very limited. Techniques so far used to ap- praise metros in developed countries do not, for instance, appear adequate to evaluate overall benefits and opportunity costs in cities of developing countries. The problem is less acute for types of projects in which the Bank already has considerable experience. Their influence on the pattern of urban development, however, differs widely. A wrong decision, wrong in the light of the priorities of an appropriately structured urban investment program, on a telephone or many other utility projects is unlikely seriously to affect the longer term pattern of urban growth. The project may be too soon, may not be of the highest priority, but it is likely to be beneficial and unlikely to prejudice de- sired future urban patterns. At the other extreme, an urban transport A £Ifl project, an expressway for example, will radically affect the future development of the city and may involve much higher total urban expenditure than other solutions. In between, there are many projects of sufficient size to raise the problem of the proportion of develop- ment funds preempted. A further limitation stems from the dearth of effective local urban agencies with which to collaborate and of projects adequately pre- pared for financing. The concentration of foreign aid on national or autonomous agencies has resulted in an absence of financial and techical assistance to major fields of urban investment. The develop- ment of municipal institutions and channels capable of dealing with the broad problems of urban areas and of preparing projects in this context has been slow. Judging by experience in other sectors, a combination of technical assistance with project lending may be the most efficient method of assisting in the critically important task of reinforcing, or at times creating, appropriate agencies. Shortage of municipal finance, exacerbated by inadequate tax structures, is a further serious limitation. Because of these limitations, the Bank's activities specifically di- rected to the complex of urbanization problems will be initially fo- cussed on a relatively small number of urban centers providing continuity of involvement over several years. In view of the inherent difficulties, the readiness of both national and urban authorities seri- ouslv to commit themselves to working out more rational urban policies and develop appropriate institutions must very strongly in- fluence the choice. To gain experience, the selection will, however, contain both large and smaller cities, and cover both relatively wealthy and poorer countries. It is recognized that such a policy of concentration entails risks of changes in local conditions involving substantial parts of the program. Shortage of expertise also indicates a need for selectivity in the types of project undertaken. Particular attention will be given to projects that exert a catalytic or dynamic influence on the pattern of urban growth through the project itself, or through the linkages with other sectors and the overall planning involved. Looked at from a different viewpoint, attention will be directed to projects which ob- viate, or substitute for, more expensive or lower priority alternatives. For similar reasons, projects will be linked to policies which affect their functioning in the wider context of urban development. The Bank's lack of experience and the general limitation of knowledge must be clearly acknowledged. With local conditions so important in determining appropriate and feasible policies, extensive dialogue with local authorities and experts, and the possibility of use of local consultants will necessarily be involved. It is also recognized that less ambitious projects undertaken with more restricted evaluation either singly or as part of a package may provide the basis for building up collaboration with municipal authorities while studies for longer term urban development are being undertaken. This approach can guard against the search for "ideal" projects precluding good projects in the interim. It is also proposed to undertake some projects of new types of special relevance to developing country urban conditions, such as site and services projects, on a wider geographical basis if particularly favorable opportunities for gaining experience present themselves and the possibility exists of repeat loans of a similar type. The Bank is now developing a program along the general lines indicated above. Preliminary economic reviews have been made of seven cities-Bombay (in some depth), Istanbul, Singapore, Bangkok, Kingston, Taipei and Djakarta. Specific urban project preparation has been undertaken also in six cities-Caracas, Istanbul, Dakar, Sao Paulo, Kuala Lumpur, Georgetown; and more general project iden- tification in four-Bombay, Kingston, Djakarta and Seoul. Urban transport has been the principal project field in three cities, overall urban development (including transport) in two. A "site and services" project has been appraised in one city. Technical assistance for trans- port and land use planning studies has been extended in seven cities, Singapore, Kuala Lumpur, Bangkok, Nairobi, Teheran, Amman and Bogota. Reviews have been made of experience in the urban field, particularly "site and service" schemes and urban land control measures. The Program of Urban Activities A program of project activities directed specifically to urbanization problems has been established for FY1972-76. It involves some 30 urban centers and around $700 million of Bank lending and is sum- marized in Table 3. The program is necessarily more tentative than for longer established sectors of Bank lending where close connec- tions have already been built up with borrowing agencies. Several of the projects are not yet specifically identified, being dependent on the results of initial studies. Some may be dropped and others substituted as part of this process and in order to bring the overall program more closely into line with the general approach to selection outlined above. The supporting study program is extensive. In four city areas- Bogota, San Jose, Klang Valley (Malaysia), and Singapore-the Bank has already agreed to be executing agent for the United Nations AC') Table 3: Urbanization Projects FY1972-76 Major Element of Project 1972 1973 1974 1975 1976 "Site and Services," Urban Extension Areas, etc. 1 2 7 7 12 Integrated Urban Region 1 4 4 Roads and Traffic Improvement 1 1 2 1 Mass Transit 1 2 3 2 Wholesale Markets 1 1 1* Other Urban Infrastructure and Land Development 1 1 1 1* Gross Total 3 6 12 16 21 Net Total Allowing for Eliminations and Slippage 2 4 8 10 14 *Not yet identified. Development Programme (UNDP) for studies of transport and urban pattern requirements designed to lead to identification of specific projects. Technical assistance will be provided to several other muni- cipal and central government agencies for similar studies, including Nairobi, Teheran, Bangkok and Amman. The preliminary phases of more general studies to assist urban and national authorities in de- fining appropriate priorities for urban projects now under way in the five countries indicated earlier will be extended and studies in some other cities are contemplated. In the virtual absence of "ready made" projects conforming to the requirements of an overall urban strategy and ready for appraisal, considerable Bank assistance of this type appears a prerequisite for building a pipeline of suitable proj- ects. Some of the projects identified are expected to be financed by other agencies. The program of urbanization projects is being coordinated with programs of projects in more traditional sectors of Bank lending. In Istanbul, for example, an overall program related to priority invest- ment needs has already been initiated, involving water supply and electricity distribution as well as urbanization projects including a wholesale market, land development and urban transport. Public utility projects are scheduled for almost all of the cities in the 1972-76 urbanization project program. The Bank is coordinating its work with other agencies. The general shortage of experience and the desirability of a package approach to maximize the impact of urban projects accentuate the need for such effort. The United Nations Centre for Housing, Building and Planning (UNCHBP) is performing most valuable work in, for instance, devising housing programs compatible with technical and financial resources and on the social organization of site and services schemes. Two joint A :-' programs with the UNCHBP are already under way.' Several other agencies are involved in specific areas of urban growth (See Annex 3). The Bank is assisting the UNCHBP in instituting regular inter-agencv exchanges of information on activities with a view to sharing infor- mation and completing necessary studies by the most appropriate agency, avoiding duplication of effort, and reducing the burden on developing country administrations resulting from numerous un- coordinated missions. An area of Latin America has been selected for trial exchange of information. In the field of water supply and sewerage, a joint program has been established with the World Health Organization, in education with Unesco. Formulation of National Urbanization Policies The Bank will take further steps to help developing countries focus attention on urbanization problems and alternative courses of action, including the balance between rural development and the develop- ment of urban centers of differing sizes. It is the governments of the developing countries who must assume the responsibility of solving their urbanization/unemployment problems. Without strong com- mitment on their part, the difficulties of surmounting the obstacles to a more rational approach can hardly be overcome. National development programs, budgets, and pricing and taxation policies largely determine the pattern of use of national development resources. The supplementary finance provided from the national budget to augment local resources, and the programs negotiated with foreign aid agencies, affect both the regional balance of resource allocation and the composition of investments. The incidence of national taxes, foreign exchange and interest rates, import and invest- ment permits, and price controls are of no less importance. The problems of urbanization have seldom been explicitly ex- amined in this context of total national resource allocation. Re- sponsibilities for preparation and implementation of development programs affecting the urban balance are usually split among many ministries and agencies. As a broad generalization, none has been ready to provide the leadership for a general policy in respect of ur- banization-or to concede it to others. There is, however, rapidly growing recognition that analysis of the problems of urbanization should be part of national development programming and policy- making. 'The world-wide review of site and services experience, and community organization of the Senegal site and services project. Attention to these basic problems will be progressively incorpo- rated in the Bank's system of economic missions, economic reports and country programs. It is believed that the focus provided for specific examination of the problems through discussions with na- tional authorities and local experts can serve to stimulate interest and action. It should also help in orienting the Bank's own program. Determination of precise objectives in the urbanization/unemploy- ment field may be a chimera, at this stage at least; growing awareness and progress towards new decision patterns are not. Technical Assistance in Urban Programming and Institution Building Where interest shown and readiness to take action are apparent, the Bank will consider in association with other agencies-bilateral or private, as well as international-the provision to planning agen- cies of technical assistance on urbanization problems. The Economic Development Institute (EDI) of the Bank will also provide help in overcoming one of the worst bottlenecks, the critical shortages of personnel in developing countries capable of performing the pro- gramming tasks relevant to the urbanization problems at both na- tional and municipal level. A first EDI course in this field is now under preparation. The improvement of management in municipal agencies, or "insti- tution building," is obviously crucial to the success of attempts to produce more efficient urban growth. In the Bank's public utility and other lending, an important element of technical assistance is often provided for this purpose. Technical assistance of at least comparable degree is envisaged in the new fields of urbanization project lending outlined below. Urban Transport The importance of mass transit to the movement of the labor force within the urban centers and its central role in determining urban growth patterns merit the inclusion of mass transit projects in the Bank's urbanization program. However, for such projects to contrib- ute fully to solution of the urbanization problems, evaluation must be in the context of the urban transport system as a whole, and the impact on urban patterns and different income groups. Policies to- wards restraint on private automobile use have particular relevance to evaluation of the projects (see pages 436-441). An extension of Bank lending to bus systems as the main form of mass transit is contemplated. The studies of transport systems noted on page 462 are expected to lead to between five and ten projects in- corporating improvementof bus systems within the FY1972-76 period. Terminals and maintenance facilities may be included as well as buses. Weaknesses in present bus systems are a compound of inadequate equipment, poor administration (including lack of coordination be- tween different bus entities) and inappropriate pricing policies. All three elements will necessarily be involved in project preparation and appraisal. Provision of bus equipment under suppliers' credit has often con- tributed to a less than thorough analysis of operations and the needs for complementary investments and policies. Differences in terms of the credits, rather than suitability and performance, have often been deciding factors in the choice, with little relation to policies affecting transport demand or overall transport programming. Incorporation of suppliers' credit within an overall bus system improvement project may be possible. The possibility also exists of lending through the in- termediary of an urban bus agency or metropolitan development agency, particularly when several independent bus operators are involved. Urban railways, particularly the improvement of facilities which already exist, are also expected to form part of the urbanization pro- gram as it develops, as may light railways or tram systems with rights- of-way. As in the case of metros, improved methods for recovering some of the benefits reflected in increased urban property values are likely to be important to financial viability. Urban Roads. Highways and streets carry virtually all intra-urban freight and most of the passenger traffic. While the development po- tential of roads between urban centers has received strong attention by the Bank in the past, little attention has been devoted to the cor- responding intra-urban road development potential. As noted earlier, populations and areas of many of the cities are tripling within 20 years. Heavy investments in urban roads cannot be avoided. Good intra- urban distribution networks are required if the full benefits of intra- urban road projects are to be realized. Similar considerations apply to extension of port facilities. The discussion on Urban Transport and Urban Patterns (pages 435, 442) indicates that well conceived road projects can contribute greatly to the efficiency of urban centers and promote patterns of urban growth which economize in use of total resources. But an ad hoc ap- proach to relieving congestion may be not only expensive and ineffec- tive but also compound future problems. The Bank's approach to 466 lending for urban roads will accordingly require particular attention to the influence of the roads on alternative patterns of urban growth, to the main beneficiaries, to interdependencies with mass transit, and to policies towards restraining the use of private automobiles and guiding land use. Several urban road projects are included in the program of urbani- zation projects for FY1972-76, either as the main element or as part of urban extension areas or the wider urban region projects. In one case, the project would provide a diversion route for a rapidly ex- panding port, in another an improved link between a major city and a satellite town as part of a wider urban region development. Two other "urban corridor" development projects are contemplated which are likely to include road construction. Almost every city investigated could benefit greatly from compara- tively inexpensive traffic engineering and controls. Capacity of exist- ing streets could in many cases quickly be increased by 50 to 100% by installation of modern coordinated signal control systems, inter- section redesign and other control measures. Besides providing al- most immediate improvements in traffic flows and reductions in user costs, such measures can in some cases postpone or even eliminate the need for other much more costly projects. In some cases, traffic movements on an entire central street system might be substantially improved for less than the cost of building a mile of urban expressway. In other cases, short links between existing arteries could substantially reduce congestion by increasing effective capacity of existing roads or obviating the need to enter congested central districts. One project involving traffic signals and other improvements is included in the program and another in this general category may be added. While the general considerations indicated above are rele- vant, the inter-linkages with mass transit improvements are likely to be particularly important. Projects may include technical assistance and training in traffic control. Terminals. Freight terminals which economize on sorting and con- solidation of goods shipments are of particular importance to the efficiency of urban areas. The number of trucks carrying goods on city streets may be reduced to a small fraction of their previous number by the construction of freight terminals at appropriate sites. At pres- ent, most cities in developing countries are suffering from inadequate and poorly sited terminals as a result of their rapid expansion. Besides improving the efficiency of freight handling and reducing costs, in- cluding those of distribution, new freight terminals can result in sub- stantial benefits by redLicing traffic congestion. Two projects are under study for wholesale markets in cities where 467 facilities are quite inadequate. Some retail distribution facilities and housing (above the commercial facilities) are likely also to be in- volved. One project has been identified in which the major element would be an inter-modal passenger terminal serving railways, buses and taxis. It would permit transfers between regional and urban- suburban rail and bus lines and could be instrumental in achieving effective coordination among these services and eliminate much con- gestion. Because of its inter-modal character, construction is outside the range of any single-mode transport agency. Housing and Urban Works Projects The problem of housing in urban centers is distinguished from other urban problems not least by the magnitudes of the needs and investment involved. Bank projects cannot realistically be expected to make a major direct contribution to the resources required. More- over, the resources required are primarily of a local nature, and hous- ing does not yield foreign income. For most developing countries, to incur foreign debt to cover more than a small operational housing expenditure would rapidly result in serious foreign debt servicing problems. It is recognized, however, that less direct methods to stimulate housing construction, particularly for low-income groups where the need is greatest, can provide a very important contribution not only to social well-being but also to the growth of national output and employment. Moreover, the siting of new housing and its relation to the siting of employment can greatly facilitate urban patterns which economize in total resource use. These considerations, more fully de- veloped on pages 445-450, point to Bank lending for relatively small projects in the housing field which can have a catalytic impact on local programs, particularly in their influence on institution building. In line with these considerations, the urban project program in this field is concentrated on "site and services" and similar projects to provide urbanized land on which the occupants can build their own dwellings using self-help methods. One project is already being ap- praised in Senegal and several others are envisaged. The first project is illustrative. The land provided by the Government will be equipped with roads, water, sewage and power facilities. Most water and sew- age facilities will be provided on a community basis. Education and health facilities are included in the project as well as sites for indus- trial, commercial, recreational and other developments. Exclusive of the social facilities and power grid, charges to occupants are expected McR to cover costs which will be only a small fraction of those of minimal housing schemes in the city.' A new institutional organization will be built up with technical assistance to form the basis of a national pro- gram on similar lines. The emphasis on use of local labor and employ- ment will result in a somewhat higher proportion of local costs than has been customary in Bank lending. As noted on page 445, public schemes for minimum housing, at least in the poorer countries, do not effectively reach the poorer groups forming a majority of the population even though they gen- erally involve heavy subsidies either directly or indirectly through low interest rates. As other agencies have found, there are also consider- able difficulties of management. The possibility is not excluded at a later stage when more experience has been gained, of direct financing of permanent housing where the particular circumstances warrant for example a pilot scheme. Such housing may also form a limited integral part of a wider urban project. Generally, however, direct financing of housing is not contemplated at this time. Indirect lending for housing through financing institutions is considered below. One project has so far been identified for improvement of low- income settlements. Generalizations on the appropriate composition of such programs is difficult. Provision of minimum services of water, drainage, sewage (often cesspits), streets and street lighting, construc- tion materials, schools and vocational training centers, may all con- stitute elements. As for site and services projects, security of tenure is an essential factor. Some elements may be provided directly out of public finances and be revenue earning, for example water supply. Others can be provided on a loan basis, for example construction ma- terials. But in general, the benefits whether the obvious ones of im- proved living conditions and stimulation of self-help output and savings, or the more indirect ones of increased economic value of land, will not be directly revenue producing. The local element of expenditure will be high. However, a few Bank projects on a pilot basis appear justified in view of the low costs involved in relation to overall impact. Bank assistance for such programs or portions of them can be linked to action to improve the position of municipal finances through administrative and land tax reforms which are often involved, and be directed also to securing economies of overall urban develop- ment. The rapidity of extension of urban areas offers great opportunities for economizing in resource use by more rational development of 'In some cases, public subsidies may be justified, but at much lower levels than those customarily involved in public housing schemes. new areas. Integrated extension areas also offer the possibility of cap- turing a large part of the increase in land values produced by public action and growth of population. The rationale for such projects has already been outlined in greater detail on page 444. Bank assistance can play an important role in stimulating well organized and managed schemes. Preliminary investigations are proceeding for identifying two or three such projects for inclusion in the lending program. "Site and services" type schemes may form a part of such projects. While direct lending for housing construction is not at present con- templated, indirect Bank assistance to housing through the stimula- tion of housing finance institutions more clearly fits the general con- siderations outlined on page 468. The non-availability of adequate housing is a serious cause of discontent among skilled labor and white-collar groups who must frequently live far from their jobs or accept high rents and poor housing closer in. Immobility due to short- age of housing and poor locations may greatly increase transport de- mands. Absence of financing institutions to collect savings and extend loans for housing for these and for the lower income groups often presents a major bottleneck. Experience shows that the volume of new savings collected by such institutions is a multiple of the amount of "seed capital" initially needed to start them. The International Finance Corporation is prepared to consider "seed capital" for independent housing finance associations and sim- ilar institutions providing mortgage facilities where such projects form part of a program to stimulate savings and develop the capital market, and where subsidization of housing for relatively richer groups of the population is not involved. Discussions are at present being held with three countries. The Bank can also lend indirectly for similar purposes through local development corporations. Particular attention will be paid to the income groups for which housing is proposed, the mobili- zation of small savings and the possibilities for supporting improve- ments to substandard housing. Urban Region Projects Evaluation of many of the projects discussed above will necessarily involve consideration of a region considerably wider than the munici- pal limits. The interactions between growth in cities and that in sur- rounding areas extending 20 miles or more have a direct relevance to projects such as major urban roadways. Similarly the growth of satellite towns has repercussions on the major city with which they are intimately connected. The program contains three or four projects 470 which are concerned with a package approach to the development of such urban regions or corridors. The elements of the package, how- ever, are likely to fall under the previous headings. The Design of Urban Projects The Bank is already involved in detailed appraisal of standards of construction and services and frequently succeeds in introducing changes in project design which cut costs without endangering effi- ciency.' Further emphasis will now be given to opportunities for low-cost urbanization services of minimum "utility standards." Bank projects should occupy a leading position in this field. Urbanization projects such as "site and services" and urban extension areas will provide evaluations of alternative basic service systems that can have much wider applicability for Bank projects in urban areas. The relation of capital costs, particularly the foreign exchange ele- ment, to local costs and employment will be given increased atten- tion. Projects which would reduce current urban public expenditures will also merit particular consideration. Projects which would cut current costs of education per student through new techniques of teaching and capital costs through more intensive use of existing buildings are of particular moment in view of the large and rising share of education in total public expenditures and the opportunities provided by urban concentrations. The difficulties of achieving such economies should not be under- estimated. The initial definition of public utility projects by municipal agencies, and the stress laid on the maintenance of high standards by the authorities concerned, national leaders, consultants and others, often makes the suggestion of alternative lower-cost systems and of pricing policies-for example to restrict peak demand-a highly sen- sitive issue. Many normative judgments may be involved both related to the existing socio-political structures and to the balance of near- term to long-term benefits. In these conditions, close working association and discussion of alternatives over a series of projects may provide a more fruitful influence than an attempt to impose conditions at an early stage too greatly at variance with what is locally acceptable at that time. How- ever, appropriate clauses will be included where possible in terms of reference of consultants to cover evaluation of alternative lower levels of services and the risks of failure of supply and maintenance costs involved. More intensive consideration of the flexibility of demand in 'See for example the Sector Working Paper on Water Supply and Sewerage. relation to costs, and feasible peak-pricing policies, may also serve to ensure adequate consideration of the wider issues. Sector review missions and reports will provide a particularly opportune occasion for discussion of such issues and the position of the sector in the general urban development policy. Research Research of operational significance is required in many fields. Given the limitations of staff, the Bank's contribution can comprise only a very small portion of the total. The objective accordingly is a blend comprising analysis of project experience, an "intelligence" function designed to keep abreast of research being conducted else- where and limited original research to fill gaps not covered by others when needed to support project activity. Collaboration with other agencies is particularly important for economizing in staff and avoid- ing dissipation and duplication of effort. The following paragraphs outline areas of particular concern rather than attempting to delineate at this time those aspects which should be chosen for a direct research effort by the Bank. Strong emphasis is placed on the highly important but neglected field of analysis of municipal financing and tax systems in relation to development. The analysis should cover existing sources and expendi- ture patterns, elasticities of revenues to income variation and to rate increases, the regressivity or progressivity of tax structure, the influ- ence of taxes on location and land values, the scope for capturing rent and betterment values, the possibilities for financing public involve- ment in the land market, relations with state and national revenues, and the impact of alternative development patterns. Municipal ex- penditure patterns have also been remarkably little investigated. Both fields are of cardinal importance in assessing alternative strategies for tackling urbanization problems or taking advantage of opportunities for economies. The Bank has already initiated some studies of the impact of provision of utilities on land values. Many difficulties will be faced by governments in moving from a growing awareness of problems towards new decision patterns until more is known about urban and inter-urban structure and economics. A considerable amount of work, most of it of quite recent origin, is being conducted to develop satisfactory analytical frameworks in various institutions. As in the field of national and sectoral economics, a considerable effort will be required within the Bank to keep abreast of developments elsewhere and, in conjunction with work on eco- nomic missions and country programs, to complement this with ap- propriate adaptations or innovations. Closely allied to both the above subjects is the analysis of methods for more closely equating private and social costs. Systematic collec- tion of data is a prerequisite for such studies, and the Bank has under way one study to provide information on the extent to which appro- priate data for these purposes are available. Important questions for such analysis include the ways and costs of providing tax and credit in- centives for smaller growth centers and of improving the sectoral and spatial allocation of investment packages within metropolitan areas. Private automobiles represent a particularly acute case of diver- gencies between social and private costs and benefits. The Bank has begun a general study of this. The first phase is designed to collate existing studies and information on experience in the control of urban automobile use. The second phase is intended to be a cooperative ef- fort in which other agencies as well as the Bank will review in greater depth the problems and policy alternatives for developing countries. A study in greater detail of vehicle user costs and charges has been made in Central America' and another is under way in Venezuela. Also in the field of transport, continuing research is required on mass transit, urban roads and the location and design of ports and other inter-modal terminal facilities, including wholesale markets. The field includes least-cost means of reducing average commuting times, feasible means of financing mass transit systems and the effects of large existing ports and other major terminal facilities within major cities. Site and services projects, both as individual programs and in the wider context of integrated urban extension areas represent a further field for cooperative effort including analyses of employment and savings aspects. A first joint review of experience by the Bank and the UNCHBP is already in progress. A considerably wider effort is needed, as is research into certain other aspects of urban housing such as the extent to which costs vary with living space, and quality and "com- pleteness" of construction. Urban land and land policies have been identified as of central im- portance in the urbanization process. The UNCHBP has already com- missioned and conducted substantial investigations and intends to continue its work in this field. In addition, some further research may be necessary in order to provide a basis for judgments on allocation of funds to infrastructure services for improvement of low-income settlements, urban renewal or urban land extension. "'A Study of Road User Charges in Central America" by A. Churchill and others. World Bank Occasional Paper No. 15. The Bank can help assure adequate attention to the stimulation and assessment of new techniques. The construction and construction materials industries appear particularly in need of a concerted effort to develop new materials and evolve new techniques to improve pro- ductivity in developing countries. The Bank might aid directly in the assembly and comparison of data on incremental costs and benefits in urban development. Such information would be particularly valu- able for judgment on alternative strategies and appraisal of specific projects. Much information is likely to be available in consultants' reports and municipal files as well as in individual studies. An appre- ciable effort of interpretation and analysis is required, for example of the overall differential in urban costs between different densities of housing, and the identification of the more variable elements of distribution of urban services. In a few fields, the Bank will itself un- dertake more intensive research on the relation between levels of service and their costs and benefits. The development of urban employment analysis through exam- ination of interdependencies among economic activities has great importance for the definition of appropriate urban policies and proj- ects. Considerable gaps exist in knowledge of the operation, finance and distribution of earnings of small-scale activities, even though they affect a large proportion of the urban population. The indirect as well as direct employment effects of housing stand in need of greater anal- ysis. The effects of legislative tax measures and legal or customary work restrictions (e.g. on double-shift working) on urbanization prob- lems present a further field for study. It is a paradox that use of ca- pacity of expensive investments often appears lower in the cities of developing countries than in developed ones. Inevitably, the return on Bank investments and feasible national urbanization/employment policies are affected. Much more information is also desirable on the pattern and causes of migration, and the experience of new migrants. 474 Annex 1 BACKGROUND DATA ON URBANIZATION Chart 1: Degree of Urbanization Compared with GNP Per Capita 100 40n . _ _ _ OUSA 20 _ _ _1, 4 ! *1 v | rahilip *Chil. BuH 6C l e - 'i - Vi_tn_m StSlwwna 10~~~~~~~~~ _rzi S ETC,ochosloe*' okja I Chi.. g Gt Cyprun 0 _hn Burudiy||ia 0 l SP IIaIIaII IIa aGNgPpt S S Pordaner* Capita rfactrtos Ecuador 0 Siamaica d .d Tb.g 1o Coasercapita-World Bopr17 2. Un I Ion- l 1969 Deiiino ubn"i ae ndferin binl tn ards 13 Cuurv fitte to tye t Haiti P C i ( t ot 9 U 1. GP pe capta-Wrld ank tlas 19T1. d 3.Prcn of pouato in 0 20muite 4fmr ta 00 , 000 or abl 1,0in mai ,tstar frm 0e00api Yearbook:r 1910,a NWYorlBakAta, 1971. Tbe9 4. Figures based on most recent census or estimated from sample surveys ranging from 1960 to 1970. (1) Curve fitted to type y a ecloGP 475 Annex 1 Table 1: Past and Projected Total and Urban Population of the Developing Nations of the World Bank 1920-2000 (1) Total regional population of Bank's developing nations (millions).* (2) Total urban (20,000 and over) population (millions).* (3) Annual rate of total population growth (%) for 20 years preceding. (4) Annual rate of urban population growth (%) for 20 years preceding. Developing Region 1920 1940 1960 1980 2000 East Asia and (1) 131.5 182.8 268.7 450.3 730.6 Pacific(l (2) 5.7 15.3 41.3 104.6 232.7 (3) - 1.6 1.9 2.6 2.4 (4) - 5.1 5.1 4.8 4.1 South Asia(' (1) 333.1 422.2 579.9 941.7 1,381.2 (2) 18.2 34.6 76.6 168.9 343.9 (3) - 1.2 1.6 2.5 1.9 (4) - 3.3 4.0 4.0 3.6 Southern Europe (1) 111.5 140.0 176.6 234.8 314.0 and the Middle (2) 24.5 36.2 59.4 91.1 141.1 East(2) (3) - 1.1 1.1 1.4 1.5 (4) - 2.0 2.5 2.2 2.2 Africa(3) (1) 135.1 179.8 255.4 419.3 716.9 (2) 5.7 11.2 30.8 76.7 189.5 (3) - 1.2 1.4 2.5 2.7 (4) - 3.4 5.2 4.7 4.6 Latin America (1) 89.5 129.9 212.4 378.4 638.1 (2) 12.9 25.5 69.7 163.4 342.0 (3) - 1.9 2.5 2.9 2.6 (4) - 3.5 5.2 4.4 3.8 Totals for all (1) 800.7 1,054.7 1,493.0 2,424.5 3,780.8 countries (2) 67.0 122.8 277.8 604.7 1,249.2 above (3) - 1.4 1.8 2.5 2.2 (4) - 3.1 4.2 4.0 3.7 'Source: UN Population Studies, No. 44, Growth of the World's Urban and Rural Population, 1920-2000. United Nations, New York, 1969. (ST/SOA/Series A/44.) Regional totals have been combined where possible to approximate World Bank groupings of developing regions. (°) Burma is included in the totals tor East Asia and the Pacific and not South Asia. North Korea and North Vietnam (not Bank members) could not readily be separated and are included in East Asia; they do not significantly affect the trends shown. (°) Including Italy and Portugal which could not readily be separated. In mid-1969, the combined populations of these "developed" countries was 62.7 million (World Bank Atlas 1971t or approximately 31% of the regional population at that time. ( Excludes South Africa. 476 Annex 1 Table 2: Growth of City Population (100,000 and over) in Developing Nations of the World Bank 1920-1960 (1) Total city population (millions). (2) Annual % growth rate over preceding 20 years. Developing Region 1920 1940 1960 East Asia and (1) 3.4 10.3 30.7 the Pacific(1) (2) - 10.1 9.9 South Asia (11 (1) 8.1 17.7 45.8 (2) 5.9 7.9 Southern Europe (1) 11.5 18.9 34.9 and the Middle EaSt(2) (2) - 3.2 4.2 Af rica(3) (1) 2.4 5.7 19.4 (2) - 6.9 12.0 Latin America (1) 4.9 11.1 35.6 (2) - 6.3 11.0 Total of all (1) 30.3 63.7 166.4 countries above (2) -. 5.5 7.3 Sources and Notes: See Table 1. 477 Table 3: Growth in Total Urban and Principal City Population 1960-1970 in Bank Developing Countries Principal City Annual Annual Annual Population as Total Growth Urban Growth Principal Growth % of Urban Country Population Rate Population Rate % Urban City Population Rate Population 1970 1960-70(') 1970 1960-70(1) 1960 1970(l) 1960-70(l) (thousand) (%) (thousand) (thousand) (%) A. Eastern Africa Botswana 626 3.0 140 3.3 20 22 Gaberones _(2) _ Burundi 3544 2.0 93 7.2 2 3 Bujumbura - East African Community 32655 3.0 2661 7.6 6 8 Kenya 10861 3.0 1000 5.1 8 9 Nairobi 500 6.0 50 Tanzania 13252 3.4 1049 8.6 5 7 Dar es Salaam(3) 350 9.0 33 Uganda 8542 2.5 612 9.2 4 7 Kampala 325 9.2 53 Ethiopia 24754 1.9 1643 2.4 6 7 Addis Ababa(3) 631 3.9 38 Lesotho 946 2.9 16 7.9 1.0 2 Maseru - - - Malagasy 6864 2.4 891 4.3 11 13 Tananarive(3) 432 5.6 48 Malawi 4429 2.5 240 10.1 3 5 Blantyre Limbe(3) 174 - 73 4 Mauritius 857 2.7 469 5.3 42 55 Port Louis 146 - 31 Rwanda 3624 3.2 10 8.3 .2 .3 Kigali - - Somalia 2941 3.9 713 6.0 20 24 Mogadiscio 200 - 28 Sudan 15631 2.9 1303 4.3 7 8 Khartoum 452 3.4 34 Swaziland 419 4.9 18 9.4 3 4 Mbabane - - - Zaire 17405 2.1 3000 3.9 15 17 Kinshasa(3) 580 10.0 19 Zambia 4326 3.1 1137 6.4 19 26 Lusaka 225 - 20 B. Western Africa Cameroon 5839 2.2 748 6.8 8 13 Douala 229 4.0 31 Central African R. 1585 2.6 1207 8.4 14 24 Bangui 180 - 15 Chad 3562 1.6 275 7.9 4 8 Fort Lamy - - - Congo Peop. R. 895 1.5 345 5.4 26 39 Brazzaville 190 4.0 Dahomey 2732 2.9 438 6.0 12 16 Cotonou(3) 175 - Gabon 485 1.0 101 6.5 12 21 Libreville - Gambia 364 2.0 37 3.8 9 10 Bathurst - - - Ghana 8808 2.8 2990 6.8 23 34 Accra 752 6.8 25 Guinea 4011 2.7 451 6.6 - 11 Conakry 220 7.0 49 Ivory Coast 4289 2.9 1244 9.3 16 29 Abidjan 550 11.0 Liberia 1168 1.7 344 10.4 13 30 Monrovia - - - Mali 5010 2.0 439 5.7 6 9 Bamako 225 5.6 Mauritania 1161 1.9 28 2.4 2 2 Nouakchott - - - Niger 3845 3.0 118 2.4 3 3 Niamey - Nigeria 66000 3.9 13826 6.0 17 21 Lagos(3) 800 - 6 Senegal 3958 2.5 1081 4.0 24 27 Dakar 600 6.0 52 Sierra Leone 2500 1.4 360 3.8 11 14 Freetown(3) 175 - 49 Togo 1851 2.5 305 8.2 10 17 Lome 180 - 59 Upper Volta 5363 2.2 253 5.1 4 5 Ouagadougou - - C. East Asia and Pacific British Solomon Islands 153 2.1 6 4.7 3 4 China (Taiwan) 14402 3.0 9272 4.1 59 64 Taipei 2150 4.7 23 Fiji 540 3.2 106 3.9 18 20 - - - - Indonesia 118184 2.3 21136 4.2 15 18 Djakarta(3J 4500 4.7 21 Khmer R. 6850 2.3 878 4.8 10 13 PhnomPenh 700 7.6 80 Korea, South 32168 2.6 12547 4.1 28 39 Seoul(3) 4661 8.0 37 Malaysia 9376 3.1 4291 5.9 35 46 Kuala Lumpur(3) 650 4.0 15 Papua and New Guinea 2348 1.2 29 7.3 7 1.2 - - - - Philippines 38290 3.5 8867 4.3 22 23 Manila 4100 4.3 46 Singapore 2113 2.6 2113 2.6 100 100 Singapore 2113 2.6 100 Thailand 31239 3.0 4659 4.5 11 13 Bangkok 2100 6.0 45 D. South Asia Afghanistan 16715 1.8 1128 2.3 6 7 Kabul 500 3.2 44 Ceylon 12685 2.5 2100 3.6 15 17 Colombo 1250 2.8 60 India 544621 2.4 102433 2.9 18 19 Calcutta 5153 1.8 5 Iran 28805 3.0 11306 4.8 33 39 Teheran 3250 6 29 Nepal 11143 2.0 552 8.0 3 5 Katmandu 440 7.9 80 Pakistan 114191 2.1 18178 4.3 13 16 Karachi 3246 5.6 18 E. Europe, Middle East and North Africa Cyprus 631 .9 302 3.9 36 48 Nicosia 112 - 37 Egypt, Arab R. 33283 2.4 14544 4.0 38 44 Cairo 5600 4.1 38 XI Finland 4768 .7 3259 2.7 56 68 Helsinki 757 3.2 23 , Table 3: Growth in Total Urban and Principal City Population 1960-1970 in Bank Developing Countries (continued) Principal City Annual Annual Annual Population as Total Growth Urban Growth Principal Growth % of Urban Country Population Rate Population Rate % Urban City Population Rate Population 1970 1960-70') 1970 1960-70)') 1960 1970(a) 1960-700) (thousand) (%) (thousand) (thousand) (%) Greece 8876 .7 5552 1.6 57 63 Athens 2425 3.0 44 Iceland 212 1.8 152 2.7 67 72 Reykjavik - - - Iraq 9089 2.9 3954 4.0 39 44 Baghdad 1250 4.4 32 Israel 2908 3.2 2365 3.7 77 81 Tel-Aviv 900 3.0 - Jordan 2421 3.6 1064 4.8 39 44 Amman(3) 350 4.5 33 Lebanon 2600 2.1 1034 4.0 33 40 Beirut 600 3.4 58 Morocco 15519 2.9 5484 4.9 29 35 Casablanca 1462 4.2 27 Portugal 9723 .9 3549 1.7 34 37 Lisbon 1500 1.2 42 c Southern Yemen 1000 2.6 325 4.2 28 33 '3 Spain 32958 .8 19369 1.7 44 59 Madrid 2990 2.8 15 Syria 6101 2.9 2351 3.4 37 39 Damascus(31 670 2.4 28 Tunisia 4882 1.6 2118 3.2 37 43 Tunis(3a 700 .8 33 Turkey 35225 2.5 11007 4.2 27 31 Istanbul 2600 5.0 24 Yemen Arab Republic 5000 - 301 5.9 3 6 - Yugoslavia 20649 1.2 7986 4.8 27 39 Belgrade 1165 3.7 15 F. Central America and Caribbean Costa Rica 1767 4.2 640 4.6 35 36 San Jose 435 5.4 68 Dominican Republic 4324 3.6 1601 5.7 31 37 Santo Domingo)3a 650 5.9 41 El Salvador 3499 3.6 1392 4.0 38 39 San Salvador 375 4.6 27 Guatemala 5172 3.1 1892 4.9 31 37 Guatemala City 770 5.0 41 Guyana 737 2.8 221 3.2 29 30 Georgetown 200 3.1 41 Honduras 2703 3.4 700 5.2 22 26 Tegucigalpa 281 5.9 40 Jamaica 2020 2.2 727 4.3 29 36 Kingston 560 5.0 77 Mexico 50624 3.8 29468 5.2 51 58 Mexico City(3a 3541 5.0 12 Nicaragua 1989 3.5 869 4.6 39 44 Managua 350 5.9 40 Panama 1465 3.1 685 4.4 41 47 Panama 440 4.9 64 Trinidad & Tobago 1106 2.9 589 6.0 40 53 Port of Spain - - - Venezuela 10390 3.5 7934 5.6 62 76 Caracas 2147 5.5 27 G. South America Argentina 24089 1.9 16978 2.4 57 71 Buenos Aires 9410 3.0 56 Bolivia 3956 1.4 1002 2.4 23 25 La Paz 500 2.3 50 Brazil 93545 2.8 50025 4.6 45 54 Sao Paulo 8405 6.4 17 Chile 9510 2.0 7007 3.4 65 74 Santiago 2600 3.1 37 Colombia 21168 3.2 11648 5.0 46 55 Bogota 2500 7.0 22 Ecuador 6089 3.4 2238 4.7 33 38 Guayaquil(3) 800 5.9 36 Paraguay 2378 3.1 852 3.5 34 36 Asuncion 445 3.6 52 Peru 13581 2.1 6256 3.3 41 46 Lima 2500 5.1 40 Uruguay 2889 1.3 2433 2.9 72 84 Montevideo(3) 1530 4.7 63 °) Source: Kingsley Davis, World Urbanization 1950-1970, Vol. 1: Basic Data for Cities, Countries and Regions, University ot California. Berkeley 1969. The 1970 figures are projections. The annual growth rates for these principal cities where recent data is available have been adjusted. (2) When no population figure is given for principal city, the population is under 100.000. - ( () Figures refer to the city area as opposed to the metropolitan area. (I I-. Annex 1 Table 4: Estimates of Migrants as a Percentage of Recent Population Increases Total Population Migrants as a Percentage Increase of Total Population City Period (Thousands) Increase Abidjan 1955-63 129 76 Bogota 1956-66 930 33 Bombay 1951-61 1,207 52 Caracas 1950-60 587 54 1960-66 501 50 Djakarta 1961-68 1,528 59 Istanbul 1950-60 672 68 1960-65 428 65 Lagos 1952-62 393 75 Nairobi 1961-69 162 50 Sao Paulo 1950-60 2,163 72 1960-67 2,543 68 Seoul 1955-65 1,697 63 Taipei 1950-60 396 40 1960-67 326 43 Sources: Etude Soeio-Economi,que doe a Zone Urhaine D'Abidjan, Societe d'Economie et de Mathematique Appliquees, Paris, 1967. Bogota rsansport and Urban Development Study. Phase I (draft report), Vol. 1, Freeman, Fox, Wilbur Smith and Assoc. Restrepo y Uribe Ltda., Bogota, 1970. Report on the Development Plan tor Greater Bombay 1564. Municipal Corp. of Greater Bombay, Bombay, 1964. Ca,acas 1500. Plan da Desarrollo Urbano. I Etapa Del Estudio, Consejo Municipal del Distrito Federal Oficina Municipal de Planeamiento Urbana, Caracas, 1968. R. Oliver, "Greater Djakarta, The Capital City of Indonesia", Washington, D.C. 1970. "Istanbul Metropolitan Alan Planlama Calasmalari", Mimarlik Special Issue, May 1970, 55-77. Urban Government for Metropolitan Lagos. B. A. Williams and A. H. Walsh, Praeger, N.Y. 1968. "Population, Employment and Earnings," (draft report), Nairobi Urban Study Group, 1971. Caractarizacco Preliminar Da Regiso Oa Grande Sao Paulo, Gegran, Sao Paulo revised edition 1969. Yoon, Jong-Joo, "A Study on the Migration Magnitude of Seoul," Journalof Population Stioies. No.2, Seoul, 1966. Taipei City Statistical data, Taipei Municipal Government, Taipei, 1950, 1960. 1967. 482 Table 5: Percentage of Urban Population in 1970 of 138 Countries Grouped by GNP per Capita in 1968 P.n Capit. GNP in S964 in us$ Less than 100 I D200 200 500 500 5000 1000-2000 200 and more Rwanda 0 Mauritania 2 Fiji 20 Romania 39 Libya 28 Switzerland 60 Burundi 3 British Solomon Rhodesia 20 Hungary 43 Puerto Rico 48 Sweden 66 Niger 3 Islands 4 Gabon 21 Bulgaria 47 Austria 51 Canada 75 Malawi 5 Papua & New Korea, North 22 Panama 47 Italy 51 U.S.A. 75 Nepal S Guineai) 4 Saudi Arabia 25 Cyprus 48 Czechoslovakia 52 Kuwait 80 Upper Volta S Swaziland 4 Honduras 26 South Africa 50 Norway 55 Yemen 6 Mozambique 6 Zambia 26 Ireland 51 Netherland Antilles 58 Afghanistan 7 Uganda 7 Ivory Coast 29 Trinidad & Tobago 53 Luxemburg 66 Ethiopia 7 Kenya 9 Guyana 30 Poland 56 New Zealand 66 Tanzania 7 Angola 11 Turkey 31 Spain 59 Finland 68 Chad 8 Khmer R. 13 Albania 35 U.S.S.R. 62 France 68 Sudan 8 Cameroon 13 Costa Rica 36 Greece 63 Belgium 69 Mali 9 Thailand 13 Jamaica 36 Argentina 71 Iceland 72 Guinea 11 Sierra Leone 14 Paraguay 36 Venezuela 76 Netherlands 72 Laos 13 Ceylon 17 Portugal 37 Hong Kong 100 United Kingdom 79 Malagasy 13 Philippines 23 British Honduras 37 Singapore 100 Denmark 80 Burma 16 Central African Dominican Rep. 37 Israel 81 Dahomey 16 Rep. 24 Guatemala 37 Germany, Fed. Pakistan 16 Bolivia 25 Ecuador 38 Rep. of 82 CO Togo 17 Viet-Nam, Rep. of 26 Surinam 38 Japan 83 a. Zaire 17 Senegal 27 Congo, Peop. R. 39 Germany, East 84 Haiti 18 Liberia 30 Yugoslavia 39 Australia 88 Indonesia 18 Southern Yemen 33 El Salvador 40 India 19 Ghana 34 Iran 39 Nigeria 21 Morocco 35 Lebanon 39 Botswana 22 Algeria 38 Iraq 43 China (Mainland) 23 Syria 37 Jordan 44 Somalia 24 Korea, Rep. of 39 Nicaragua 44 Viet-Nam, North 24 Tunisia 43 Barbados 45 Egypt, Arab R. 44 MalaysiatOi 46 Peru 46 Mongolia 52 Brazil 54 Colombia 55 Mauritius 55 Cuba 58 Mexico 58 China, Rep. of 64 Chile 74 Uruguay 84 Median 13 23 38 55 68 75 Noti The O pertnta-e oi urban population is the prnponrtrn ou the pupulation detined as .ub.n by the ountryns own detinition ti the total nountry population shown in peroentage. erc Pentaening t unban popuiation in Papum only. S (I Percentage on urban populaatin in west Malaysna only. Soure. P.e ca.pte GN P trom World Bant dum.unts. P.-netage on Mran Pnpuialhon tnn. Mogiey DaOvis, World 1azt/ome a 17a. V.I. 1, UOmversity ot caUiomI9a,59. t6 Annex 1 Table 6: Extent of Slums and Uncontrolled Settlements in Various Cities in Developing Countries Uncontrolled Settlement City At Percentage Population Total ot City Country City Year (Thousands) (Thousands) Population Africa Senegal Dakar 1969 500 150 30 Tanzania Dar es Salaam 1967 273 98 36 Zambia Lusaka 1967 194 53 27 Asia China (Taiwan) Taipei 1966 1,300 325 25 India Calcutta 1961 6,700 2,220 33 Indonesia Djakarta 1961 2,906 725 25 Iraq Baghdad 1965 1,745 500 29 Malaysia Kuala Lumpur 1961 400 100 25 Pakistan Karachi 1964 2,280 752 33 Republic of Korea Seoul 1970 440 (d.u.) 137 (d.u.) 30 Singapore Singapore 1966 1,870 980 15 Europe Turkey Total Urban Population 1965 10,800 2,365 22 Ankara 1965 979 460 47 1970 1,250 750 60 Izmir 1970 640 416 65 North and South America Brazil Rio de Janeiro 1947 2,050 400 20 1957 2,940 650 22 1961 3,326 900 27 Brasilia 1962 148 60 41 Chile Santiago 1964 2,184 546 25 Colombia Cali 1964 813 243 30 Buenaventura 1964 111 88 80 Mexico Mexico City 1952 2,372 330 14 1966 3,287 1,500 46 Peru Lima 1957 1,261 114 9 1961 1,716 360 21 1969 2,800 1,000 36 Venezuela Caracas 1961 1,330 280 21 1964 1,590 556 35 Maracaibo 1966 559 280 50 Source: U.N. General Assembly, Hotising, Building and P/anning. Problems and Priorities in Hluman Settlements, Report of the Secretary General August 1970, Annex 111, p. 55. Definitions vary Additional details are given in the source quoted. Note: d.u. = dwelling units 484 Annex 1 Table 7: Public Transportation in Selected Cities Average Average Daily Public Type of Number of Daily Transport Trips City Population Public Transport Units Owned Passengers per Habitant (Thousands) (Thousands) Abidjan 550 Buses 200 220-250 0.45 (1970) Addis Ababa 500 Buses 135 80 0.16 (1968) Bombay 5,200 Buses 1,338 1,810 0.351 (1967) Suburban rail 1,160 trains/day 2,130 0.4107 Caracas 1,719 Buses 925 1,300 0.76 (1966) Por Puestos(l) 6,000 New Delhi 3,000 Buses 820 750 0.26 (1966) Suburban rail 110 35 Sao Paulo 5,253 Buses/Trolley 7,000 4,718 0.90 (1967) Suburban rail Tehran 2,711 Buses 1,600 1,900 0.76 (1966) Calcutta 7,200 Buses 886 1,500 0.21 (1965) Suburban rail 450 trains/day 650 0.09 0.42 Trams 459 850 0.12J Bogota 2,340 Buses 4,170 3,782 1.62 (1969) Source: World Bank documents. (1) Collective taxis operating on regular routes Table 8: Automobile Registrations in Selected Cities City's % of National Registered Persons Total Population Automobiles per Car Popu- Auto- City Year National City National City National City lation mobiles (millions) (thousands) Abidjan 1968 4.1 0.51 43.9 30.9 93 16.5 12.5 70.4 Bangkok 1966 31.7 2.50 99.0 72.0 320 34.8 7.9 72.8 Bogota 1969 20.2 2.34 146.0 52.0 138 45.0 11.6 35.6 Bombay 1966 475.0 5.00 420.0 61.0 1,131 82.0 1.1 14.5 Caracas 1966 8.9 1.72 420.0 151.5 21 11.3 19.3 36.1 Djakarta 1969 118.0 4.50 212.2 77.2 557 58.0 3.8 36.6 Nairobi 1965 9.5 0.45 70.0 41.6 135 10.8 4.7 59.4 San Jose (C.R.) 1966 1.5 0.19 27.2 13.0 56 14.2 12.3 48.1 Sao Paulo 1967 86.6 5.25 1,417.9 366.1 61 14.4 6.1 25.8 Tehran 1966 25.1 2.71 174.0 85.5 144 31.7 10.8 49.1 Source: Automobile Manufacturers Association Facts and Figures and World Bank documents. 485 Annex 1 Table 9: Alternative Estimates of Passenger Capacity- Peak Hour Volumes/Track or Lane Rail Rapid Bus Rapid Private Auto Transit Transit on Expressway 1. 40-60,000 47,000 7,500 2. 62,000 23,000 7,500 3. 40,000 20,000 4. 25,000 4,000 1,100 5. 28,000 9,000 2,250 6. 35-40,000 22,500 3,200-3,600 (6,000) 7. 28,000 8. (see Table 10) 90,000 10,000 9. 36-48,000 14-18,000 4,000 10. 30-72,000 6- 7,500 Sources: 1. Highway Research Board, Highway Capacity Manual. Special Report N o. 87, 1965, p. 345. Calculations are theo- retical, not based on actual experience. The assumptions are:-Rail: Headway of 1.5 minutes 40 10-car trains, 24,000 seated passengers plus standing passengers. 940 buses per hour per bus lane, headway-3.8 seconds, bus seating capacity-50, speed-33 mph. Auto: 1,500 automobiles per hour, seating capacity-5. If 1.6 people per car is assumed the number of passengers would be 2.400. 2. Wilbur Smith & Associates, The Potential for Bus Rapid Transit. February 1970, Tables 8 and 10, and p. 147. Rail: Actual peak-hour passenger load on IND-8th Ave. Express, New York. This is the greatest volume among the seven different lines surveyed by the Institute of Traffic Engineers in 1965. Bus: Actual passenger load on the exclusive bus lane, the ramp to bus terminal, New York-the highest volume carried on the 14 expressways and local streets surveyed by the Institute of Traffic Engineers. Auto: Actual peak-hour passenger volume on Chicago's Congress Expressway. 3. Wilbur Smith and Associates, Future Highways and Urban Growth, 1961, p. 143, adopted from Hornburger, Wolfgang S., "Rapid Transit: Present and Future," Traffic Ouarterly. Volume XIV, April 1, 1960. 4. D. J. Reynolds, "Urban Motorways and Urban Congestion," in George A. Smerk (ed.) Readings in Urban Transporta- tion. Indiana University Press. Bloomington, Indiana, 1968, p. 225. 5. Walter S. Rainville, "The Importance of Urban Transit and Its Effective Passenger Capacities," in George A. Smerk (ed.) Readings in Urbaa Transportation, Indiana University Press, Bloomington, Indiana, 1968, pp. 176-177. Subway: A rapid transit train of eight cars operating on a 2-minute headway. Bus: A transit bus on an expressway lane operating on a peak-hour headway of 30 seconds (120 buses per hour) at 1,500 load factor. 1 bus lane = 4 auto lanes. Auto: An expressway lane operating at a rate of 1,500 vehicles per hour and carrying an average of 1.5 occupants. 6. Wilbur Smith and Associates, future Highways and Urban Growth. New Haven, 1961, pp. 144-147. Rail: Includes standing space and assumes 10-car trains, 50 seats per car, and 90-seconds headways. Bus: Assumes use of 400-450 fifty-seat buses with no stops and unimpeded use of freeway lanes. If buses stop at stations along freeways, an estimated 120 buses with 6,000 seats can be accommodated. Auto: Assumes an occupancy of 1.8 persons per car. 7. U.S. Dept. of Transportation, Resource Book on the Federal-Aid Highway Program, Sept. 1971, p. 322. Actual operations on a freeway lane have shown that close to 660 buses carrying almost 28,000 persons in one hour are possible with capacity to spare. S. U.S. Department of Transportation, Resource oonk On the Federal-Aid Highway Program, Sept. 1971, pp. 335-336. Calculations of theoretical possibilities. Rail: See Table 10. Bus: Based on maximum bus capacities established by different experts for an exclusive bus freeway lane (Texas Transportation Institute-1,250 buses/hour, FHWA -1,450 buses/hour, General Motors-1,200 buses/hour. Auto: 2,000 automobiles per hour on a freeway lane under maximum flow conditions. Assumes an occupancy of 5 persons per car. The estimate is for the maximum figure based on the lowest alternative, with an average of 50 seats and 25 standees possible per bus. 9. Walter S. Rainville, "The Importance of Urban Transit and Its Effective Passenger Capacities." in Smerk (ed.) pp. 175-178. Rail: Rapid transit 10-car train, 90-120 passengers/car at headways of 1.5 minutes and 40 trains per hour. Bus: 240 transit buses on a freeway operating on a peak-hour headway of 15 seconds. Auto: 2,000 vehicles per lane per hour with an occupancy of 2 persons per vehicle. 10. Wilbur Smith & Associates, "Center City Transportation Project: Urban Transportation Concepts." September 1970, Table 28. 486 Annex 1 Table 10: Capacities of Subways a. Passenger Capacities on Actual and Proposed Subway Systems Theoretical Standeesj Cars/ Trains/ Capacity City Seats/Car Car Train Hr. (Persons/Hr.) New York 50 250 10 32 99,000-16,000 Toronto 83 217 6 30 54,000-14,940 Rotterdam 80 210 4 30 34,000- 9,600 San Francisco 72 - 10 40 28,000-28,000 Montreal 40 100 9 24 30,200- 8,640 Pittsburgh 28 42 10 30 21,000- 8,400 Washington, D.C. 81 94 8 30 72,000-19,440 b. Maximum Passenger Capacity vs. Actual Usage Present City Capacity Volume Usage (Existing Systems) New York 99,200 61,400 0.62 Chicago 37,500 10,373 0.28 Cleveland 11,200 6,211 0.55 Toronto 54,000 35,166 0.65 (Planned Systems) San Francisco 88,800 28,000 (est.) 0.32 Washington, D.C. 42,000 31,900 (est.) 0.76 c. Reported Peak-Hour Track Loads City Persons per Track per Hour Toronto 30,000 Chicago 28,000 Philadelphia 20,000 Boston 18,000 Cleveland 10,000 Sources: Tables a and b from the U.S. Department of Transportation, Resource Book on the Federal Aid Highway PFro- gram. Sept. 1971. pp. 338-339 and Table c from Wilbur Smith and Associates, Future Highways and Urban Growth. New Haven, 1961, p. 126. 487 Annex 2 URBAN LAND POLICY Introduction Three principal problems affect policies on the use of urban land in the developing countries-the difficulty of acquiring land for de- velopment within the urban areas, the inefficient pattern of urban development and land use, and the severe limitation of resources for further urban development. The severity of these problems varies greatly between cities be- cause of the great diversity of urban patterns around the world- differences in cultures and in the organization of society; in legal and communal tenure of land; in major economic activities and political- economic organization; and in historical patterns of development. All of these factors have influenced, and have in turn been affected by, patterns of land use and land ownership. In virtually all cities of developing countries there are problems of underutilization of existing serviced land and of constraints on urban land extension. Underutilization frequently results in part from his- torical patterns of development-the types of buildings constructed, the layout of streets, etc.-and withholding of land from the market for investment speculation, to which may be added deficiencies in zoning laws and/or their enforcement. Development is restricted by the poverty of the bulk of the population and the paucity of municipal government resources. If a substantial proportion of the land is owned by a small number of large landowners, their monopoly/ oligopoly behavior may further retard inner city development and extension of the urban area. High and rising land prices, though partly an inevitable result of rapid urban population and economic growth, are accentuated, perhaps greatly, by such factors. Rising prices en- courage speculative withholding of both serviced and unserviced land, which aggravates all three basic problems. Other factors retard- ing urban land extension are constraints on public sector provision of infrastructural support for private sector development either ex ante or ex post or simple failure of owners to perceive suitable investment returns from urban land development. As a result, substantial areas of unplanned vacant land and low density development are common features, with the percentage of vacant land reportedly ranging as high as 75%. In some cities substantial tracts of land are owned by government and in a few of these (of which Singapore is a notable example) this "opportunity" has been seized with vigor and no little imagination. In most cities, however, because such land falls under the control of many ministries or government entities, because there is no develop- 488 Annex 2 ment program, or for some combination of these and other reasons, the land is not brought into productive use. Many other factors may have an impact on curtailing development or contribute to undesired patterns of development. Particular difficulties arise, for example as in Bombay, where rent controls inhibit the construction of medium and low income housing in or near the center city where most jobs exist, the absence or inadequacy of public services (e.g. water and telephones) discourages the location of economic activities in out- lying areas of the city where many workers live, and heavy subsidi- zation of the transport network tends to fossilize this historically determined pattern. Objectives The more commonly identified problems of rapid rise in land prices and speculative withholding need to be viewed against this back- ground of scarcity of land readily available for construction, ineffi- cient patterns of urban development and severe resource limitations. Rather extensive legal powers frequently exist on paper to deal with these problems. However, a lack of clarity as to objectives has meant that these have not always been appropriately designed. Even where they have been, they have usually not been used with great frequency or determination. The objectives would seem to be fivefold: 1. To increase the supply of urban land for development. 2. To mobilize the necessary financial (and other) resources for de- velopment. 3. To influence development so as to improve patterns of land use and the spatial and sectoral allocations of investment. 4. To achieve a greater measure of distributional equity on both the resource and the expenditure side, keeping in mind externalities and secondary income effects. 5. As part of 2 and 4, to obtain for public use a satisfactory portion of betterment values in private land which are induced by public investment. Instruments' A number of attempts have been made to develop a taxonomy for the instruments or measures for achieving these objectives. None have proved free from problems. To record experience in the market economies to date, however, three types of measures may be dis- tinguished. 'This section draws heavily upon the very usetul papers prepared for the United Nations Interregional Semiiinar on Urban Land Policies and Land Use Control Measures held in Madrid, 1-13 November-, 1971. AgA Annex 2 A. Non-tax Measures designed to influence private development. B. Tax Measures designed to influence private development and/or raise revenues for public development. C. Direct Public Involvement in the land and land development mar- ket designed to substitute public for private sector development, to encourage certain private sector development and/or raise revenues. It should be pointed out, at the outset, that for purposes of brevity and clarity these measures are discussed individually. However, in practice many are highly interdependent and the use of one alone may be ineffective. Similarly, they vary greatly in the administrative burdens imposed and their practical usefulness should be assessed with this in mind. A. Non-tax Measures Non-tax measures cover regulation of land ownership, trading and use and public assistance for development. The most important regu- lations affecting land ownership are those having to do with a forced transfer of ownership rights when land is required for public pur- poses. Others seek to improve the functioning of the land market by forcing vacant land into use and facilitating trading. Regulation of land trading has generally been confined to the establishment of price controls. Land use controls exist in most countries, whether or not any formal planning exists, and include zoning and sub-division regula- tions; building regulations and development permits; and planning concepts such as controlled areas and urbanized or "urbanizable" limits. Various schemes of public assistance for land assemblage and joint adjustment of land use planning have been adopted in recent years. 1. Regulation of Land Ownership Almost every country has adopted regulations which limit private ownership rights in cases involving the general public welfare. In North America eminent domain and police power are the devices. According to eminent domain, the government may assume owner- ship of private land for the benefit of the community, provided just compensation is paid. Police power refers to the right of the authori- ties to limit the exercise of private rights in real property without com- pensation if the health, safety, morals or general welfare of the public is enhanced thereby. Similar laws exist in Western Europe. Generally, the individual landowner may challenge the decision to make his land available for a public project. It is then up to the courts to examine the issues and decide whether the authorities' proposed action is justified. 490 Annex 2 In France the preemption right gives the authorities priority in buy- ing land from private owners. In India, New Delhi has embarked upon a land purchase program based on compulsory sale of private land. So far half the designated 66,000 acres falling within the urbanized limits defined in the master plan have been acquired. This land can- not be resold, but may only be leased for development purposes specified in the plan. Expropriation of vacant land has a long history. in the Ottoman Empire, for example, the concept of Machlul provided that unculti- vated agricultural land becomes state property after three years. The power to expropriate vacant land has been legislated recently in Cuba, Chile, Ceylon and Spain. In China (Taiwan) a law provides that, if an owner's land exceeds a maximum prescribed area, he must sell the excess portion within two years. Otherwise, the local authority may purchase the land. In such cases of compulsory sale, the land ownership regulations in Taiwan and a number of other countries forbid payment to owners of prices higher than those obtaining prior to announcement of the intention to use the land for public purposes. 2. Regulation of Land Trading The most frequently used land trading regulations have been price controls and price freezing. Both are intended principally to restrain speculative increases in land values. Cuba has established a ceiling on the maximum allowable unit price of land; in addition, the sale of real estate is banned to prevent accumulation, although units can be exchanged. France and Spain have adopted land-price freezing in connection with plans for priority development zones. General registration and evaluation of land prices can be found in Australia, Denmark and Japan. Japan has established a "guidepost" system of price control which involves periodic official announce- ments of land values for selected standard tracts of land. A similar sys- tem exists in Spain. In South Korea the law establishing the transfer tax requires posting by government, at least twice a year, of "standard market prices." The Netherlands has pioneered in controlling land values through active participation in the market. Feeding their own substantial land holdings into the market at appropriate times, municipalities have achieved quite remarkable results. Consequently, the most densely populated country in Europe has had the lowest rate of increase of urban land prices, and the weight of land in urban construction costs has remained stable for 50 years. A f~ Annex 2 3. Regulation of Land Use Land use regulations are intended to guide the physical growth of the city according to a master plan or whatever other framework for development exists. As with all control measures, interest groups are bound to exert pressure on decision makers, influencing them to adopt plans consistent with their diverse interests. Thus, in addition to planning considerations, such master plans as exist reflect socio-eco- nomic and political realities prevailing at the time of their adoption. Zoning and subdivision laws vary between countries. A particularly restrictive version is found in North America where, among other things, zoning regulations have typically been used to "protect prop- erty values" in higher income residential areas. In Britain and other countries of Western Europe much broader concepts of zoning have been in use for many years. Among Latin American countries, Chile and Venezuela have sig- nificant land use controls. In Venezuela the subdivision developer must cover the costs of infrastructure services provided by local gov- ernment. Before a subdivision can be approved, the Municipal Coun- cil must confirm that the city has not designated the area for other uses. In Chile, the central government exercises control over land use on a national scale. In theory, this enables government to direct physi- cal development not only within city limits but also on the peripheries and throughout the country. In France and Japan the national plans specify zones of priority de- velopment and deferred development. One function of this system is to exercise some control over the urbanization process by defining urbanizable limits. In India, the city of Chandigarh has adopted a Periphery Control Act under which the area within five miles of the city boundaries on all sides is declared a controlled area. Develop- ment of this territory must proceed in accordance with a periphery land use plan intended to prevent squatting and uncontrolled devel- opment. Other land use regulatory devices are building regulations, in- cluding codes and by-laws, and development permits. Where such regulations exist, developers are required to submit plans to the authorities for approval to ensure that they conform with building codes.Without such approval, workcannotcommence. Development permits have been employed effectively in Britain where the permit covers the type and timing of development. Once an appeal for a per- mit is filed the authorities must respond within a period not exceed- ing two months. This time limit is important in light of long delays experienced in such countries as Nigeria, Ghana and Malaysia. In Italy AG') Annex 2 and Spain the central government exercises final control over the granting of building permits. In the United States the states handle building permits and related regulations. As a result, regulations vary widely and this has been a factor preventing the development of in- dustrialized housing. The controlled development of infrastructure is another potential policy tool which, particularly if incorporated into a comprehensive land development plan, would assist municipalities in directing land into desired use patterns. To be sure, land can be brought into use without prior or adequate provision of power, water, sewerage, roads and other services; and developers frequently are motivated to pro- vide infrastructure as part of a development scheme. However, there is no doubt that government participation in the financing and/or provision of infrastructure can be an important factor encouraging development according to desired patterns. By the same token, the withholding or proscription of infrastructure can effectively discour- age development where it is not desired. It is somewhat surprising that conscious application of such policies has been so infrequent; however, such methods have been used as part of the general plan- ning process in Britain and in some towns of Canada. 4. Land Regulations Related to Public Assistance Some countries have adopted regulations designed to facilitate de- velopment schemes through public assistance. The main idea behind this approach is that cooperation between the private and public sec- tors may remove difficulties associated with the raising of capital, land assemblage, and improvement of land use patterns. Again, programs have varied from country to country. In India, a "pooling" technique has been utilized in assembling large tracts in- stead of negotiating with each owner-a long and costly process. The land is assembled, given a new layout, provided with infrastructure services and redistributed among the original owners. Since this process raises land values considerably, values before and after devel- opment are taken into account in compensating the owners. Also, the reduced size of plots as a result of provision for parks and other public space is considered for the purpose of determining just compensation and ownership rights. A similar method, called land readjustment, has been used extensively in Japan since 1909 and was introduced more recently in South Korea. In Western Europe several forms of public assistance in land devel- opment exist. In France, Spain, Italy, and the Federal Republic of Germany the demand for land in different municipalities has resulted in the creation of landowners' associations. These bodies attempt to Annex 2 coordinate with the public authorities the implementation of land development programs. Another frequently used technique for mutual benefit is the mixed private-public company. The landowners' association pools the land of a group of owners, and the joint com- pany undertakes development of the land according to plans ap- proved by the municipality. The private landowners retain collectively some ownership rights in the form of leasing privileges and develop- ment rights. B. Tax Measures All taxes have as one objective the raising of revenues. Given the limited resources of municipalities, this must be regarded as a most important feature of the overall fiscal system if not of individual taxes. In addition, taxation of vacant land, particularly at penal rates, is de- signed to bring the land into productive use. Other tax measures are used to influence the desired location and type of land development, to improve distributional equity and/or to discourage speculation. The betterment tax is of special interest since it operates on rising land values; and, if properly applied, can be a very important source of revenue. However, there are difficulties in attempting to establish betterment tax policies. These difficulties stem from differing views as to the causes of increases in real property values. For example, in- creases in land values may occur as a result of the general increase in demand (resulting from population and economic growth), from pub- licly financed betterment and/or from privately financed betterment. The rationale for and methods of taxation differ correspondingly. Basically, it is the betterment values arising from the first two factors that one would like to "capture" for public sector use. In so doing, it is necessary also to attempt to distinguish between real increases in value and those caused by inflation. 1. Taxation of Vacant Land The main purpose of a tax on vacant land is to bring land into pro- ductive use to meet the needs of a growing urban population and economy. Ideally, the tax would motivate landowners either to de- velop their sites (in accordance with a master plan if one exists), or to sell the property to interested private or public concerns willing to undertake development. A number of countries have such taxes. In China (Taiwan) vacant land which is not developed within a prescribed period is subjected to a tax ranging from three to ten times the land value tax, which is the standard real property tax. In South Korea there is a "Real Prop- erty Speculation Check Tax" applicable to "untransferred unoccu- aq4 Annex 2 pied" land. The law includes in the definition of unoccupied or vacant land lots with structures that cover no more than one-tenth of the total area, thereby encouraging also the development of under- utilized land. In Chile municipalities levy a tax based on the assessed values of all vacant land in urban areas. This rate begins at 3% and is increased annually to a maximum of 6%. In Uruguay, in addition to a tax on vacant land, there is a special tax on "unused" lots where the value of the land is greater by a factor of 4.5 or more than the value of struc- tures on the land. Both taxes are on a progressive scale. Syria also has adopted a progressive tax that increases from 1% to 5% of assessed value. Although in principle a very well-conceived instrument, in most countries the vacant land tax suffers from a number of deficiencies- including (1) improper and outdated valuation, (2) widespread tax evasion, and (3) inefficiency of the tax administration generally. 2. Taxation to Influence the Location and/or Type of Development Land taxation has been recognized as a potentially effective means of achieving desired patterns of land development in urban areas. However, it has not been fully utilized as yet in most parts of the world because of inexperience, political hesitation and the pressure of interest groups. As the urbanization process continues and urban problems mount, it can be expected that municipalities will increas- ingly resort to land taxation as a policy tool to influence the location and type of urban development. In principle these taxes should be highest in areas which attract investments and lower in second choice areas. Hopefully this would contribute to a more balanced distribu- tion of land uses, providing the population with employment centers, commercial and cultural facilities in the areas where they live. In France, under a law aimed at discouraging land development not in accordance with recommended uses, a developer is taxed 90% of the added value accruing to his property following additions or reno- vations not in conformity with the municipal plan. A similar law in Spain imposes an 8% tax on the added value to the property in urban areas, and 2% in rural zones. These low rates can be expected to be considerably less effective than those in France, provided the latter are enforceable. The land tax system in the State of Hawaii-follow- ing a different approach-establishes tax rates according not to the existing land use patterns, but rather in conformity with their ap- proved use in the master plan. Though not perfect, this system pro- vides considerable incentives to reduce discrepancies between actual and desired land use. AQ,G Annex 2 3. Land Transfer Taxation The rapid rise in land values in urban areas has no doubt encour- aged speculation and the development of land markets involving, usually, a small group of traders controlling substantial resources. As public authorities have become aware of this situation and, in par- ticular, of the fact that land prices in many cities of developing coun- tries currently exceed those in developed countries, there have been attempts to devise instruments to exercise a measure of control over the land market, so that the growing need for urban land can be satis- fied at more reasonable cost. A popular tax measure aimed at checking speculative land trans- actions has been the land transfer tax. In principle it discourages spec- ulative trading by cutting into speculators' profits, at the same time providing the public with a share of market values. On the other hand, the tax does nothing to discourage speculative withholding, and- unless the tax is confined to, or the rate is much higher for trans- actions taking place within a short period of time-discourages all transactions, including those which would bring land into produc- tive use. The rate of the land transfer tax varies from country to country. In Afghanistan and Jordan it amounts to 1% of the sales price of land, in France 1.25%, Lebanon 3% (5% for foreigners), West Germany 5% and Spain 7.4%. In most if not all countries the low rates imposed have had little or no success in discouraging speculative land trans- actions. In South Korea the law imposes a special tax payable by the seller on the transfer of land. This is not so much a transfer tax as a capital gains tax applicable only at the time of transfer. The tax rate is based on a complicated formula involving the difference between acquisi- tion and sales price, capital expenditures made during ownership and administrative costs involved in the sale. The tax figure is further ad- justed in accordance with the "wholesale commodity price index." The result is that the tax approaches 50% of the capital gain. 4. Taxation on the Income from Real Property Urban real estate produces income in the form of rentals. In cities with high rates of automobile ownership, even vacant land may pro- vide income if used for commercial parking. Taxation of income from real property is an important source of revenue for most governments. In countries as diverse as the United States and India income taxes apply equally to income from real es- tate. However, these taxes operate as well in a reverse way. Expenses 496 Annex 2 deductible against total income include: maintenance costs on build- ings, service charges and local taxes, costs of collecting rent, mortgage interest and (in the United States) accelerated depreciation. It is rea- sonable to assume that laws so favoring landowners are bound to increase speculative pressures and do violence to commonly held concepts of equity. The British tax system distinguishes between income earned from employment ("earned income") and that from real property or other wealth ("unearned income"). The latter is taxed at a higher level with a minimum rate of 40%. In Nepal the urban property tax is increased by 50% if the land is used as an income source. In Iraq there is a tax and progressive surtax ranging from 1% to 16% on real estate income. In Jordan there is a tax of 17% on the rent earned from real property. 5. Taxation on Real Property and Betterment Values Taxation on real property normally constitutes the principal source of revenue from land. As urbanization proceeds, real property and betterment taxation are viewed equally as a control measure and an indispensable financial source for municipal governments. A well- designed tax program would provide the municipal authorities with a fair share of the benefits accruing to landowners, while encouraging the latter to maintain and develop their property in accordance with the interests of the community. The forms and rates of taxation on real property and betterment values vary widely. In Taiwan, there are two taxes: (1) the Land Value Tax, and (2) the Land Value Increment Tax. The basis for the former is the unimproved value of land, with assessment on the basis of the owner's self-declaration of value. The local government reserves the right of compulsory purchase at the owner's declared value; this power serves as a deterrent to tax evasion through understatement. This tax is annual and progressive from 1.5 to 7%. The Land Value Increment Tax is levied on the net increment in land value at the time of its transfer or after a lapse of ten years since the last assessment. This tax too is progressive, rising from 20 to 80%. Two other taxes may be levied to cover betterment values: (1) the Annual Tax on Construc- tional Improvement, and (2) the Constructional Benefit Charge. The first applies to improvements brought about by the owner's invest- ments. The second covers increases in values attributable to public works. Though in principle this appears to be a marvelous combina- tion of taxes, there are severe problems with assessment and en- forcement. In India, municipalities levy property and betterment taxes, but the income from these sources has been far below expected levels be- 4c7 Annex 2 cause of deficiencies in the assessment system and collection ma- chinery. As a result, many municipalities have refrained from levying these taxes. In Latin America, inflationary pressures have contributed to spec- tacular increases in property values. In fact, investment in land has become one of the most attractive means of protection against infla- tion. This situation requires more frequent assessments of real prop- erty values in order to adjust taxes to new market values. However, municipalities have generally failed to do this, utilizing unrealistically low values for the tax base. C. Direct Public Involvement The case for direct public involvement in urban land markets de- rives from the fact that the desired results cannot be achieved solely by seeking to influence private sector behavior-even by the use of substantial tax and non-tax incentives. It would seem that this involve- ment should be developed with three principal objectives in mind: 1. Increasing the supply of adequately serviced urban land at a rate sufficient to reduce land prices and/or reduce their rate of increase to a "tolerable" level; 2. Economizing on total resources by bringing about a better pat- tern of development; and 3. Capturing a substantial share of the betterment values associated with provision, by the public sector, of urban infrastructure and services. These objectives correspond closely to the three problems de- scribed at the outset. To be successful, such intervention should be so arranged as to permit public authorities to enter or leave the market at any stage-as dictated by circumstances. But, first, it is necessary to deal with the problem of acquisition subject to fiscal constraint. Acquisition can take place in a number of ways. To the aforementioned means of eminent domain, compulsory purchase and the exercise of preemption rights, should be added such non- compulsory means as purchase on the free market. In addition, the problem of finance must be overcome. It seems likely that, once a public market for the purchase, development and resale or rental of land is developed, the process can be financially self-sustaining. To start such a process in a situation where fiscal re- sources are slight, it might be worth considering exchanging bonds for property-in much the same way that the Japanese financed their rural agricultural land reform in the last century. In any case, public involvement in urban land markets must be comprehensive. The "package" of policy inputs might include appropriate combinations 498 Annex 2 of the following: declaration of value by owners (in the knowledge that this declaration will be used for both tax and expropriation pur- poses); the designation of zones of priority and deferred develop- ment with appropriate time limits and price-freezing provisions; an "appropriate" rate of annual property tax; annual expropriation of significant amounts of land; a "balancing" of expropriation quantities and tax rates to create incentives for the "declared value" to be as near as possible to actual value; public development of the expropri- ated land; and a mechanism for capturing betterment values. 1. Acquisition The best known examples of this far-sighted approach are provided by the municipalities of several Western European countries. Most of these countries currently acquire land for a variety of purposes-the leading examples being Sweden, Norway, the Netherlands, Germany, France and Spain. The Swedish approach began in Stockholm before the end of the 19th century and is now approved for all municipalities. It involves purchases in the private market, supported by the right of municipal preemption. The city council of Stockholm has accumulated enough land about the city to effect housing developments for the working population and entire satellite towns, and has found this policy ex- tremely rewarding financially. This has enabled it to build up reserves almost from the start of the program. As already indicated, a similar situation exists in the Netherlands. In West Germany, in the period 1948-1965, public authorities in- vested $4 billion in land acquisition, considerably increasing existing municipal land reserves. In 1963, 31% of the land in towns of 100,000 population or more was under public ownership. In cities of one mil- lion and over, the municipalities controlled 37% of the land. In less developed countries, the situation varies widely. In Asia, Hong Kong, Singapore, China (Taiwan), Afghanistan and India have programs. In Latin America, with the exception of Cuba and Chile, few countries have a positive policy of public land ownership. In Argentina, the state owns substantial tracts of valuable land in Buenos Aires and elsewhere, but no development program exists for this land. In fact, there has recently been a tendency to sell off tracts to meet the immediate fiscal problem. Ecuador, which built up a reserve of pub- licly held land during the 1940's, has recently not only ceased to acquire land, but begun a systematic process of donating substantial quantities of public land to social institutions. In Brazil, the majority of urban real estate is in private hands, except in the new capital of Brasilia. 4QQ Annex 2 The experience in land acquisition points to many advantages of creating a Public Land Bank, which perhaps explains an accelerating trend in the developing countries, as in the world generally, towards increased public ownership of land. In developing countries in which the private land ownership concept is not firmly established, as for example in some areas of Africa, there would seem to be a strong case for municipalities to clearly define in law their control over this land. 2. Development Public investment in the land market covers, in addition to acquisi- tion, development for such purposes as low and moderate income housing, schools, green belts, etc. However, the two aspects are strongly interrelated. Expropriation, by whatever means, has been gaining favor in recent years largely because an increasing number of countries have come to regard housing, particularly for the urban poor, as an extremely important social and political matter-an im- portant public service that fully warrants the power of condemnation. Housing and roads provided the main stimulus for the development of land acquisition policies around the world, though the United Kingdom (where new towns were also an important factor) and the Netherlands (where public land has been used to regulate urban land prices for general planning purposes) provide at least partial excep- tions to the rule. Initially, in most countries, expropriation was limited to the right of eminent domain and the latter applied only to specific -and of these often only important-development schemes. These restrictions are rather rapidly being relaxed in most countries. In the U.S. and Canada the concept has been stretched gradually to cover first housing and then general development uses (including private sector uses). In Spain and Italy, all that is needed to use this technique is general plan approval to develop a certain area; and in France, acquisition by expropriation is explicitly permitted in order to build up reserves. Land acquisition and development is administered at the local level in most countries of Europe and North America, though this may be about to change in favor of regional concepts. National institutions are responsible in Britain and Spain, and regional institutions in France. Elsewhere, control is generally at the national level. However administered, acquisition and development are encumbered by lack of information, confusion or disagreement concerning title and/or assessed value. In many cases, the process is substantially delayed while litigation on such matters is proceeding. Recently, some coun- tries have tackled this issue by taking possession first and dealing with such matters later. Ceylon has attempted to convert confusion over 500 Annex 2 title from a liability into an asset by specifically providing for land with defective titles to be expropriated for public use. In the United States, different programs exist under federal legisla- tion providing for federal aid to cities for urban renewal, including low and moderate income housing. In Western Europe, many munici- palities actively engage in residential construction for low income citizens. Hong Kong, Singapore, and Israel provide additional exam- ples of public involvement in general land development and housing. The United Kingdom was the first country to adopt a national new town policy as a means of easing congestion in the older cities. 3. Redistribution Acquisition, development and redistribution of public land follow different paths in different countries. The financial aspects of public land acquisition, development and resale or lease take on major im- portance, since no plans can be executed without adequate resources. Most countries rely upon some combination of "prior financing" and the realization of profit upon resale or lease. However, since there is no case where public sector financing provides adequate funds for accumulating reserves, primary emphasis has been placed upon the latter approach. Sweden, United Kingdom, France and Spain have recently budg- eted some funds for the purposes of reserve accumulation, but the amounts involved are almost derisory compared to the requirements. Norway has adopted a rather unique practice of public sector rental of private lands, which helps to mitigate the financial problem. Other countries have relied to some extent upon borrowed capital. In France and Sweden, short and medium-term financing are uti- lized for land acquisition in cases where the public authority buys land for early resale to construction firms. If the land is intended to be kept in public ownership for possible lease to private developers, the financing is based upon long-term credit. The more favorable financ- ing terms given in this case are designed to encourage leasing of land so that it may be retained in public control. In some instances mu- nicipalities adopt a policy of selling for housing but leasing for com- mercial and industrial uses. The wisdom of such a distinction has recently received considerable testimony in Britain where the expira- tion of long-term residential leases has caused severe personal and political problems. As for redistribution, in some countries, e.g. the Netherlands, land is resold to the private sector in its undeveloped state. In the bulk of countries (particularly among developing countries), most land is turned over to housing corporations. But in a fair number of instances 501 Annex 2 it is turned over to general development institutions. Examples in the developing countries are the Bombay and New Delhi Development Corporations in India and the Guayana Development Corporation in Venezuela. These new multi-function institutions work on a range of problems extending from housing to unit development in urban areas to new town development. Redistribution of land to the private sector is by sale in most coun- tries, by sale or lease in Sweden. In Hong Kong and Singapore, where much of the land started out under public control as Crown Lands, the authorities have kept it under public ownership, leasing it (and whatever public facilities have been added) to the private sector. This has obvious advantages, particularly since the period of the leases is shortened to a point where it is much easier to capture "unearned increments" and bring about appropriate changes in land use patterns. 502 Annex 3 INTERNATIONAL ASSISTANCE IN URBAN DEVELOPMENT International assistance, both multilateral and bilateral, has been slow in focussing on the need of less developed countries for help in dealing with urbanization. In part, this has been due to a lack of understanding of the gravity of conditions and the fact that cities have not been perceived as positive integral parts of the national develop- ment process. The scarcity of funds has also mitigated against alloca- tions for urban development per se, a new, relatively undefined area of development investment. While these attitudes in part still exist, there is a growing awareness of both the magnitude of the urbaniza- tion phenomenon in the less developed countries and the need to act upon its inherent opportunities and dangers for development. Available information on the kind and amount of international assistance in urban development, especially financial data, is frag- mentary. However, while incomplete, the information presented here is sufficient to demonstrate that to date the amount of involvement has been quite limited. A. Multilateral Assistance According to estimates made by the United Nations Centre for Housing, Building and Planning (UNCHBP), in the 1966-1967 bien- nium a total of $700 million was expended for international programs in housing, building and planning. Of this total $570 million was ac- tually reported by international agencies, and an additional $130 mil- lion was added by UNCHBP to account for those agencies which did not respond to their inquiry. The estimate for unaccounted expendi- tures represents 22.5% of the total which was reported. The average expenditure of $350 million annually is also approximately what UNCHBP estimated for the 1963-1965 period. United Nations The United Nations undertakes several different kinds of activities in the area of urban development. These include the collection, analy- sis and dissemination of information; research on specific problems; training programs and pre-investment projects. The Centre for Hous- ing, Building and Planning is the UN unit that undertakes these activi- ties at the headquarters level, and the regional economic commissions do similar work within their respective regions, with the exception of pre-investment studies. The special UN agencies, such as WHO, FAO -sn'l Annex 3 and ILO, also are involved in these activities. The United Nations Development Programme (UNDP) is responsible for the program- ming and financing of most technical assistance and pre-investment projects of the UN parent body. The special agencies undertake their own activities and, along with the Office of Technical Cooperation of the agencies for UNDP pre-investment projects. Projects vary from urban and regional development planning and training undertaken by an expert team of 10 to 15 members lasting three years to the ap- pointment of a single housing finance advisor for several months. Urban development has received a relatively low level of commit- ment in both staff resources and technical assistance funds in the UN. However, although the amount continues to remain small, there is evidence of an increasing trend. In 1963 expenditures of the UN regu- lar budget and the UNDP together for housing, building and physical planning was about $1.35 million; in 1970, it was about $3.86 million. The following table summarizes the amount of funds expended by the UN bodies in international programs for housing, building and physical planning. Table 1: UN Funds Expended in International Programs for Housing, Building and Physical Planning (Amounts in $ thousands) Total Amount Average annual amount UN Source 1963-1970 in 1966-1969 Un regular budget 12,817(0, 1,681 UN extra-budgetary (UNDP, Funds in Trust) 16,109 2,837 UN special agencies (WHO, FAO, 1LO)(b' - 6,950 Total 28,926 11,468 (a) This does not include operating expenditures of the administrative agency (UNCHBP) for 1963, 1964, 1965. (b) UN special agencies not included: Unesco, UNICEF, UNIDO, World Bank Group. Source: Centre for Housing, Building and Planning, UN, New York. June 1971. Aside from the above information on housing, building and plan- ning, in the available data on UN projects, the urban development component is not specifically identified. The following table of the UNDP Special Fund capital commitments during the First Develop- ment Decade (1960-1970) indicates the magnitude of projects in housing, building and physical planning in comparison to other UN designated sectors, and Table 3 contains a selected list of ongoing UNDP Special Fund projects. The UN has recently expressed a considerable interest in greater coordination of its pre-investment studies and technical assistance with World Bank lending in the field of urban development. rIflA Annex 3 Table 2: UNDP/SF Capital Commitments 1960-1970 (Amounts in $ thousands) No. of Total Project UNDP UN Project Sector Projects(a) Cost Contribution Agriculture 457 1,010,078 429,930 Education and Service 106 468,510 341,369 Health 15 51,331 39,164 Housing, Building and Physical Planning 21 48,410 29,843 Industry 323 664,685 387,311 Multi-Sector 69 150,521 64,323 Public Administration and Other Services 57 134,405 81,383 Public Utilities (including transport) 180 354,728 188,615 Social Welfare 6 9,204 5,753 Total 1,234 2,891,872 1,567,691 (a) Excludes cancelled projects. Source: Status of Approved Projects in the Special fund Compoonent As of December 31. 1970, OOiSFifleports UROP, New York, 1971. Regional Political and Economic Bodies European Economic Community (EEC) EEC through Fonds Europeen de Developpement has given some assistance for urban development in African, Asian, and Latin Ameri- can countries in the form of technical assistance and capital loans. Between 1964 and 1969 urban development-related aid from this source amounted to approximately $39.7 million. This includes ex- penditure on water supply and drainage, sites for dwellings, improve- ment of central marketing facilities and airports. Organization of American States (OAS) Six years ago OAS established a Division of Urban Development for the purpose of giving technical assistance in two areas to member countries: the formulation of urban development policy and the transmission of housing and construction technology. In 1970 the Division's budget amounted to some $2.2 million and its projected budget for 1975 amounts to $2.8 million. Regional Development Banks Inter-American Development Bank (IDB) Since its establishment in 1960, IDB has played a major role in international aid for urban development. Drawing on funds primarily from the Social Progress Trust Fund ($575 million, contributed by the United States), IDB began its urban development activity by financing 505 Annex 3 Table 3: Selected On-Going UNDP/SF-Financed Studies in Urban Areas-1970 Commencement Executing UNDP/SF Date of Country and Project Agency Contribution Project AFGHANISTAN Central Authority for Housing and Town Planning, Kabul UN $ 865,000 1966 BRAZIL Assistance in Strengthening the Operations Research and Control System of the National Housing Bank (BNH) UN 546,500 1969 CHINA Urban and Housing Development UN 727,800 1967 Sewerage Planning in the Greater Taipei Area WHO 517,400 1969 The Preparation of a Consolidated Community Development Programme UN 405,000 1969 COLOMBIA Transport and Urban Development Study, Bogota IBRD 207,000 1969 GHANA Preparation of a Master Plan for Water Supply and Sewerage for the Accra-Tema Metropolitan Area (Phase II) WHO 1,726,400 1967 IRAN Pre-lnvestment Survey of Sewerage Needs and Facilities in Teheran WHO 1,110,000 1969 IRELAND National Institute for Physical Planning and Construction Research UN 725,000 1964 IVORY COAST Water Supply and Sewerage for Abidjan WHO 503,100 1970 NEPAL Development of Water Supply and Sewerage, Greater Kathmandu and Bhaktapur WHO 752,400 1970 NIGERIA Master Plans for Wastes Disposal and Drainage, Ibadan WHO 867,500 1969 PAKISTAN Location and Planning of Cities in the East UN 1,009,700 1967 Master Plan for the Metropolitan Area of Karachi UN 1,208,100 1970 PERU Experimental Housing Project, Lima UN 1,065,500 1968 PHILIPPINES Master Plan for a Sewerage System for the Manila Metropolitan Area WHO 714,000 1967 SENEGAL Establishment of a Master Plan for Water Supply and Sewerage for Dakar and Surrounding Areas WHO 2,170,300 1968 506 Annex 3 Table 3: Selected On-Going UNDP/SF-Financed Studies in Urban Areas-1970 (continued) Commencement Executing UNDP/SF Date of Country and Project Agency Contribution Project SINGAPORE Assistance in Urban Renewal and Development UN $1,576,400 1967 SYRIA Centre for Housing and Construction, Damascus UN 681,000 1969 TOGO Building Construction Centre, Cacavelli UN 1,132,700 1968 TURKEY Master Plan for Water Supply and Sewerage for the Istanbul Region WHO 1,222,900 1966 UGANDA Master Plans for Water Supply and Sewerage for the Greater Kampala and Jinja Areas WHO 616,800 1968 VENEZUELA Urban Research and Planning UN 767,300 1968 two types of projects: water and sewerage, and housing construction (excluding land and administration costs). Subsequently, its housing finance policy changed to one of supporting housing in combination with related water, sewerage and road facilities, and at present IDB finances housing only within the framework of integrated urban de- velopment programs. Housing projects per se are no longer being considered by the IDB. Recently, IDB has initiated technical assis- tance, including pre-investment studies, for several large urban food Table 4: IDB Urban Development Assistance, 1961-1970 (Amounts in $ thousands) Total Capital Completed Projects Ongoing Projects(a) Commitment No. No. NO. Project Type Projects Amount Projects Amount Projects Amount Electric Power 2 12,822 8 170,632 10 183,454 Food Marketing Facilities 1 600 1 6,300 2 6,900 Housing and Related Community Services(b) 20 113,214 12 105,280 32 218,494 Manufacturing Plants 12 31,562 1 4,587 13 36,149 Water & Sewerage 31 158,403 38 261,308 69 419,711 Totals 66 316,601 60 548,107 126 864,708 (a) As of December 30, 1970. Source: Banco lnteramericano de Desarrolle Undecimo Informe Anual 1870, Washington, 1971. Aggregated amounts for project types, particularly housing, electric power and water and sewerage may be lower than actual due to difficulties in distinguishing between urban and rural investments. (b) This includes local roads, water and sewerage, and some community development technical assistance as integral parts of housing developments. 507 Annex 3 marketing facilities. In the ten years (1961-70) of lending activity, approximately 21% of IDB's capital commitment, some $864.7 mil- lion, has been for the following urban development-related projects. Based on the principle of securing 50% participation by govern- ments in its lending, IDB has been responsible for approximately $800 million investment in housing and related community services in some 16 countries. In addition, through technical assistance it has helped Latin American Governments establish five national housing agencies and reconstruct several other national housing agencies and banks. In the field of housing finance, IDB has developed saving and loan systems, mortgage insurance schemes, and secondary mortgage markets in several Latin American countries. African Development Bank (DBA) and Asian Development Bank (ADB) The other two more recently established regicnal development banks, the African Development Bank and the Asian Development Bank, are also involving themselves in urban development. In 1970, the DBA was financing one water supply and sewerage study ($300,000) and had several other similar projects in the pipeline. The ADB, by the end of 1970, had made three loans for water supply totalling $20.5 million, and a loan of $18 million for a metropolitan f reeway. B. Bilateral Assistance Information collected by the United Nations indicates that bilateral aid for housing, building, and physical planning is about five times that of multilateral aid. Information from this source also indicates that the aggregate expenditure of France, the United Kingdom and the United States for recent years equals about 95% of total bilateral aid in this area of development assistance (see Table 5). Agency for International Development (AID) Most of the US Government aid to developing countries is chan- nelled through the US State Department's Agency for International Development (AID). Information from AID indicates that between 1949 and mid-1970 approximately 4%, some $459.4 million, of its capital commitment has been for the urban development related projects shown in Table 6. AID assistance for housing is conducted mainly through the Office of Housing (OH) and the Latin American Housing and Urban Devel- opment Division of the Office of Development Resources. The OH was established in 1970 to administer the Housing Investment Guar- 508 Annex 3 antee Program (HIG) and to provide other housing services as re- quested by AID regional bureaus. The HIG program accounts for most of present US Government commitment to housing assistance. Its basic purposes are, inter alia, to support in developing countries local institutions which mobilize local savings, specialize in providing long-term housing credit, and to stimulate housing demonstration projects. For this program the US Congress has authorized some $780 million to guarantee the housing investments in developing countries of US private lending institutions such as insurance companies, com- mercial banks, and federally chartered savings and loan associations. Of this amount, $67 million has been committed for Africa, $106 mil- lion for Asia, and about $346 million for Latin America. In spite of the high interest rates (83/4% to 107/8% to the institutional borrower and 11% to 13% to the individual) the demand for the program is said to be growing. Table 5: Bilateral Assistance in International Programs for Housing, Building, and Physical Planning of Selected Countries(a) (Amounts in $ thousands) As Percentage of the Total Average Annual Official Development Amount Assistance Given to Donor Country 1966-1969 Developing Countries (1968) Australia 1,587(b) Austria 53 0.1 Denmark 178 2.1 France 57,645 7.7 Japan 120 0.03 Sweden 1,714 2.0 United Kingdom 31,254 6.1 United States 88,383 2.6 Source: Funds Eupended in International Programs for Housing, Building and Planning (E/C.6/110), UN, New York, 1971 (forthcoming). Information submitted by the governments of donor countries, and L. B. Pearson, Partners in Development. Report of the Commission on International Development. Praeger, N.Y. 1969. (a) Other countries have also provided assistance to developing countries but data are not available. I n comparing the expenditures of the various countries it should be kept in mind that countries differ in their policies toward assisting through bilateral or multilatera! means. For example, of its total development assistance the US channels only 6.4% through multilateral means, and France 4.5%. On the other hand, Denmark channels 48.6% and Sweden 52.8% through multilateral means. (b) Average of 1966-1967 and 1969-1971. Since the creation of the Alliance for Progress in 1960, the Latin America Division has made bilateral loans of approximately $190 million and has made available about $60 million in local currency for housing. AID loans for housing and urban development have generally been on soft or concessional terms which at one time were at such favor- able terms as 3/4 of 1% for forty years but today are at around 3% 509 Table 6: AID Urban Development Assistance, 1949-1970 (Amounts in $ thousands) Total Capital Project Type(a) Capital Grants and Loans Technical Assistance Commitment Completed Projects On-going Projects Total Completed Projects On-going Projects Total No. Pro- No. Pro- No. Pro- No. Pro- No. Pro- No. Pro- No. Pro- jects Amount jects(d) Amount jects Amount jects Amount jects Amount jects Amount jects Amount City and Regional Planning - - - - - - 2 128 1 32 3 160 3 160 Environmental Sanitation 33 54,511 15 63,233 48 117,344 70 28,661 5 1,879 75 30,540 123 147,884 Highways(b) 1 131 8 30,390 9 30,521 - - - - - - 9 30,521 U, Housing(c) 25 71,989 10 87,558 35 159,547 76 11,272 21 16,159 97 27,431 132 186,978 --I CD Project Support for Housing 2 13,698 - - 2 13,698 3 74 - - 3 74 5 13,772 Potable Water 2 1,832 15 66,422 17 68,254 1 141 9 8,719 10 8,860 27 77,114 Urban Transit and Traffic Engineering 1 306 1 2,000 2 2,306 7 315 1 402 8 717 10 3,023 Totals 64 142,067 49 249,603 113 391,670 159 40,591 37 27,191 196 67,782 309 459,452 (a) The categories ot environmental sanitation, housing, and project support for housing may include some assistance to rural areas. (b) Inter-city highways. (c) The housing investment guarantees are not included in this category. (d) As of June 30, 1970. Source: Technical Assistance Completed Projects; Capital Assistance Completed Projects; Capital Assistance Projects, Otfice of the Controller, AID, W253, 6/30/70. w Annex 3 per annum for forty years with a period of grace of ten years and 3% interest during that period. While more than half of AID's loans for housing have been to pub- lic bodies for low-income housing, AID's interest has been on the creation of capital generating institutions for middle income families and on cooperative techniques. Recently, however, serious consider- ation has been given to AID's earlier concern with aided self-help techniques and low-income housing and the use of cooperative tech- niques for that purpose. Financial assistance for "site and services," "core" housing, and "shell" housing will undoubtedly ensue. AID has also made substantial contributions in the form of grants for technical assistance, again primarily in Latin America. The Office of Engineering's urban development unit is presently formulating strategy for coordinating separate infrastructural projects and building complexes to better support development goals and result in more integral assistance. That office also has initiated a study in labor-intensive construction technologies for low-income housing and supporting urban infrastructures. The AID approach to housing and urban development has not been central policy directed. The Urban Development staff of the Bureau for Technical Assistance has been assigned the task of helping the Agency to determine what role, if any, is appropriate for US technical assistance in further support of developing country efforts to address the problems of rapid urban growth. An intensive international survey of urban development problems, approaches, resources, and oppor- tunities has been concluded, and a study is in preparation, in consul- tation with other elements of the Agency, including field missions. A policy decision is anticipated early in fiscal year 1973. Apart from a $29 million urban sector loan to Colombia, capital assistance in urban development has been project specific, focusing on physical infrastructure. However, there are several capital assis- tance projects under review in the Latin America Bureau which incorporate a more comprehensive approach to infrastructural de- velopment, including a technical assistance component. Caisse Centrale de Cooperation Economique (CCCE) CCCE is the disbursement agency for a large portion of the French Government's investment funds in Africa, Indo-China and the French overseas territories and departments. CCCE also works as a delegated agency for aid of the Fonds Europeen Developpement (FED) to Madagascar, the thirteen African countries where CCCE has offices, and the French overseas territories and departments. At this time in- vestment figures are only available for housing. Between 1966 and 511 Annex 3 1969 CCCE lent approximately $28.94 million for housing; $5.58 million in 1969 alone. The four-year aggregated amount aided in the construction of some 27,700 housing units. Commonwealth Development Corporation (CDC) CDC, a Government statutory corporation established in 1948, is an important channel for British development assistance. It is em- powered to undertake, either alone or in association with others, projects for the development of a wide range of economic enter- prises, including agriculture, forestry, fisheries, mining, factories, elec- tricity and water supply, transport, housing, hotels, constructimJn and engineering. The corporation was initially empowered to operate only in developing Commonwealth countries, but since 1969 its op- eration has been extended. The corporation operates under borrow- ing powers of up to $612 million: it borrows principally from the United Kingdom Exchequer. At the end of 1971 it had 210 ongoing projects representing a total estimated commitment of about $496 million. The following types of urban development-related projects accounted for approximately 65% of this amount, or some $324 million. Table 7: CDC Urban Development Aid, 1971 Project Type Capital Commitment (S million) Housing 117 Electricity, Power and Water 121 Transport 19 Industrial and Property Development Companies(a) 67 Total 324 (a) Includes housing and other property development, development of industrial estates and the building of factories tor sale. Source: Commonwealth Development Corporation In the area of housing finance, CDC has aided in establishing 21 mortgage finance companies in African, East Asian and Caribbean countries. These companies finance up to 90% of the value of mort- gage loans. Some of the companies also develop and build housing projects when necessary. There is also a wholly-owned CDC sub- sidiary, Commonwealth Housing Corporation Ltd (CHCL) which com- bines the functions of technical adviser to CDC and to CDC's housing projects and also of acting more generally as consultants on the pro- motion, planning and execution of housing developments in overseas territories. 512 Annex 4 BANK LENDING FOR URBAN PURPOSES Table 1: Bank Lending(') for Urban Purposes as of January 31, 1972 $ million Total Bank and IDA lending 20,000 of which to developing countries 17,300 to developed countries 2,800 Total lending urban-related(21 15,770 of which to developing countries 13,610 to developed countries 2,160 Total lending urban-related excluding inter-urban transport and telecommunications to developing countries 7,920 Urban lending as % total to developing countries 79% Urban lending (excluding inter-urban transport and telecommunications) as % total to developing countries 46% (1) Bank andIDAincluding$220 millionBankloansto IFC butnotthetotalof IFCinvestments($SS50millionatend-1971). Figures are rounded and are net of cancellations, refundings and terminations totalling $477 million of Bank loans and $70 million of IDA credits. (2) Power, Industry, Development Finance Companies, Water Supply and Sewerage, Education and Tourism, Trans- portation and Communication, and IFC operations. Table 2: Urban Projects of the World Bank in Selected Cities Cities Past Lending thru January 31, 1972 $ million Buenos Aires power, water supply, industry 292 Lagos power, port, urban roads, industry, education 180 Sao Paulo power, water supply, industry, port 260 Teheran industry, power, telecommunications 175 Bogota power, water, industry 260 Caracas power, water, telecommunications and urban expressways 170 Bombay (n) industry, power, port 135 Rio de Janeiro power, commuter railway 135 Karachi power, port, industry 128 Calcutta(') power, port, industry 120 Kuala Lumpur power, water, industry, telecommunications 113 Istanbul industry, power, port, education 100 Medellin power, industry 100 TOTAL 2,168 (l) Totals for Calcutta and Bombay are substantially underestimated since no satisfactory basis exists for apportioning to them parts of the SS00 million for industrial imports, the $265 million for development finance companies, the $180 million for telecommunications, and $651 million for Indian Railways (which include commuter lines into these cities). 513 7 ;C o " I I I Economics Although the World Bank generally awards loans for specific projects, its experience has shown that the merits of projects can best be judged within the framework of broad sectors of the economy. Papers included in this volume outline the important aspects of the sectors to which the World Bank directs the majority of its development financing. They offer a unique perspective on the characteristics of the Bank's activities and indicate the breadth of the Bank's lending programs and the philosophy behind them. Each paper describes the distinctive economic, financial, and institutional characteristics of the sector. The papers outline the roles played by each sector in the general process of economic development. They review the scale and approach of the World Bank's operations in each sector. In addition, each paper sum- marizes the Bank's philosophy about how its own operations, as well as those of other lending institutions, can contribute to the physical, financial, and institutional development of individual sectors within its member countries. This volume will be of value not only to those directly concerned with economic development, but also to the economists who teach or do research in development problems. In addition, it provides an otherwise unavailable profile ofthe World Bank at work. Published for the International Bank for Reconstruction and Development THE JOHNS HOPKINS UNIVERSITY PRESS Baltimore, Maryland 21218