PUB8507
Wi
World )evelopment Report 1990
Published for The World Bank
Oxford University Press
Oxford University Press
OXFORD NEW YORK TORONTO DELHI
BOMBAY CALCUTTA MADRAS KARACHI
PETALING JAYA SINGAPORE HONG KONG
TOKYO NAIROBI DAR ES SALAAM
CAPE TOWN MELBOURNE AUCKLAND
and associated companies in
BERLIN IBADAN
© 1990 by the International Bank
for Reconstruction and Development / The World Bank
1818 HStreet, N.W, Washington, D.C. 20433 U.S.A.
First printing June 1990
All rights reserved. No part of this publication may be
reproduced, stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior
permission of Oxford University Press. Manufactured in the
United States of America.
The denominations, the classifications, the boundaries,
and the colors used in maps in World Development Report
do not imply on the part of The World Bank and its
affiliates any judgment on the legal or other status of any
territory, or any endorsement or acceptance of any boundary.
The World Bank does not guarantee the accuracy of the data included in
this publication and accepts no responsibility whatsoever for any
consequences of their use.
ISBN 0-19-520850-1 clothbound
ISBN 0-19-520851-X paperback
ISSN 0163-5085
The Library of Congress has cataloged this serial publication as follows:
World development report. 1978-
[New York] Oxford University Press.
v. 27 cm. annual.
Published for The World Bank.
1. Underdeveloped areasPeriodicals. 2. Economic development
Periodicals. I. International Bank for Reconstruction and Development.
HC59. 7. W659 330.9'172'4 78-6 7086
This book is printed on paper that adheres to
the American National Standard for Permanence of Paper
for Printed Library Materials, Z39.48-1984.
Foreword
This Report is the thirteenth in the annual series often constitute a deprived group even among the
addressing major development issues. Like its poor; in most areas literacy rates and wages are
predecessors, it includes the World Development much lower for women than for men, and access
Indicators, which provide selected social and eco- to social services and employment is more diffi-
nomic data on more than 120 countries. The Re- cult. The plight of poor women is troubling in it-
port addresses the most pressing issue now facing self. It is even more troubling because the health
the development community: how to reduce pov- and education of mothers greatly influence the
erty. Its main conclusions are summarized below. well-being and future of their children.
The world economy enjoyed moderate growth in A review of development experience shows that
the closing years of the decade. But the auspicious the most effective way of achieving rapid and po-
picture was not uniform. The industrial countries litically sustainable improvements in the quality of
saw favorable developments in growth, trade, and life for the poor has been through a two-part strat-
investment. Real per capita incomes grew (and egy. The first element of the strategy is the pursuit
poverty declined) in South Asia and, even more of a pattern of growth that ensures productive use
markedly, in East Asia. But in some countries of of the poor's most abundant assetlabor. The sec-
Latin America and in most of Sub-Saharan Africa, ond element is widespread provision to the poor of
real per capita incomes, living standards, and in- basic social services, especially primary education,
vestment have slipped. For the poor in these coun- primary health care, and family planning. The first
tries, the 1980s was a lost decade. component provides opportunities; the second in-
In 1985 more than one billion people, or almost creases the capacity of the poor to take advantage
one-third of the total population of the developing of these opportunities. The strategy must be com-
world, were living on less than $370 per capita a plemented by well-targeted transfers, to help
year. The percentage of the population living in those not able to benefit from these policies, and
poverty was especially high in South Asia and by safety nets, to protect those who are exposed to
Sub-Saharan Africa. Other aspects of the quality of shocks.
lifealready lower, on average, in developing than Although domestic policy is critical to the reduc-
in developed countriesvaried dramatically tion of poverty, international assistance is needed
among regions, with the bleakest figures for re- to support countries' efforts. Simply increasing re-
gions with the greatest incidence of poverty. For sources, however, will not solve the problem. Aid
example, in 1985 life expectancy was 76 years for is most effective when it complements the recipi-
the developed world but only 50 years for Sub- ents' efforts. The allocation of aid should be more
Saharan Africa and 56 years for South Asia. Some closely linked to a country's commitment to pur-
regions were close to achieving universal enroll- sue development programs geared to the reduc-
ment in primary education, but in Sub-Saharan tion of poverty.
Africa net primary enrollment was only 56 percent, The Report's projections for the 1990s show
and in South Asia it was about 75 percent. Women buoyant growth of about 3 percent a year in the
111
industrial countries and about 5.1 percent in the rica. Nevertheless, the key measures for combat-
developing worldcompared with the 4.3 percent ing poverty are known, and the resources to sup-
achieved by developing countries in the 1980s. If port such an effort are there to be mobilized.
this forecast is correct, prospects for improving the Containing the number of poor in Sub-Saharan Af-
quality of life are bright in most regionsexcept rica until population growth can be brought under
where rapid population growth is expected to off- control and reducing the number elsewhere by 400
set the beneficial effects of economic growth. million are feasible goals for the century's end. The
In East Asia and South Asia the number of poor main obstacle is not the availability of resources
is expected to fall dramatically, and social indica- but the willingness of governments in both devel-
tors should continue to improve. The countries of oping and developed countries to commit them-
Latin America and the Caribbean, Eastern Europe, selves to these goals.
and the Middle East and North Africa will see only Like previous World Development Reports, this
modest reductions in the number of poor, but im- Report is a study by the staff of the World Bank,
provements in social indicators will continue. In and the judgments in it do not necessarily reflect
Sub-Saharan Africa the expected growth in gross the views of the Board of Directors or the govern-
domestic product of 3.7 percent a year, although ments they represent.
significantly higher than in the 1980s, will not be
sufficient to offset the effects of rapid population
growth, and the number of poor will increase.
Even so, adequate provision for the social sectors
should allow a rapid reduction in child mortality
and a reversal of the decline in primary school en- I
( cI\.c4&
rollment experienced in the 1980s. Barber B. Conable
The hurdles to be overcome in decreasing pov- President
erty throughout the world remain formidable. The World Bank
They are especially daunting in Sub-Saharan Af- June 1, 1990
This Report has been prepared by a team led by Lyn Squire and comprising Ehtisham Ahmad,
Robert L. Ayres, Gary Fields, Helena Ribe, Mark Suridberg, Jacques van der Gaag, Dominique van de
Walle, and Michael Walton. The team was assisted by Lara Akinbami, Fernando J. Batista, Robin
Burgess, Elaine K. Chan, Pierre Englebert, Carlos Alberto Herran, Kathryn A. Larin, Natasha
Mukherjee, and Anna-Birgitta Viggh. The work was carried out under the general direction of Stanley
Fischer.
Many others in and outside the Bank provided helpful comments and contributions (see the biblio-
graphical note). The International Economics Department prepared the data and projections pre-
sented in Chapter 1 and the statistical appendix. It is also responsible for the World Development
Indicators. The production staff of the Report included Les Barker, Kathy DahI, Connie Eysenck,
Kenneth Hale, Jeffrey N. Lecksell, Nancy Levine, Hugh Nees, Joyce C. Petruzzelli, Kathy Rosen, Walt
Rosenquist, and Brian J. Svikhart. Library assistance was provided by Iris Anderson. The support
staff was headed by Rhoda Blade-Charest and included Laitan Alli, Trinidad S. Angeles, and Maria
Guadalupe M. Mattheisen. Clive Crook was the principal editor.
Fondly remembered and acknowledged, too, is Pensri Kimpitak (1945-1990), illustrator and graph-
ics designer for the Report since its inception in 1978, whose final illness prevented her participation
in this edition.
iv
Contents
Acronyms and initials ix
Definitions and data notes x
Overview I
Poverty today I
Policies for attacking poverty 2
The politics of poverty 3
Reaching the poor 3
Aid 4
What can be achieved? 4
The challenge for the 1990s 5
Diverging trends in the world economy 7
Recent developments in the world economy 7
The diverging performance of developing countries in the 1980s 10
Prospects for the 1990s 15
2 What do we know about the poor? 24
Three poor families 24
Measuring poverty 25
The characteristics of the poor 29
From diagnosis to treatment 38
3 Progress on poverty: lessons for the future 39
Changes in poverty 39
Factors influencing country performance 45
What is the critical tradeoff? 51
4 Promoting economic opportunities for the poor 56
The pattern of growth and the incomes of the poor 56
Increasing the participation of the poor in growth 64
Reaching resource-poor areas 71
Policies for poverty-reducing growth 73
5 Delivering social services to the poor 74
Social sector policies 74
Investing in people 79
Providing access for the poor 82
Lessons for the next decade 88
V
6 Transfers and safety nets 90
Public provision of transfers and safety nets 91
Food pricing and distribution policies 92
Public employment schemes 97
Caring for those left out 100
The role of transfers and safety nets 101
7 The 1980s: shocks, responses, and the poor 103
Short-run policy and household welfare 103
Macroeconomic policy and the poor 104
Public expenditure restructuring and the poor 116
Lessons for adjustment in the 1990s 120
8 International factors in reducing poverty 121
Trade and poverty 121
Debt and poverty 126
Aid and poverty 127
9 Prospects for the poor 138
Poverty at the end of the century 138
Regional differences in poverty 139
The potential for action 143
Bibliographical note 144
Statistical appendix 159
World Development Indicators 169
Boxes
1.1 Reform in the Eastern European economies 9
1.2 Politics and economic performance 12
1.3 World military expenditures in the 1990s 17
1.4 Mexico's economic prospects after Brady 21
1.5 Going bananas in the European Community, 1992 22
2.1 How should we measure living standards? 26
2.2 Does how we measure poverty really matter? 28
2.3 Village-level perspectives on asset poverty 33
2.4 Moving in and out of poverty in rural India 35
3.1 Development in a Javanese village 41
3.2 Poverty in Africa 42
3.3 The mechanics of changes in poverty 47
3.4 Growth, real wages, and poverty: the United Kingdom and the United States, 1770 to 1920 50
3.5 Political coalitions and the poor 52
3.6 Tradeoffs and poverty: a simple simulation model 53
4.1 Agricultural policies in Brazil favor large farmers 59
4.2 Infrastructural development and rural incomes in Bangladesh 60
4.3 The impact of technological change on women 61
4.4 Does rural-to-urban migration help or hurt the poor? 62
4.5 The merits of selective land reform: Northeast Brazil 65
4.6 Land tenure systems in Sub-Saharan Africa: the case of Rwanda 66
4,7 Learning from informal finance in Sub-Saharan Africa 67
4.8 Developing appropriate technologies for subsistence farmers in Bihar 70
4.9 Groundwater irrigation in South Asia: reaching the landless 71
4.10 Protecting the environment and reducing poverty: China's experience in the Loess Plateau 72
5.1 Adult health: a neglected issue of growing importance 78
5.2 Education and economic growth 80
5.3 Safe motherhood 83
vi
5.4 A community child care and nutrition program in Colombia 85
5.5 Mirzapur, Bangladesh: water supply, sanitation, and hygiene education 86
5.6 Family planning buffaloes and vasectomy festivals in Thailand 88
6.1 The historical antecedents of public action 91
6.2 Reaching the poor through public action 92
6.3 Drought relief in Africa: food interventions in Botswana 97
6.4 At a rural public employment site in India. . . 99
6.5 Drought relief in Africa: public employment in Botswana 100
7.1 Macroeconomic adjustment, household responses, and the role of women: the experience of an
urban community in Ecuador 104
7.2 Macroeconomic policies and the poor in Ecuador: results from a modeling analysis 105
7.3 The poor in Eastern Europe in the years of crisis 108
7.4 Macroeconomic policy choices and income distribution in the Philippines 111
7.5 The effect of macroeconomic adjustment on farmers in Sub-Saharan Africa 113
7.6 The political economy of adjustment 115
7.7 Packaging interventions: the role of social action programs 118
8.1 Industrial country trade policy and the poor in the Dominican Republic 122
8.2 Bangladesh is not spared from MFA quotas 123
8.3 The generalized system of preferences: who benefits and how much? 124
8.4 Foreign aid for development 129
8.5 The World Bank's experience with rural development projects 131
8.6 The World Bank's experience with urban poverty projects 132
8.7 Food aid and poverty 135
Text figures
1 Population and poverty in the developing world, 1985 2
2 Poverty in the developing world, 1985 and 2000 5
1.1 Changes in the distribution of income and population in the developing world 10
1.2 Growth of OECD industrial production and developing country exports, 1970 to 1989 13
1.3 Real export performance of developing countries, 1980 to 1989 13
1.4 Trade balance as share of GDP in severely indebted middle-income countries, 1970 to 1989 14
1.5 Changes in the terms of trade, by developing region, 1980 to 1988 14
1.6 Real interest rates, 1963 to 1989 15
1.7 Real per capita income growth in developing countries in the 1980s and forecast values,
1990 to 2000 23
2.1 Poverty and average standards of living, developed and developing countries 27
2.2 Poverty and infant mortality rates, India, 1983, by urban and rural areas of states 30
2.3 Poverty and illiteracy in India, 1981 32
2.4 Seasonality in rural Bangladesh: poor households in six villages, July 1984 to June 1985 36
3.1 Progress and stagnation in the developing world 40
3.2 The impact of recession on social indicators in developing countries 44
3.3 Trends in malnutrition in the 1980s: Ghana and Lesotho 45
3.4 Inequality and wages, Colombia, 1965 to 1986 49
3.5 Increasing the assets of the poor 54
3.6 Increasing current income 54
4.1 Agricultural purchasing power and rural wages 57
4.2 Direct and indirect taxation of agriculture in the 1970s 57
4.3 Net resource flows between the government and the agricultural sector, Thailand,
1964 to 1982 58
5.1 Trends in health expenditure, number of physicians, and health outcome 76
5.2 Trends in education expenditure, student-teacher ratios, net enrollment rates,
and completion rates 77
6.1 Seasonality and workers employed under the Employment Guarantee Scheme,
Maharashtra, India 98
7.1 Poverty in Indonesia and Malaysia, 1984 and 1987 107
7.2 Fixed investment in Indonesia and Malaysia, 1979 to 1988 108
7.3 Poverty and wages in Brazil and Costa Rica 110
vi'
7.4 Investment pause versus investment crash 111
7.5 Consumption and agricultural purchasing power, Sub-Saharan Africa and Ghana 112
7.6 Agricultural terms of trade, Nigeria and Tanzania 114
7.7 Who gained from public spending on education? Chile, 1974 to 1986 117
7.8 How public employment schemes alleviate unemployment: Chile, 1970 to 1989 119
8.1 Commodity composition of exports, 1988 125
8.2 Sources of developing countries' net aid receipts, 1987 128
Text tables
1.1 Performance indicators in the world economy, 1989 8
1.2 Performance indicators, by developing region, selected periods 11
1.3 Prospects for the 1990s 16
1.4 Comparative indicators for severely indebted low- and middle-income countries 20
2.1 How much poverty is there in the developing countries? The situation in 1985 29
2.2 Rural and urban poverty in the 1980s 31
2.3 Poverty and landholding in Bangladesh, 1978-79 32
3.1 Twenty years of progress in developing countries 39
3.2 Changes in selected indicators of poverty 41
3.3 Changes in poverty in the 1980s 43
3.4 Have social services reached the poor in developing countries? 43
3.5 Variation in country experience 45
3.6 Social sector spending as a percentage of GNP 46
3.7 Poverty, economic growth, and recession 48
4.1 Effective protection and labor intensity of manufacturing, India, 1986 62
4.2 The contribution of manufacturing to employment, selected countries, 1970 to 1980 63
4.3 Diversity in the coverage of credit programs 67
4.4 Grameen Bank loans, by purpose of loan and sex of borrower, 1985 68
5.1 Expenditures on primary education as a percentage of GDP, 1985 and 2000 87
5.2 Projections of social indicators to 2000 89
6.1 Distribution of the annual income transfer from the general food subsidy, Egypt, 1981-82 93
6.2 Distribution of the monthly transfer from food subsidies, Sri Lanka, 1978-79 94
6.3 General and targeted subsidies, Jamaica, 1988 95
7.1 External shocks in the 1980s, by region 107
7.2 Changes in key variables, Indonesia and Malaysia, 1984 and 1987 108
7.3 Changes in key variables, Brazil, 1980 to 1987 110
7.4 Changes in key variables, Ghana, 1980 to 1988 112
7.5 Fiscal contraction and social spending, Sub-Saharan Africa and Latin America 116
8.1 The effect of complete trade liberalization on selected countries 123
8.2 The relative importance of aid in selected developing countries, 1987 127
8.3 Per capita aid receipts, 1988 128
9.1 Social indicators, by developing region, 1985 and 2000 139
9.2 Poverty in 2000, by developing region 139
Statistical appendix tables
A.1 Population growth, 1965 to 1990, and projected to 2000 159
A.2 Population and GNP per capita, 1980, and growth rates, 1965 to 1989 160
A.3 Population and composition of GDP, selected years, 1965 to 1989 160
A.4 GDP, 1980, and growth rates, 1965 to 1989 161
A.5 GDP structure of production, selected years, 1965 to 1988 162
A.6 Sector growth rates, 1965 to 1989 162
A.7 Consumption, investment, and saving, selected years, 1965 to 1988 163
A.8 Growth of export volume, 1965 to 1988 164
A.9 Change in export prices and terms of trade, 1965 to 1988 165
A.10 Growth of long-term debt of low- and middle-income economies, 1970 to 1989 166
A. 11 Investment, saving, and current account balance before official transfers, 1965 to 1988 167
A.12 Composition of debt outstanding, 1970 to 1988 168
vi"
Acronyms and initials
CMEA Council for Mutual Economic Assis- IFAD International Fund for Agricultural
tance (Bulgaria, Czechoslovakia, Ger- Development
man Democratic Republic, Hungary, IFC International Finance Corporation
Poland, Romania, and U.S.S.R.) IFPRI International Food Policy Research
DAC Development Assistance Committee of Institute
the OECD IMF International Monetary Fund
DPi' Diphtheria, pertussis, and tetanus (vac- LIBOR London interbank offered rate
cine) NATO North Atlantic Treaty Organization
EC The European Community (Belgium, NGO Nongovernmental organization
Denmark, Federal Republic of Ger- ODA Official development assistance
many, France, Greece, Ireland, Italy, OECD Organisation for Economic Co-opera-
Luxembourg, Netherlands, Portugal, tion and Development (Australia, Aus-
Spain, and United Kingdom) tria, Belgium, Canada, Denmark, Fin-
ECE Economic Commission for Europe land, France, Federal Republic of Ger-
EFTA European Free Trade Association many, Greece, Iceland, Ireland, Italy,
EGS Employment Guarantee Scheme (Ma- Japan, Luxembourg, Netherlands, New
harashtra State, India) Zealand, Norway, Portugal, Spain,
EPI Expanded Programme on Immuniza- Sweden, Switzerland, Turkey, United
tion Kingdom, and United States)
EPZ Export processing zone OPEC Organization of Petroleum Exporting
ESCAP Economic and Social Commission for Countries (Algeria, Ecuador, Gabon,
Asia and the Pacific Indonesia, Islamic Republic of Iran,
Eurostat European Statistical Office Iraq, Kuwait, Libya, Nigeria, Qatar,
FAO Food and Agriculture Organization Saudi Arabia, United Arab Emirates,
GATT General Agreement on Tariffs and and Venezuela)
Trade PPP Purchasing power parity
GDP Gross domestic product Unesco United Nations Educational, Scientific,
GNP Gross national product and Cultural Organization
GSP Generalized system of preferences UNDP United Nations Development Pro-
G-7 Group of Seven (Canada, France, Fed- gramme
eral Republic of Germany, italy, Japan, UNIDO United Nations Industrial Develop-
United Kingdom, and United States) ment Organization
IBRD International Bank for Reconstruction UNICEF United Nations Children's Fund
and Development USAID United States Agency for International
ICRISAT International Crops Research Institute Development
for the Semi-Arid Tropics WHO World Health Organization
IDA International Development Association WFP World Food Programme
ix
Definitions and data notes
Demographic terms economies were retained; (b) all economies with a
GNP per capita of $6,000 or more were classified as
Total fertility rate. The average number of chil- high-income economies, and (c) the subgroups "oil
dren that would be born alive to a woman during exporters" and "exporters of manufactures" un-
her lifetime if she were to bear children at each age der "developing economies" were dropped. In ad-
in accordance with the prevailing age-specific fer- dition, "high-income oil exporters" is no longer a
tility rates. separate group; "industrial economies" has been
Infant mortality rate. The probability of dying renamed OECD members, which is a subgroup of
between birth and age 1, per thousand births. the new category high-income economies; and a new
Under 5 mortality rate. The probability of dying aggregate, total reporting economies and its subcate-
between birth and age 5, per thousand births. gory oil exporters has been added. As in previous
Life expectancy at birth. The number of years a editions, this Report uses the latest GNP per capita
newborn infant would live if prevailing patterns of estimates to classify countries. The country com-
mortality at the time of its birth were to stay the position of each income group may therefore
same throughout its life. change from one edition to the next. Once the clas-
Net primary enrollment rate. The number of chil- sification is fixed for any edition, all the historical
dren age 6 to 11 enrolled in primary school as a data presented are based on the same country
percentage of the population age 6 to 11, adjusted grouping. The country groups used in this Report
for each country's age structure for primary are defined as follows.
school. Low-income economies are those with a GNP per
capita of $545 or less in 1988.
Country groups Middle-income economies are those with a GNP
per capita of more than $545 but less than $6,000 in
For operational and analytical purposes the World 1988. A further division, at GNP per capita of
Bank classifies economies according to their gross $2,200 in 1988, is made between lower-middle-
national product (GNP) per capita. (Other interna- income and upper-middle-income economies.
tional agencies maintain different classifications of High-income economies are those with a GNP
developing countries; a table describing the classi- per capita of $6,000 or more in 1988.
fications was included in World Development Report Although the dividing line between low and
1989.) middle income used in the World Development Re-
Country classifications were revised in the 1989 port has always been a specific level of GNP per
edition of the World Development Report and its sta- capita, editions prior to 1989 were ambiguous
tistical annex, the World Development Indicators. about the line between middle income and high
The principal changes were: (a) the "developing income. Industrial market economies and high-
economies" group was dropped, but references to income oil exporters were shown separately, but
the specific income groups low- and middle-income some economies remained in the middle-income
x
group despite having a GNP per capita above that Côte d'Ivoire, Ecuador, Honduras, Hungary, Mex-
of some countries classffied as high-income. The ico, Morocco, Nicaragua, Peru, Philippines, Po-
high-income cutoff point of $6,000 removes that land, Senegal, Uruguay, and Venezuela. The 1989
anomaly. edition used the category "seventeen highly in-
Low-income and middle-income economies are debted economies," which did not include the
sometimes referred to as developing economies. People's Republic of the Congo, Honduras, Hun-
The use of the term is convenient; it is not in- gary, Nicaragua, Poland, and Senegal and did in-
tended to imply that all economies in the group are clude Colombia, Jamaica, Nigeria, and Yugoslavia.
experiencing similar development or that other OECD members, a subgroup of high-income
economies have reached a preferred or final stage economies, comprises the members of the Organi-
of development. Classification by income does not sation for Economic Co-operation and Develop-
necessarily reflect development status. (As in last ment except for Greece, Portugal, and Turkey,
year's edition of the World Development Indica- which are included among the middle-income
tors, high-income economies classified by the economies.
United Nations or otherwise regarded by their au-
thorities as developing are identified by the symbol Geographic regions (low-income
t.) The use of the term "countries" to refer to and middle-income economies)
economies implies no judgment by the Bank about
Sub-Saharan Africa comprises all countries
the legal or other status of a territory.
south of the Sahara except South Africa.
Non reporting nonmembers are Albania, Bulgaria,
Europe, Middle East, and North Africa comprises
Cuba, Czechoslovakia, German Democratic Re-
eight European countriesCyprus, Greece, Hun-
public, Democratic People's Republic of Korea, gary, Malta, Poland, Portugal, Romania, and
Mongolia, Namibia, and the Union of Soviet So- Yugoslaviaall the economies of North Africa and
cialist Republics. In the main tables of the World
the Middle East, and Afghanistan. For some analy-
Development Indicators, only aggregates are ses in the World Development Report separate group-
shown for this group, but Box A.2 contains key ings are used for Eastern Europe and for Middle
indicators reported for each of these countries. East and North Africa (or Middle East, North Af-
rica, and other Europe, depending on the database
Analytical groups used). Eastern Europe comprises the former cen-
trally planned economies in the group (Hungary,
For analytical purposes, other overlapping classifi- Poland, and Romania) and Yugoslavia.
cations based predominantly on exports or exter- East Asia comprises all the low-income and
nal debt are used in addition to geographic country middle-income economies of East and Southeast
groups. The lists provided below are of economies Asia and the Pacific, east of and including China
in these groups that have populations of more and Thailand.
than 1 million. Countries with less than 1 million South Asia comprises Bangladesh, Bhutan,
population, although not shown separately, are in- India, Myanmar, Nepal, Pakistan, and Sri Lanka.
cluded in group aggregates. Latin America and the Caribbean comprises all
Oil exporters are countries for which exports of American and Caribbean countries south of the
petroleum and gas, including reexports, account United States.
for at least 30 percent of merchandise exports.
They are Algeria, Bahrain, Brunei, Cameroon, Data notes
People's Republic of the Congo, Ecuador, Arab Re-
public of Egypt, Gabon, Indonesia, Islamic Repub- Billion is 1,000 million.
lic of Iran, Iraq, Kuwait, Libya, Mexico, Nigeria, Trillion is 1,000 billion.
Norway, Oman, Qatar, Saudi Arabia, Syrian Arab Tons are metric tons. equal to 1,000 kilograms,
Republic, Trinidad and Tobago, United Arab Emir- or 2,204.6 pounds.
ates, and Venezuela. Dollars are current U.S. dollars unless other-
Severely indebted middle-income countries (abbre- wise specified.
viated to "Severely indebted" in the World Devel- Growth rates are based on constant price data
opment Indicators) are nineteen countries that are and, unless otherwise noted, have been computed
deemed to have encountered severe debt-servicing with the use of the least-squares method. See the
difficulties. They are Argentina, Bolivia, Brazil, technical notes of the World Development Indica-
Chile, People's Republic of the Congo, Costa Rica, tors for details of this method.
xi
The symbol. . in tables means not available. able after 1987. Historical data in this Report may
The symbol - in tables means not applicable. differ from those in previous editions because of
The number 0 or 0.0 in tables means zero or a continuous updating as better data become avail-
quantity less than half the unit shown and not able and because of new group aggregation tech-
known more precisely. niques that use broader country coverage than in
All tables and figures are based on World Bank previous editions.
data unless otherwise specified. The cutoff date for Economic and demographic terms are defined in
all data in the World Development Indicators is the technical notes to the World Development In-
April 30, 1990. dicators.
Data for secondary sources are not always avail-
xl'
Overview
During the past three decades the developing poorest of the world's poor. It seeks first to mea-
world has made enormous economic progress. sure poverty, qualitatively as well as quantita-
This can be seen most clearly in the rising trend for tively. It then tries to draw lessons for policy from
incomes and consumption: between 1965 and 1985 the experience of countries that have succeeded in
consumption per capita in the developing world reducing poverty. It ends with a question that is
went up by almost 70 percent. Broader measures also a challenge: what might be achieved if govern-
of well-being confirm this picturelife expectancy, ments in rich and poor countries alLike made it their
child mortality, and educational attainment have goal to attack poverty in this closing decade of the
all improved markedly. Viewed from either twentieth century?
perspectiveincome and consumption on the one
hand, broad social indicators on the otherthe de- Poverty today
veloping countries are advancing much faster than
today's developed countries did at a comparable In the countries that have participated in the over-
stage. all economic progress that has taken place since
Against that background of achievement, it is all the 1960s, poverty has declined and the incomes
the more staggeringand all the more shameful even of those remaining in poverty have in-
that more than one billion people in the develop- creased. In some cases this change has been dra-
ing world are living in poverty. World Development matic. Indonesia, for example, took less than a
Report 1990 estimates that this is the number of generation in the 1970s and 1980s to reduce the
people who are struggling to survive on less than incidence of poverty from almost 0 percent of the
$370 a year. Progress in raising average incomes, population to less than 20 percent. On a variety of
however welcome, must not distract attention social indicators, some developing countries are
from this massive and continuing burden of now approaching the standards of the developed
poverty. world. In China, which accounts for a quarter of
The same is true of the broader measures of well- the developing world's people, life expectancy
being. Life expectancy in Sub-Saharan Africa is reached 69 in 1985. But in many countries eco-
just 50 years; in Japan it is almost 80. Mortality nomic performance was weaker, and the number
among children under 5 in South Asia exceeds 170 in poverty fell more slowly. Where rapid popula-
deaths per thousand; in Sweden it is fewer than tion growth was an important additional factor, as
10. More than 110 million children in the develop- in much of Sub-Saharan Africa, consumption per
ing world lack access even to primary education; in head stagnated and the number in poverty rose.
the industrial countries anything less than univer- The 1980soften called a "lost decade" for the
sal enrollment would rightly be regarded as unac- poordid not, in fact, reverse the overall trend of
ceptable. The starkness of these contrasts attests to progress. The incomes of most of the world's poor
the continuing toll of human deprivation. went on rising, and under 5 mortality, primary
This Report is about poverty in the developing school enrollment ratios, and other social indica-
worldin other words, it is concerned with the tors also continued to improve. The setbacks of the
1
and the pressure of rapidly expanding populations
Figure 1 Population and poverty in the mean that many of the poor live in areas of acute
developing world, 1985 environmental degradation.
The weight of poverty falls most heavily on cer-
tain groups. Women in general are disadvantaged.
Percentage of In poor households they often shoulder more of
population
the workload than men, are less educated, and
have less access to remunerative activities. Chil-
dren, too, suffer disproportionately, and the future
quality of their lives is compromised by inadequate
nutrition, health care, and education. This is espe-
cially true for girls: their primary enrollment rates
are less than 50 percent in many African countries.
The incidence of poverty is often high among eth-
nic groups and minorities such as the indigenous
Percentage of
poor peoples in Bolivia, Ecuador, Guatemala, Mexico,
and Peru and the scheduled castes in India.
In many but not all cases low incomes go hand in
hand with other forms of deprivation. In Mexico,
for example, life expectancy for the poorest 10 per-
cent of the population is twenty years less than for
the richest 10 percent. In Côte d'Ivoire the primary
enrollment rate of the poorest fifth is half that of
the richest. National and regional averages, often
0 Sub-Saharan Africa bad enough in themselves, mask appallingly low
o East Asia life expectancy and educational attainment among
o South Asia the poorest members of society.
o Europe, Middle East,
and North Africa
Q Latin America and Policies for attacking poverty
the Caribbean
In the 1950s and 1960s many saw growth as the
primary means of reducing poverty and improving
the quality of life. For example, the Indian Plan-
1980s fell heavily on particular regions. For many ning Commission viewed rapid growth as the
in Sub-Saharan Africa and Latin America incomes main (although not the only) instrument for
fell during the decade, and the incidence of pov- achieving this objective. In the 1970s attention
erty increasedalthough the social indicators, at shifted to the direct provision of health, nutri-
least in Latin America, proved somewhat more tional, and educational services. This was seen as a
resilient. matter for public policy. World Development Report
The burden of poverty is spread unevenly 1980, marshaling the evidence available at the
among the regions of the developing world, time, argued that improvements in the health, ed-
among countries within those regions, and among ucation, and nutrition of the poor were important
localities within those countries. Nearly half of the not only in their own right but also to promote
world's poor live in South Asia, a region that ac- growth in incomes, including the incomes of the
counts for roughly 30 percent of the world's popu- poor.
lation (Figure 1). Sub-Saharan Africa accounts for a The 1980s saw another shift in emphasis. Coun-
smaller, but still highly disproportionate, share of tries, especially in Latin America and Sub-Saharan
global poverty. Within regions and countries, the Africa, struggled to adjust after the global reces-
poor are often concentrated in certain places: in sion. The constraints on public spending tight-
rural areas with high population densities, such as ened. At the same time, many began to question
the Gangetic Plain of India and the island of Java, the effectiveness of public policy, and especially
Indonesia, or in resource-poor areas such as the policy toward the poor. Against this background,
Andean highlands and the Sahel. Often the prob- World Development Report 1990 reexamines how pol-
lems of poverty, population, and the environment icy can help to reduce poverty and explores the
are intertwined: earlier patterns of development prospects for the poor during the 1990s.
2
The evidence in this Report suggests that rapid ing countries a period of painful macroeconomic
and politically sustainable progress on poverty has adjustment was unavoidable. In the longer term
been achieved by pursuing a strategy that has two the economic restructuring associated with adjust-
equally important elements. The first element is to ment is perfectly consistent with the two-part
promote the productive use of the poor's most strategy. In the short term, however, many of the
abundant assetlabor. It calls for policies that har- poor are at risk. During the transition period the
ness market incentives, social and political institu- poor can be protected through a judicious mix of
tions, infrastructure, and technology to that end. macroeconomic policies (for example, pricing pol-
The second is to provide basic social services to the icy reforms that benefit poor farmers) and mea-
poor. Primary health care, family planning, nutri- sures to moderate declines in private consump-
tion, and primary education are especially tion. Experience also shows that it is possible to
important. shift public spending in favor of the poor, even
The two elements are mutually reinforcing; one within an overall framework of fiscal discipline,
without the other is not sufficient. In some coun- and to target transfers more accurately. In addi-
tries, such as Brazil and Pakistan, growth has tion, increased capital inflows can be used to help
raised the incomes of the poor, but social services cushion the impact of adjustment on the poor.
have received too little attention. As a result, mor-
tality among children remains unusually high and The politics of poverty
primary enrollment unusually low, and the poor
The framework of political and economic institu-
are not as well equipped as they might be to take
tions is important because policies to reduce pov-
advantage of economic opportunities. Some other
erty involve a tradeoff. This tradeoff is not, in the
countries, by contrast, have long stressed the pro-
main, between growth and the reduction of pov-
vision of social services, but growth has been too
erty. Switching to an efficient, labor-intensive pat-
slow. In Sri Lanka, for example, primary enroll- tern of development and investing more in the hu-
ment rates and under 5 mortality rates are excep-
man capital of the poor are not only consistent
tionally good, but the potential for raising the in-
with faster long-term growth; they contribute to it.
comes of the poor has gone to waste for lack of Since these actions mean that a larger share of in-
economic opportunity. come and more public spending will go to the
Progress has been greatest in the countries that
poor, the principal tradeoff, especially in the short
have implemented both parts of the strategy. By
run, is between the interests of the poor and those
promoting the productive use of labor these coun-
of the nonpoor. The two-part strategy is, therefore,
tries have furnished opportunities for the poor, more likely to be adopted in countries where the
and by investing in health and education they poor have a say in political and economic decision-
have enabled the poor to take full advantage of the
making.
new possibilities. In Indonesia and Malaysia this Although the two-part strategy does involve a
approach has brought about a substantial reduc- politically sensitive tradeoff between the poor and
tion in poverty along with rapid improvements in the nonpoor, it is likely to prove more feasible than
nutrition, under 5 mortality, and primary other strategies. Large-scale redistributions of land
enrollment. have sometimes been successful. l.n Japan and the
Even if this basic two-part strategy is adopted, Republic of Korea, for example, land redistribution
many of the world's poorthe sick, the old, those was central to the reduction of rural poverty and
who live in resource-poor regions, and others laid the basis for the other policies advocated in
will continue to experience severe deprivation. this Report. Where it can be done, redistribution of
Many others will suffer temporary setbacks owing land should be strongly supported. But the politi-
to seasonal variations in income, loss of the family cal obstacles to such reform are great. In most
breadwinner, famine, or adverse macroeconomic countries the two-part strategy outlined here,
shocks. A corprehensive approach to poverty re- which sees investment in education as the best
duction, therefore, calls for a program of well- way of augmenting the assets of the poor, is more
targeted transfers and safety nets as an essential likely to succeed.
complement to the basic strategy.
During the 1980s many developing countries Reaching the poor
had to cope with macroeconomic crises. Their ex-
perience drew attention to a new concern: the Even when macroeconomic adjustment is not a
need to frame adjustment policies that give due primary issue, the strategy requires an increase in
weight to the needs of the poor. In many develop- certain categories of public spending that specffi-
3
cally benefit the poor. If these are to be affordable Treaty Organization would pay for a doubling of
and hence sustainable, they must be cost-effective. aid. The resources can be made availableal-
Experience since the 1970s shows, however, that though little will be achieved unless they are used
reaching the poor with targeted programs can be effectively.
difficult. Nongovernmental organizations have The analysis in this Report provides the basis for
made important contributions here. Self-selecting a better aid strategy. External assistance should be
programs, which exclude the nonpoor by offering more tightly linked to an assessment of the efforts
benefits that are of interest only to the poor, are that would-be recipients are making to reduce pov-
another promising approach. Low-wage public erty. This principle already underlies procedures
employment programs, for instance, have pro- for allocating the resources of the International De-
vided an effective safety net for the poor in certain velopment Association (IDA). Carrying out this
parts of South Asia and have been especially valu- principle would mean that countries committed to
able in preventing famine. Chile's experience sug- the two-part strategy would be the main recipients
gests that such schemes may also be helpful dur- of aid. This reflects the conviction that aid works
ing recessions. well only when it complements a sound develop-
To be truly cost-effective, interventions must be ment strategy.
not merely well targeted but also carefully de- In countries where policies (on prices and public
signed to meet the specific needs of poor people. spending, for instance) are inconsistent with ef-
This means developing technologies suited to the forts to reduce poverty, external resources would
risky environment that confronts small farmers, achieve far less. Yet there are many poor people in
devising credit schemes to serve small borrowers, such countries. Indeed, these are the very coun-
combining feeding programs for especially vulner- tries in which poverty is going to get worse. The
able groups with education on health and nutri- judgments that have to be made in such cases will
tion, and so on. Successful programs have usually be extremely difficult. Aiming moderate quantities
involved the poor both at the design stage and of aid directly at highly vulnerable groups seems
during implementation. the appropriate response. Health clinics that serve
Public spending that is well designed and accu- the poor, immunization programs for children,
rately targeted can play an important part in the and targeted nutrition programs are the sorts of
fight against poverty. But such programs, however intervention that might be supported by the aid
cost-effective, are no substitute for efforts to attune community in such circumstances.
the broad stance of economic policy to the needs of Many countries will fall between these two ex-
the poor. Attacking poverty is not primarily a task tremes. In such cases, intermediate amounts of as-
for narrowly focused antipoverty projects, vital sistance would be appropriate. Careful judgment
though these may be. It is a task for economic pol- is needed to determine how this aid can best be
icy in the large. used to make policy more responsive to the needs
of the poor.
Aid These principles certainly bear on the operations
of the World Bank, but they should be regarded as
Aid has often been an effective instrument for re- applicable to the aid community as a whole. If the
ducing povertybut not always. Donors some- aid strategy outlined here were adopted and fol-
times have other objectives. In 1988 about 41 per- lowed consistently by bilateral donors, nongovern-
cent of external assistance was directed to middle- mental organizations, and multinational agencies,
and high-income countries, largely for political its effectiveness would be greatly increased.
reasons. Even when aid has been directed to the
poor, the results have sometimes been disappoint- What can be achieved?
ingespecially in countries in which the overall
policy framework has not been conducive to the The Report projects that if, over the next ten years,
reduction of poverty. the industrial countries grow at about 3 percent a
The world is at a turning point: the geopolitical year (the historical trend) and real aid flows in-
tensions that have prevailed since World War II are crease at a similar rate, per capita incomes in the
easing rapidly. This offers a unique opportunity to developing countries will grow by between 5.1
cut military spending and increase international percent a year in East Asia and 0.5 percent a year
assistance. A cut of just 10 percent in military in Sub-Saharan Africa. With some progress to-
spending by the countries of the North Atlantic ward adopting the strategy advocated by the Re-
4
port, the number of poor in the developing world
could fallcounting gains and setbacksby more Figure 2 Poverty in the developing world,
than 300 million by the end of the century. More 1985 and 2000
determined efforts to reduce poverty could see
even greater progress.
This relatively favorable outcome hides marked Number of poor (millions)
0 100 200 300 400 500 600
differences among regions (Figure 2). The number
of people in poverty would fall most sharply in I I I
East Asia. Progress would be good in South Asia,
modest in Latin America and the Caribbean, and South Asia
negligible in developing Europe, the Middle East,
and North Africa. Compared with a reduction of
400 million in poverty elsewhere in the developing
world, in Sub-Saharan Africa slow economic Sub-Saharan
Africa
growth and rapid population growth would mean
an increase of nearly 100 million in the number of
poor people. By the end of the century Sub-
Saharan Africa will account for more than 30 per-
East Asia
cent of the developing world's poor, as against 16
percent in 1985.
Progress in other aspects of living standards is
expected to continue. By 2000 mortality for chil- Europe,
Middle East and
dren under 5 will range from about 30 deaths per North Africa
thousand in East Asia to a still-high 135 per thou-
sand in Sub-Saharan Africa. Most regions will be
able to achieve universal primary school enroll- Latin America
ment with, at most, a modest increase in the share and
the Caribbean
of gross domestic product (GDP) allocated for this
purpose. Again, Sub-Saharan Africa is the excep-
tion. Universal primary enrollment can be
achieved by 2000, but it will cost more than 2.5 o 1985 0 2000
percent of GDP, compared with current spending Equals 50 miffion
of less than 1.5 percent.
Progress in Asia depends critically on develop-
ments in China and India. If economic reform fal-
ters in China or if India is unable to maintain its
recent momentum, the prospects for reducing to pose further obstacles to the reduction of pov-
poverty will be far less bright. Alternative projec- erty. Trends in regional growth would then di-
tions based on slower growth and policies less fa- verge more noticeably. Although progress in East
vorable to the poor show an additional 150 million and South Asiathe regions with the greatest
poor, mainly in India. number of poor--would be relatively unaffected,
The plight of Sub-Saharari Africa is unfortu- Latin America would face a serious setback, and an
nately all too clear. Even to hold the number of already intolerable situation in Sub-Saharan Africa
poor at the 1985 level will require a massive effort: would deteriorate still further.
GDP growth of 5.5 percent a year, a radical restruc-
turing of industry, improved incentives and tech- The challenge for the 1990s
nology for agriculture, and increased allocations of
resources to primary education, health care, nutri- No task should command a higher priority for the
tion, and family planning. This cannot be achieved world's policymakers than that of reducing global
unless governments strengthen their reform ef- poverty. In the last decade of the twentieth century
forts and donors increase their assistance. it remains a problem of staggering dimensions.
Slower growth in the industrial countries, higher Despite the difficulties, which are especially daunt-
interest rates, and a smaller-than-expected rise in ing in Sub-Saharan Africa, the means for reducing
the terms of trade could combine, as in the 1980s, poverty are at hand. This Report argues that the
5
principal elements of an effective strategy are well Containing the number of poor in Africa and re-
understood and that the external resources needed ducing the number elsewhere by almost 400 mil-
to support it could be made available at little cost to lion (compared with 1985) would be a substantial
the industrial countries. Where developing coun- achievement. It is an ambitious goalbut one that
tries are committed to reducing poverty, the indus- is within reach by the end of the century.
trial countries should respond with increased
assistance.
6
Diverging trends
in the world economy
Looking back on the 1980s, much of the world can however, this first chapter begins by examining re-
count itself fortunate. One after another, great cent developments in the world economy and the
dangers appeared, only to recede without serious prospects for the 1990s. Just as the external envi-
repercussions. The international debt crisis no ronment of recent years goes some way toward
longer threatens the integrity of the world's finan- explaining the disappointing performance of many
cial system; the recession of the early 1980s, which developing countries, so too the economic outlook
drove unemployment in Europe to record highs, describes the foundation on which future efforts to
gave way to one of the longest peacetime expan- attack poverty will have to be built. It will be far
sions of this century; the stock market crash of easier to reduce poverty if the platform is one of
October 1987 failed to cause a second Great De- low inflation, lower real interest rates, and open
pression. It was a turbulent period, to be sure. But trade than if fluctuating prices, high real interest
by the standard of the fears expressed at the time, rates, and restricted trade prevail. As always,
it was a decade of disasters that never happened. progress in the developing countries is closely
Aptly enough, it closed with momentous changes bound up with the policies of the industrial coun-
in several Eastern European countrieschanges tries.
that mark the beginning of a new and uniquely
promising era in the history of the world. Recent developments in the world economy
Yet millions of the world's most vulnerable peo-
ple must take a far gloomier view of the past ten The 1980s closed happily for the industrial
years. Many developing countries have not merely countriesgrowth was moderate to high, output
failed to keep pace with the industrial countries; was at or near potential, unemployment was well
they have seen their incomes fall in absolute terms. down from the levels seen earlier in the decade,
The living standards of millions in Latin America inflation was under control, and world trade was
are now lower than in the early 1970s. In most of expanding strongly. In the principal industrial
Sub-Saharan Africa living standards have fallen to countries productivity growth accelerated in the
levels last seen in the 1960s. Such facts, extraordi- late 1980s, and investment grew nearly twice as
nary as they are, fail to capture the plight of the fast as GDP. Concern over inflation remained, but
very poorest, whose lives have remained blighted restrictive monetary policies appeared to be keep-
even as incomes elsewhere in the developing ing the pressures in check. Commodity prices
world have risen. For many of the world's poor, were fairly stable, although petroleum prices rose
the 1980s was a 'lost decade"a disaster indeed. by an average of 20 percent over their 1988 level.
This Report is about the poor. It is thus about the Despite a healthy growth of 3.6 percent in the
fundamental issue in economic development: the industrial countries, external imbalances were
eradication of poverty from the world. Later chap- slow to narrow. The United States finished the
ters will look in greater detail at policies that offer year with a current account deficit of $106 billion,
hope for reducing poverty. As in previous Reports, down by $20 billion from the previous year. The
7
Table 1.1 Performance indicators in the world economy, 1989
Real growth Growth of export Gross domestic
of GDP volume investment/GDP
Group and region 1980-89 1989 1980-89 1989 1980-89 1989
Industrial countries 3.0 3.6 4.8 7.6 20.9 21.5
Developing countries 4.3 3.3 6.1 8.1 24.3 24.6
Sub-Saharan Africa 1.0 3.5 0.0 10.1 16.1 15.2
East Asia 8.4 5.1 14.7 8.1 30.0 30.7
South Asia 5.5 4.8 6.1 9.6 22.3 21.4
Eastern Europeb 1.4 0.0 3.8 2.0 29.4 24.8
Middle East, North Africa,
and other Europe 2.9 2.5 64b 14b 25.9 24.1
Latin America and
the Caribbean 1.6 1.5 4.9 4.4 20.1 20.6
Data for 1989 are preliminary.
Estimates.
Japanese current account surplus fell by more than has fallen to less than 10 percent of GDP-a level
27 percent, to $58 billion, but in the Federal Repub- that is insufficient even to replace worn-out
lic of Germany the surplus rose by more than 14 capital.
percent, to $56 billion, and approached that of Ja- In 1989 Latin America saw no recovery from the
pan for the first time in recent history. Although weak growth and falling per capita incomes that
concern over the willingness of capital markets to had characterized the rest of the decade. Average
finance the U.S. current account deficit has eased, income growth was 1.5 percent, but differences
saving in the United States and other industrial within the region were great. Brazil achieved 3.5
countries remains low. As a result, these econo- percent real growth, whereas the rest of Latin
mies continue to absorb a large share of the global America averaged just 0.2 percent. Debt continues
supply of capital, and this in turn contributes to to be a major obstacle to growth: net outward
high world interest rates. transfers of resources to creditors amounted to al-
In 1989 the developing countries saw growth most one-fifth of export revenues. A reappraisal of
slow to 3.3 percent, as against an average for the strategies for handling the debt crisis led to several
decade of 4.3 percent (Table 1.1). Growth was proposals for a new approach in 1989. The Brady
strongest in South Asia and East Asia-the regions initiative, announced in March 1989, calls for case-
with the world's largest concentration of poor by-case debt reduction accompanied by official fi-
people-although it was slower in both cases than nancial support that is conditional on domestic
the average for recent years. Despite strong export policy reform. Recognition of the need for debt
performance, South Asia's growth fell to 4.8 per- reduction and for full participation both by official
cent from 9 percent in 1988. Incomes in East Asia agencies and by private creditors marks an impor-
rose by a healthy 5.1 percent, but that followed a tant departure from previous debt strategies.
10 percent rise in 1988. China's growth fell to 3.9 Several agreements providing for commercial
percent, less than half the average for the decade, debt reduction have already been concluded under
as the pace of economic reform slowed and the the broad guidelines of the Brady initiative. These
government adopted austerity measures to curb agreements, which are backed with financial sup-
inflation by controlling the expansion of domestic port from the World Bank, the International Mone-
credit. tary Fund (IMF), and the Japanese government,
Elsewhere, growth was less buoyant. Real ex- vary in their structure and terms. Mexico's agree-
port growth of 10 percent helped boost growth in ment, for example, covers 85 percent of its com-
Sub-Saharan Africa to 3.5 percent in 1989, one of mercial bank debt. The Philippines bought back a
the decade's strongest years. But because popula- portion of its commercial bank debt at a 50 percent
tion was growing extraordinarily fast, average per discount and also received some new loans. Costa
capita income scarcely increased. Some countries, Rica's existing commercial bank claims were ex-
including Burkina Faso, Ghana, Mali, and Mauri- changed for new low-interest bonds. The results
tius, managed growth of more than 5 percent in so far indicate that in addition to reducing the debt
1989, but investment remains severely depressed burden these agreements will encourage the coun-
across the region. In some countries investment tries concerned to strengthen their adjustment pol-
8
Box 1.1 Reform in the Eastern European economies
The countries of Eastern Europe face the task of trans- planning was abolished and the state began making
forming their command economies into decentralized greater use of taxes, subsidies, and price controls to
and more market-oriented systems. The long-term guide the economy indirectly. During the 1970s and
gains are likely to be great, but in the short term the early 1980s progress was slow. In 1985-86, however,
reformers may face steep transitional costs. Different new impetus for reform came after the stabilization ef-
countries have already adopted their own distinctive forts of 1982-84 had failed to address the underlying
approaches. Poland has decided to "cross the chasm in structural problems. Wage regulations were made
one leap," whereas in Hungary reforms have been more flexible, enterprises were given greater autonomy
more cautious and gradual. in setting prices, and foreign trading rights were ex-
In October 1989 the Polish government announced a panded. The reforms failed, however, mainly because
far-reaching plan to first stabilize the economy and of lax monetary and fiscal management. Wage in-
then move quickly to a market-based system. The goals creases were allowed to outstrip productivity, and the
of stabilization are to reduce inflationwhich ran at an state incurred large losses. After 1987 the government
annualized rate of 650 percent during 1989and to again tried to impose fiscal and monetary discipline
eliminate the government deficit. Inflation is to be held and to stimulate nonruble exports. Although trade per-
down by decreasing real wages, stabilizing food prices, formance improved, efforts to curb the fiscal deficit
and increasing interest rates to reduce the demand for have been disappointing.
credit. But these measures will not succeed unless the The choice of gradual versus rapid reform is also a
fiscal deficit is curbed. The plan calls for massive sales choice between two sets of risks. Rapid reform is likely
of state enterprises, closure of inefficient plants, cuts in to lead to greater dislocation in the short run, whereas
price subsidies on food and domestic energy, and cut- slow reform often creates inconsistencies that thwart
backs in spending on defense and public administra- further progress. In Poland a substantial share of the
tion. It looks toward an ownership system modeled on work force may become unemployed under the
that of Western industrial countries. Freedom to estab- planned restructuring and sale of state enterprises. The
lish enterprises is to be codified and restrictions on social safety nets being put in place may prove both
housing rentals and sales removed. Some of these costly and inadequate. Hungary took the other path;
measures will fall heavily on low-income families and the government freed many prices and decentralized
those most dependent on the state. the economy, but large fiscal deficits led in the end to
Hungary's reforms began in 1968, when rigid central the failure of the pre-1990 reform efforts.
icies. This should help to restore the confidence of Expansion of external economic relations may
domestic and foreign investors. also pose difficulties. More than half the exports of
The year 1989 was a historic one for Eastern Eu- the members of the Council for Mutual Economic
rope. Many countries in the region saw nothing Assistance (CMEA)Bulgaria, Czechoslovakia,
less than a political revolution, although a largely the German Democratic Republic, Hungary, Po-
peaceful one, and the pace of change in economic land, Romania, and the U.S.S.R.are within the
policy accelerated everywhere (Box 1.1). Suddenly region. The adoption of outward-oriented market
there has emerged an enormous potential for rais- reforms will require restructuring of the CMEA
ing industrial productivity, expanding technologi- and dismantling of the current accounting prac-
cal exchange and trade relations, and thereby tices based on "transferable rubles." Many of
boosting incomes in the region. these countries' manufactured exports are unlikely
In the short and medium run, however, adjust- to be competitive in outside markets.
ment will probably involve significant costs. Aver- In other developing Europe, the Middle East,
age regional incomes failed to grow in 1989, and and North Africa the decade ended with slower
Poland's GDP is estimated to have fallen by 1 per- growth. On average, income in the region grew by
cent, with a sharper decrease likely in 1990. Ram- 2.5 percent in 1989, but this implied no growth in
pant open inflation has occurred recently in Poland per capita incomes. Weak export performance
and Yugoslavia, but both countries have had early made foreign exchange scarce and curbed invest-
success after implementing tough anti-inflationary ment. Turkey and several countries in North Africa
policies. Elsewhere in the region there are many are finding that debt service is absorbing a large
signs of repressed inflation. share of their export earnings.
9
share is rising. In South Asia fertility rates have
been successfully reduced since the 1960s, and in-
Figure 1.1 Changes in the distribution of income
and population in the developing world come growth has been boosted over the past de-
(percentage share) cade. Although South Asia still harbors the largest
number of poor, during the past twenty-five years
Sub-Saharan Africa's share of the world's poor has
Income Population grown and incomes in Latin America have gradu-
Average, 1960-65 ally deteriorated.
Why have some regions performed so much bet-
ter than others? In the end, the battle against pov-
erty depends on the answer to this question. Not
surprisingly, trends in poverty during the 1980s
reflect trends in overall economic performance. Al-
though data are scarce, the evidence shows that
$801 billion 2.1 billion where economic performance was good, poverty
decreased. Thus, in Asia poverty declined in India,
Average 1988-89 Indonesia, Malaysia, and Pakistan. In many coun-
tries of Sub-Saharan Africa, Latin America, and
Eastern Europe, however, external and internal
shocks caused poverty to increase. In all the coun-
tries in these regions for which data are available
Brazil, Colombia, Costa Rica, Côte d'Ivoire, Po-
land, Venezuela, and Yugoslaviapoverty in-
creased during at least part of the 1980s.
This performance gap should not be regarded as
an unchangeable fact of life; per capita incomes did
grow almost everywhere during the 1960s. But the
$3,000 billion 3.7 billion regions began to diverge in the 1970s. By the 1980s
per capita GDP was growing at 6.7 percent in East
o Sub-Saharan Africa 0 East Asia 0 South Asia
Asia and 3.2 percent in South Asia but was falling
[J Europe, Middle East, and North Africa
(excluding Eastern Europe) in both Sub-Saharan Africa and Latin America (Ta-
O Latin America and the Caribbean ble 1.2). These regional differences found their
counterparts in investment. Both Asian regions in-
Note: Income is in real 1980 dollars.
creased their national saving and investment rates
during these periods, in contrast to the declines
experienced in Sub-Saharan Africa and Latin
The diverging performance of developing America. Domestic policies and external economic
countries in the 1980s factors combined to determine the levels of re-
gional growth and investment.
Over the past quarter-century the distribution of
income among the developing countries has Domestic factors: the policy environment
changed markedly (Figure 1.1). The East Asian
countries account for the largest gain; their share The performance of countries in the 1980s varied
of developing country real incomes rose from 22 to according to their initial position and their ability
37 percent. All other regions had lower shares by to adjust to shocks over the decade. Many East
the 1980s, but Latin America and Sub-Saharan Af- Asian countries had a relatively healthy balance of
rica slipped furthest, by 6 and 5 percentage points, payments and strong trade performance at the be-
respectively. Population shares have also changed ginning of the 1980s, and their fiscal expenditures
over this twenty-five-year period, but much more were largely under control. When disturbances
gradually. Population growth has fallen below av- such as higher world interest rates came along,
erage in East Asia, leading to a 2 percent drop in these countries responded swiftly and succeeded
that region's share of the developing world's pop- in maintaining stability and restoring growth. In
ulation, whereas fertility rates in Sub-Saharan Af- contrast, many countries in Latin America and
rica are well above average, and the population Sub-Saharan Africa began the decade with greater
10
Table 1.2 Performance indicators, by developing region, selected periods
Growth of real per capita GDP (percent) Gross domestic investmentlGDP
Region 196 5-73 1973 -80 1980-89 1965-73 1973 -80 1980_89a
Sub-Saharan Africa 3.2 0.1 -2.2 16.2 20.8 16.1
East Asia 5.1 4.7 6.7 24.2 29.7 30.0
South Asia 1.2 1.7 3.2 17.1 19.9 22.3
Eastern Europeb 4.8 5.3 0.8 28.3 33.8 29.4
Middle East, North Africa
and other Europe 5.5 2.1 0.8 23.4 29.2 25.9
Latin America and
the Caribbean 3.7 2.6 -0.6 20.7 23.9 20.1
a. Data for 1989 are preliminary.
b. Estimates.
underlying imbalances, often hidden because bor- average, inflation in the severely indebted middle-
rowing had temporarily maintained growth. This income countries was more than 100 percent be-
group found it much harder to adjust to the shocks tween 1980 and 1987, compared with 8 percent in
of the 1980s. South Asia and 5 percent in East Asia.
Successful adjustment requires macroeconomic The recurrent bouts of high inflation that have
stability: a low and sustainable rate of inflation, a plagued Latin America are linked to the tax base,
realistic exchange rate, and a manageable level of the social structure, and domestic politics (Box
fiscal expenditures. It also requires a microeco- 1.2). Severe inflation cripples the economy and
nomic environment that is favorable to new invest- deepens economic crises in severa][ ways: it under-
ment. The countries that succeeded in both re- mines domestic confidence, reduces investment,
spects were able to sustain or improve their provokes capital flight, and often Leads to a misal-
growth performance during the 1980s. location of scarce foreign exchange. It further en-
courages dollarization (the use of foreign currency
RESTORING STABILITY. The shocks to interest rates as a medium of exchange), and it shrinks the tax
and terms of trade of the 1980s reduced real in- base by driving many economic activities into the
comes in most developing countries. Adjustment informal and illegal sectors. The result is an econ-
called for cuts in consumption and in government omy that does not respond to adjustment efforts.
spending. Countries that were particularly depen-
dent on primary exports or were heavily burdened RESTRUCTURING FOR GROWTH. For countries that
with debt needed to make even deeper cuts. By entered the 1980s with structural problems, raising
and large, the countries that weathered the storm long-run growth requires adjustment policies
were the ones that acted early. Indonesia, for ex- aimed at institutional reform and reallocation of
ample, saw its terms of trade decline by 25 percent resources. Unlike stabilization measures, which of-
in 1986; meanwhile, exchange rate movements ten hinge on quick and decisive adjustment, eco-
pushed its ratio of debt to gross national product nomic restructuring also requires long-term plan-
(GNP) to twice that of Brazil. The government ning. The trade regime, the financial sector, and
acted promptly by devaluing the rupiah and cut- the domestic regulatory framework are all crucial
ting public spending. The 1986 budget deficit was to this task.
held to 3.6 percent of GDP despite lost oil reve- The successful East Asian countries have acted
nues; the 1987 deficit fell to just 1 percent of GDP. swiftly to stabilize their economies while pursuing
In both years real GDP growth remained above 3 gradual reform programs and maintaining a com-
percent, and it has been even stronger since then. petitive exchange rate. The Republic of Korea, for
Government deficits, inflation, and unstable ex- example, pursued gradual but comprehensive
change rates are closely related in debtor coun- trade reform during the late 1970s and 1980s. Indo-
tries. Large government deficits in the late 1970s nesia supported its careful approach toward stable
and early 1980s were financed mainly by external exchange rate management with reforms of the
borrowing. With the halt in foreign loans in the trade regime, the domestic regulatory framework,
1980s many governments, particularly in Latin and the financial system. By contrast, most Latin
America, increasingly financed their deficits by American and Sub-Saharan African countries en-
borrowing at home and by printing money. On tered the decade with overvalued exchange rates
11
Box 1.2 Politics and economic performance
Several Latin American countries have gone through down, foreign exchange reserves are depleted, infla-
periods of intense political and economic change tion increases dramatically, and devaluations become
driven by a combination of social and redistributive inevitable. An erosion of external support and of access
goals, populist politics, and nationalism. Although the to foreign borrowing generally accompanies these de-
policies may come from the left or the right of the polit- velopments. In Brazil in the second year of the Cru-
ical spectrum, they and their consequences share re- zado Plan (named for the new currency introduced in
markable similarities. Examples include Argentina an effort to stabilize the economy) foreign exchange
(1946-49), Brazil (1985-88), Chile (1970-73), and Peru reserves were exhausted, inflation soared to more than
(1985-88). Ironically, the very people these programs 400 percent, and the exchange rate was sharply deval-
set out to help have often been harmed in the process. ued. In the final stages the program collapses with a
In each case the government looked for support to a surge of inflation, an outflow of capital, and a sharp
diversity of groups, especially the urban working class drop in real wages. In Chile real wages fell by 10 per-
and elements of the rural poor. The leaders promised to cent in 1972 and by 32 percent in 1973, to well below
accelerate and redistribute growth through state activ- their preprogram levels. In Peru real wages fell by 34
ism. Typically, they came to power after a period of percent in 1988, and in Brazil they fell by 29 percent
slow growth that was often the outcome of previous during 1987-88.
austerity programs. The programs have generally reduced investment,
The reformist agenda starts with expansionary mac- promoted capital flight, and left workers worse off.
roeconomic policies to promote employment and raise They have also had less visible effectsan erosion of
real wages. The results are encouraging. In Chile dur- investor confidence and loss of government credibility.
ing the first year of the Allende administration GDP Later governments are caught in a double bind. Social
increased by nearly 8 percent and real wages rose by 17 pressures to restore growth and stability are intensi-
percent. Labor's share of national income grew from 52 fied, and confidence in government is weakened. Un-
to 62 percent in that year. In Peru real wages grew by 27 der these circumstances it is difficult to halt inflation
percent in 1986 under the Garcia administration. through currency reform and a fiscal austerity program
In time the program starts to unravel. Stocks are run without a radical change in the policy environment.
that were sustained by high levels of protection Weak external demand, declining terms of trade, a
and overborrowing. Direct export taxes, nontariff diminishing supply of external finance, and a great
barriers, and quantitative controls on credit and increase in the volatility of interest rates combined
investment were also common. Adjustments were to produce an unusually adverse economic cli-
often delayed and hesitant. mate.
Toward the end of the decade many countries in
both regions did sharply devalue their currencies WORLD DEMAND AND TRADE. Growth in the out-
and begin substantial reforms of trade and domes- put and exports of developing countries is closely
tic policies. (Some, including Chile, Ghana, and correlated with demand in the industrial countries
Mexico, were already showing signs of restored (Figure 1.2). During the early 1980s growth in de-
growth.) But the response of exports and invest- veloping countries' exports fell as world growth
ment has often been slow. This may be because of slowed and industrial countries' imports stag-
continuing uncertainty about the policy regime. nated. Matters have improved somewhat since
Consistent actions are essential to convince the pri- 1983; the industrial countries have achieved an av-
vate sector that the policy stance will be main- erage GDP growth of 3.5 percent a year, and the
tained. In Sub-Saharan Africa the problems are volume of merchandise exports from developing
compounded by the lack of complementary infra- countries has expanded by 6.7 percent a year.
structure, heavy dependence on exports of pri- Imported primary commodities from developing
mary commodities, and weak entrepreneurial and countries are used mainly as intermediate inputs;
managerial capacity. Adjustment will take longer here the link to industrial country production is
under these circumstances. direct. Demand for developing countries' manu-
External factors: the global environment
factures is also related to industrial country pro-
duction and to changes in the pattern of final de-
Adverse developments in the world economy also mand. Between 1965 and 1988 the share of
had a part in the falling growth rates of the 1980s. manufactured goods in total developing country
12
EXTERNAL TERMS OF TRADE. For many of the poor-
Figure 1.2 Growth of OECD industrial production
est developing countries the purchasing power of
and developing country exports, 1970 to 1989 exports depends on the prices of a few primary
commoditiescocoa beans in Ghana, copper and
coffee in Papua New Guinea, and so onin rela-
Annual real growth (percent) tion to the prices of imports, which are mainly
12
manufactured goods. Prices for pnmary commodi-
ties, especially tropical products and food crops,
fluctuate sharply with global supply and demand.
During the 1980s prices for many primary com-
modities fell to their lowest levels since World War
II. Nonoil commodity prices declined for most of
the decade, although they recovered a little in
1988. By 1989 average commodity prices were still
33 percent lower than in 1980. Oil prices also fell
steadily between 1980 and 1985, but most develop-
-12 ing countries import oil, and so they benefited.
The decline in the terms of trade during the
-16 1980s has been most pronounced in Sub-Saharan
1970 1974 1978 1982 1986 1989 Africa and Latin America, althoi gh by the stan-
dards of the 1970s both regions started the decade
Export volumes, developing countries
Industrial production, OECD
in a favorable position (Figure 1.5). The fall in
prices during the 1980s cost Latin America and
Sub-Saharan Africa 13 and 15 percent, respec-
tively, of their exports' real import purchasing
power relative to the 1970s. In both regions
exports increased from 16 to 64 percent. The grow-
ing importance of manufacturesnotably, elec-
tronics, apparel, toys, and other consumer
goodshas strengthened the link between indus-
trial countries and East Asian exporters, in particu- Figure 1.3 Real export performance of developing
lar. The revival in world demand after 1985 had a countries, 1980 to 1989
smaller effect on the exports of other regions. Ex-
ports from South Asia have expanded rapidly Real exports (1980 = 100)
since 1985 but remain a low share of national out- 240
put. Latin America's exports have expanded
slowly and have only recently regained the level of 220
the early 1970s. Sub-Saharan Africa's exports fell 200
in the early part of the decade and stagnated
through 1988 (Figure 1.3). 180
Many highly indebted developing countries
have been running substantial trade surpluses,
largely because of their need to service debt. Be-
tween 1980 and 1981 the nineteen severely in-
debted middle-income countries had an average
trade deficit of $4 billion; during 1982-89 they
achieved an average annual surplus of $26 billion,
the equivalent of 3.3 percent of GDP (see Figure
1.4). This improvement has come largely through a
reduction in imports. It bears witness to the 1980 1983 1986 1989
squeeze on investment and consumption caused Sub-Saharan Africa East Asia
by the austerity programs undertaken in many of South Asia Latin America
these countries. Unfortunately, however, the de- Europe, Middle East, and the
and North Africa Caribbean
cline in imported intermediate and investment
goods has had adverse long-term effects.
13
ments. After 1984 this changed dramatically. Be-
tween 1984 and 1989 total net transfers were $153
Figure 1.4 Trade balance as share of GDP in
severely indebted middle-income countries, billion, bringing the average annual flow to $25
1970 to 1989 billion, or about 15 percent of the region's exports.
The halt in commercial lending coincided with,
Prcentage of GDP and was partly provoked by, falling terms of trade
6 and rising real interest ratesboth of which
pushed up the need for financing. Official lending
was also scaled back, compounding the difficul-
ties. Some countries in East Asia also found them-
selves deep in debt in the early 1980s, but they
have coped more easily with the problem. Strong
current account surpluses have reduced the re-
gion's borrowing needs, and the corresponding
capital outflow has in some cases taken the form of
accelerated amortization payments.
In Sub-Saharan Africa the story is very different.
Commercial borrowing has been a significant
source of funds only for a handful of middle-
income or resource-rich countries. Between 1984
and 1989 only 6 percent of net flows came from
private sources. As a result, more than 65 percent
of the stock of foreign debt in the region is official,
Figure 1.5 Changes in the terms of trade,
by developing region, 1980 to 1988
1970 1974 1978 1982 1986 1989
1970 - 80 average price = 100
130
roughly two-fifths of this loss was attributable to
the effect of lower petroleum prices on oil ex-
porters. Losses elsewhere were much less pro-
nounced.
EXTERNAL FINANCE. Debt can safely be used to
finance investment only if the investment gener-
ates the revenues that will be needed to repay the
loan. The debt accumulated by many developing
countries in the 1970s and early 1980s failed this
testalthough this was partly because of unfore-
seeable circumstances. The international financial
markets responded by halting most voluntary
lending to the principal debtors after 1983. The
threat of default kept banks from withholding fi- 80
nance altogether, but most of the loans went to- 1980 1982 1984 1986 1988
ward rolling over the debt and capitalizing interest
Sub-Saharan Africa East Asia
payments in arrears. South Asia Latin America
Until 1983 Latin America regularly experienced a Europe, Middle East, and the
positive net transfer of long-term debt (excluding and North Africa Caribbean
IMF credit): borrowing outpaced total debt repay-
14
Figure 1.6 Real interest rates, 1963 to 1989
Percent
10
Averages
1963-73 2.64%
1974-79 0.97%
1980-89 5.85%
3
-3
1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989
Note: The real interest rate is calculated as the London interbank offered rate (LIBOR) minus the U.S. GDP deflator.
and in 1988 more than half of this was conces- The additional burden of high real interest rates,
sional. Although debt remains a serious obstacle to in relation to their 1963-80 average, was roughly $8
growth, the region has continued to receive large billion a year for Latin America during the 1980s,
amounts of aid, and net transfers remained posi- or close to 1 percent of the region's GDP. Much of
tive through the 1980s. The structure of South the cost was concentrated between 1982 and 1985,
Asian debt resembles that of Sub-Saharan Africa, after the debt crisis first erupted. In 1984 alone this
but the region's high ratios of debt service to ex- interest rate shock is estimated to have cost 1.8
ports have been manageable thanks to relatively percent of Latin America's GDP. The cumulative
strong growth and prudent borrowing. shock to Sub-Saharan Africa and South Asia was
milderless than one-third of that experienced in
INTEREST RATE VARIABILITY. Floating rate debt be- Latin America.
came commonplace during the high-inflation years
of the mid-1970s. As long as real interest rates re- Prospects for the 1990s
main constant, floating rate debt should impose no
additional burden on debtors. It was the combina- The interaction of unfavorable external events and
tion of world recession, worsening terms of trade, inappropriate domestic policies has placed some
and a rise in real interest rates that brought on the countries on a persistent downward path. Several
debt crisis. of the severely indebted countries have become
Real interest rates were exceptionally high dur- progressively further removed from normal finan-
ing the 1980s. On average, they were more than cial relations. The prospects for these countries de-
twice as high as in the 1960s and nearly six times pend on credible changes in domestic policy and
higher than in 1974-79, when the developing on a response by the international community that
countries took on a large share of their debt (Figure will provide a breathing space for the adjustment
1.6). High rates were caused by a decline in indus- process.
trial country savings, by lack of progress in dealing If the patterns of regional income growth seen in
with global current account imbalances, and by the 1980s were to be repeated in the 1990s, the
large swings in the major currencies, perhaps ac- results would be disastrous for most of Sub-
companied by greater uncertainty regarding future Saharan Africa as well as for parts of Latin America
exchange rate movements. and South Asia. Sub-Saharan Africa, which today
15
Table 1.3 Prospects for the 1990s
Real GDP growth rates Real GDP per capita growth rates
Trend, Recent experience, Forecast, Trend, Recent experience, Forecast,
Group and region 1965-80 1980-89 1989-2000 1965-80 1980-89 1989-2000
Industrial countries 3.7 3.0 3.0 2.8 2.5 2.6
Developing countries 5.9 4.3 5.1 3.4 2.3 3.2
Sub-Saharan Africa 5.2 1.0 3.7 2.0 -2.2 0.5
East Asia 7.3 8.4 6.6 4.8 6.7 5.1
China 6.4 10.1 6.8 4.1 8.7 5.4
Other 8.1 6.4 6.3 5.5 4.2 4.6
South Asia 3.6 5.5 5.1 1.2 3.2 3.2
India 3.6 5.6 5.2 1.2 3.5 3.4
Other 3.9 5.0 4.8 1.2 2.2 2.4
Eastern Europe 5.3' 1.4' 1.9 4.5' 0.8' 1.5
Middle East,
North Africa, and
other Europe 6.3 2.9 4.3 3.9 0.8 2.1
Latin America and
the Caribbean 6.0 1.6 4.2 3.4 -0.6 2.3
a. Estimates.
has a population of about 450 million, would have prevent any disruption of global financial markets.
an additional 165 million people, and per capita Over the course of the decade lower U.S. govern-
incomes would be 20 percent lower than today's ment spending (especially on defense), a gradual
near-subsistence levels. The countries of Latin depreciation of the dollar, and buoyant growth in
America would have an additional 85 million peo- the other industrial countries should help to cor-
ple and average incomes 6 percent lower than to- rect the U.S. fiscal and current account deficits.
day's. International political developments should
The outlook, however, is for stronger perfor- make deficit correction in the United States easier
mance during the 1990s (Table 1.3). The develop- and help to spur more vigorous growth in Europe.
ing countries should grow, on average, by 5.1 per- The strategic arms negotiations between the War-
cent a year, compared with 4.3 percent in the saw Pact countries and the members of the North
1980s. This judgment reflects confidence that a Atlantic Treaty Organization, as well as unilateral
combination of improvements in domestic policy decisions by the U.S.S.R. and the United States to
and greater external assistance will gradually bring pare defense expenditures, will release resources
growth to closer to its long-run potential by the for other uses (Box 1.3). (The members of the War-
end of the decade. The disturbances of the 1980s saw Pact are Bulgaria, Czechoslovakia, the Ger-
are assumed not to recur, although the process of man Democratic Republic, Hungary, Poland, Ro-
recovery will be gradual. Real commodity prices mania, and the U.S.S.R.)
are expected to dip in the short run but then rise Despite this generally favorable outlook, the leg-
gradually and grow at an average 0.2 percent over acy of the 1980s remains evident. Although all re-
the decade. Real interest rates should ease to be- gions are expected to have positive per capita in-
tween 3 and 4 percent over the decade, as against come growth, Latin America and Sub-Saharan
an average of nearly 5.5 percent in the 1980s. Africa are unlikely to achieve their long-run poten-
The industrial countries should grow at roughly tial. With the reduction of debt under the Brady
3 percent a year-close to their long-run potential. initiative, per capita income growth in Latin Amer-
This outlook takes into account the high rates of ica might rise to 2.3 percent, as against an average
investment achieved since the mid-1980s and the of 3.4 percent a year between 1965 and 1980. If this
productivity growth evident in several countries. projection is correct, the number of poor people in
In the short run, growth in the United States is Latin America is unlikely to decrease during the
projected to slow to between 2.0 and 2.5 percent decade. In Sub-Saharan Africa per capita incomes
owing to a decrease in private and government are not likely to rise in the first half of the decade,
aggregate demand. The U.S. current account defi- although growth of about 1 percent a year is fore-
cit wifi remain below 2 percent of GDP in the early cast for 1995-2000. The combination of low income
1990s, but smooth financing of the deficit should growth and high fertility rates implies that the
16
Box 1.3 World military expenditures in the 1990s
In the 1980s world military expenditures grew to un- estimate that by the end of the decade defense outlays
precedented peacetime levels; at an estimated $1 tril- could be brought to half their current levels without
lion (Box figure 1.3), they accounted for roughly 5 per- jeopardizing the military balance between the super-
cent of total world income. (Owing to data powers.
imperfections and methodological differences, esti- The net impact of reduced defense costs on industrial
mates vary from source to source. The numbers shown and developing countries will depend on how the
here are given only as an example of the potential ben- "peace dividend" is used. In the United States a con-
efits from reductions in military spending.) But the de- servative estimate puts annual savings over the next
cade that began with accelerated spending has ended four years at about $45 billion, roughly three times the
with the promise of a sharp decrease. In 1989 defense total spent each year on foreign assistance. These re-
expenditures in the U.S.S.R. fell by an estimated $20 sources could be used to cut the fiscal deficit, boost
billion, or 6 percent of the defense budget. The United spending on domestic programs, assist industries af-
States is also scaling back defense spending, perhaps fected by the cutbacks, or expand the U.S. commit-
by as much as 10 percent over the next four years. ment to development assistance abroad.
Much deeper cuts may be feasible. Reining in defense spending in developing countries
Success hinges on the Strategic Arms Reduction should also be a priority over the coming decade. On
Talks and the Negotiations on Conventional Armed average, defense accounts for one-fifth of government
Forces in Europe. These talks seek to establish an ac- spending. Moreover, the high import content of de-
ceptable parity in strategic and conventional forces be- fense expenditures exacerbates the balance of pay-
tween the Warsaw Pact countries and the countries of ments difficulties and foreign exchange constraints
the North Atlantic Treaty Organization. Some analysts faced by many developing countries.
Box figure 1.3 Defense expenditures, 1987
Developing countries 0 Warsaw Pact NATO
$446.6
$364.5
$172.8
10.8
4.9
Total defense spending Defense spending as share of Defense spending as share
(billions of dollars) central government budget of GNP (percent)
(percent)
Note: NATO, North Atlantic Treaty Organization.
Source: U.S. Arms Control and Disarmament Agency 1988.
number of poor in the region is likely to swell Growth), the region will need to sustain and
rapidly. deepen reforms to rationalize th.e incentive sys-
Even this grim prospect assumes that Sub- tem, develop domestic infrastructure, diversify the
Saharari Africa perseveres with adjustment and productive base, and improve the efficiency of in-
continues to receive debt relief and financial aid. vestment. If this can be done, export volume could
As discussed in detail in a recent World Bank re- grow at about 3 percent a year over the decade,
port (Sub-Saharan Africa: From Crisis to Sustainable allowing the region to increase imports commen-
17
surately while raising the share of investment in pected to expand by more than 9 percent in real
national expenditure. The projection also assumes terms during the decade. In the high- and middle-
the reduction or rescheduling of all official bilateral income countries of the region exports will matter
debt falling due over the decade, in accordance less for growth as rising domestic demand as-
with the terms adopted by the Group of Seven sumes greater importance.
(G-7) countries in 1988. (The Toronto summit laid China's economy is projected to perform well,
out terms for rescheduling concessional bilateral although growth of per capita income is expected
claims and partially reducing nonconcessional bi- to fall from the 8.7 percent a year achieved in the
lateral claims on low-income African countries that 1980s to about 5.4 percent. Austerity measures de-
are following adjustment policies supported by the signed to restrain domestic inflation and rein in
World Bank or the IMF.) The gains from this assis- foreign and domestic borrowing will put a brake
tance are projected to amount to only 10 percent of on growth early in the decade. Continuing reforms
the countries' current long-term nonconcessional in pricing and labor markets and further decentral-
debt by the end of the decade. Clearly, the plan ization of investment and management should
will not relieve the region of its debt burden, but it help to bring about sustained advances in produc-
will ease the financial strain of debt service. tivity.
The projected per capita growth of 2.3 percent a Considerable uncertainty surrounds the outlook
year in Latin America will also depend on further for Eastern Europe. Needed economic reforms,
policy reform and a significantly lighter debt ser- combined with strong anti-inflationary measures,
vice burden. The projections assume lower interest will depress growth for a time despite generous
rates and buoyant exports (growth in volume of external assistance. Dismantling worker-managed
4.9 percent a year). They also assume a restoration firms and privatizing state enterprises will create
of creditworthiness and a return to precrisis invest- open unemployment on a large scale. In the me-
ment levels by the middle of the decade. The cur- dium run, however, the prospects for boosting
rent debt reduction strategies will need to be productivity and attracting new investment from
strengthened to ensure sufficient financing for Western Europe are good. By the end of the de-
new investment. cade growth should be robust, and over the de-
For other developing regions the outlook is more cade as a whole per capita income is expected to
favorable. Per capita incomes in South Asia appear grow by 1.5 percent a year.
set to continue growing at 3.2 percent a year Per capita GDP in other developing Europe, the
nearly three times as fast as between 1965 and Middle East, and North Africa is expected to in-
1980. India, in particular, is forecast to lead the crease at 2.1 percent a year. Thanks to steady
region with a 3.4 percent growth in per capita growth in world demand for oil and to an expected
GDPenough to allow substantial progress in re- decline in the share of output from countries that
ducing poverty. A critical assumption in the pro- are not members of the Organization of Petroleum
jection for India, however, is that the policies that Exporting Countries (OPEC), oil prices are forecast
allowed investment, productivity, and exports to to rise 3 percent a year in real terms. This will
expand rapidly in the 1980s will be maintained. strengthen the region's terms of trade. Debt ser-
With careful management of the real exchange vice, however, wifi continue to be a drain. Morocco
rate, export volume could grow at a robust 8 per- is the only North African country with commercial
cent a year. If measures are taken to restrain the debt big enough to merit consideration for relief
fiscal deficit, this rise should prevent debt service under the Brady initiative, but several other coun-
costs from undermining growth. tries have debt-to-exports ratios that put them
The countries of East Asia are expected to con- nearly on a par with many of the severely indebted
tinue the prudent and flexible macroeconomic poli- countries.
cies that have worked so well in the past. The re-
gion's real per capita income is projected to grow Risks in the outlook
at 5.1 percent a year. This would mean that aver-
age incomes would rise by a further 65 percent by These forecasts inevitably rest on assumptions that
2000 and that poverty would be nearly eliminated. may turn out to be wrong. By modifying some of
Strong global demand for the region's exports, the assumptions, it is possible to estimate ranges
particularly in Japan, will help to offset the effects for the forecasts. This exercise also confirms the
of a slackening U.S. domestic market. Manufac- important influence that fiscal and monetary poli-
tured exports from China and Indonesia are ex- cies in the industrial countries have on developing
18
countries. The fiscal policies of the United States slowdown there would mean that per capita in-
have particularly large international repercussions comes would fall over the decade. The outlook for
because of the country's size and the pivotal role of the incidence of poverty in Sub-Saharan Africa
the dollar in global financial markets. The persis- would worsen accordingly.
tence of current account imbalances among the Coordination of policies among the principal in-
United States, Japan, and Europe is largely the dustrial nations (notably, to stabilize exchange
result of the drop in saving in the United Statesa rates) became an important factor in international
decline for which both the government and house- economic relations during the 1980s. At their an-
holds are responsible. Overall savings in the nual summits, however, the G-7 countries have as
United States have fallen by about 3 percent of yet failed to give due consideration to the effects of
GNP since the early 1980s. If savings in the United their policies on the developing world. Finding so-
States and other industrial countries fail to rise as lutions to the problems faced by developing coun-
projectedperhaps because the United States fails tries, especially those that did not share in growth
to cut its federal deficitthen real interest rates will during the 1980s, will increasingly rest on coordi-
remain high in the 1990s. Under this scenario, and nated efforts that recognize these linkages. There
assuming that the industrial countries continue to are two tasks of immediate importance: to ease the
tighten their monetary policies, industrial country debt burden on developing countries and to lower
growth rates are likely to be about 0.5 percent barriers to world trade.
lower over the decade, and real interest rates will
probably hold steady at about the 1980s average of Dealing with debt in the 1990s
5.5 percent.
This situation would damage the growth pros- The goal of the highly indebted countries has not
pects for developing countries through four main changed: as in the 1980s it is to return to sustained
channels. First, slower industrial country growth growth and external creditworthiness. That these
would dampen import demand for developing countries have failed to restore growth and are
country exports. Second, higher real interest rates now more deeply indebted than at the outset is
would increase the debt service burden on coun- testimony to the difficulty of their task. Since the
tries with floating rate debt and those undertaking debt crisis began, many severely indebted coun-
new borrowing. Third, commodity prices would tries have restrained imports, raised exports, and
be likely to weaken, worsening the terms of trade thereby generated trade surpluses. These adjust-
for exporters of primary commodities. Finally, the ments, however, have been made at the cost of
slower-growing industrial countries would proba- compressed consumption and wages, lower in-
bly be less generous in their assistance to develop- vestment and output, and frequent recourse to in-
ing countries. flationary financing of government deficits. A
Developing countries that have heavy commer- strategy is needed to break this pattern in the
cial debts and depend on exports of primary com- 1990s.
modities would be the most vulnerable to this turn Among the severely indebted countries a distinc-
of events. Over the decade real income growth in tion must be made between low- and middle-
the developing countries as a group would be income countries because of their significant struc-
about 0.7 percent lower than the forecasts pre- tural differences (Table 1.4). The severely indebted
sented in Table 1.3. Asia would be least affected low-income countriestwenty-six in all, most of
since, compared with the group as a whole, the them in Sub-Saharan Africasuffer from deeply
region holds less commercial debt, has a greater rooted structural weaknesses. Most have weak fi-
share of manufactures in exports, and (except for nancial and infrastructural bases, depend on a nar-
the export-led countries) is less dependent on row range of primary commodities for exports,
trade. Real income growth in South and East Asia and are crippled by low nutritional and educa-
would be lower by about 0.6 percent. Latin Amer- tional standards. Rapid population growth exacer-
ica would be worst affected; its average real bates their difficulties. In contrast, the middle-
growth would be reduced by about 1.0 percent income debtors are well endowed with natural
over the 1990s. In Sub-Saharan Africa a low share resources and skilled labor and have well-
of floating rate debt offsets the region's high de- developed industrial bases.
pendence on commodity exports. As in Asia, in- For all the severely indebted countries the main
come growth wouki decline by 0.6 percent, but challenge is to design and implement credible pol-
because of the region's growing population, a icy reforms to foster growth. The direct benefits
19
Table 1.4 Comparative indicators for severely indebted low- and middle-income countries
(percent, unless otherwise specified)
Indicator Low income Middle income
Average population growth (1988) 3.1 2.0
GNP per capita (1988 dollars) 274.0 1,782.0
Gross domestic investment as a share of GDP
(current prices, 1987-88) 14.4 22.4
Exports as a share of GDP (1987-88) 19.4 16.3
Imports as a share of GDP (1987-88) 23.3 13.7
Share of manufacturing in exports (1987-88) 6.2 45.0
Share of nonfuel commodities in exports (1987-88) 52.5 39.3
Official development assistance as a share of GDP (1987) 8.2 0.5
Under 5 mortality rate (per thousand, 1985) 191 84
Primary net enrollment rate (1985) 50 89
Source: World Bank 1989f and World Bank data.
that can be won through negotiations cannot by restoring growth is greater. Experience with the
themselves lift the constraints imposed by chronic Brady initiative so far suggests that it is possible to
indebtedness; they must go hand in hand with the strengthen adjustment programs and mobilize pri-
indirect benefits of restored credibility, higher pri- vate investment through partial debt reduction
vate investment, and the repatriation of flight capi- (see Box 1.4). But there is room for improvement.
tal. New measures are needed to encourage invest- For a variety of reasons, foreign commercial credi-
ment, improve the allocation of resources, and tors remain reluctant to provide much new lend-
raise, in a less distortionary way, domestic reve- ing. Alternative profit opportunities and doubts
nues for financing government. The outlook for about the prospects for debt-distressed countries,
growth and renewed creditworthiness would then even with Brady programs in place, have acceler-
be bright. ated the exit of commercial banks. Changes in
In most cases official creditors account for more banking regulations in creditor countries could en-
than four-fifths of the total debt of the severely courage banks to take part in debt reduction pro-
indebted low-income countries. The principal grams and grant new loans. The official financial
creditor governments are trying to reduce the debt resources available under the Brady initiative
burden of this group through the mechanisms amount to $30 bfflion to $35 billion. This is a signifi-
agreed to at the 1988 Toronto summit. Sixteen Sub- cant amount of assistanceenough to reduce the
Saharan African countries have rescheduled under annual debt service of the severely indebted
the new protocol. The Special Program for Africa, middle-income countries by an estimated $6 billion
which provides concessional balance of payments a year between 1990 and 1993. But it is not enough
assistance to low-income countries that are under- to support programs for all nineteen Brady coun-
taking significant reforms, is expected to be ex- tries. Additional financing from bilateral official
tended beyond 1990, when the current program sources may become necessary.
ends. These, together with the IMF's Enhanced Insurance against shocks is also likely to be part
Structural Adjustment Facility, are the main of any successful strategy. A 10 percent deteriora-
sources of multilateral concessional assistance. tion in the terms of trade, or a 2 percent rise in
Even with favorable commodity prices and ex- world interest rates, could erase the gains from
port growth over the next decade, the severely in- debt reduction under the new approach. Protec-
debted low-income countries wifi need further as- tion against such contingencies is needed. One
sistance, including debt reduction, if they are to possibility is commodity-linked bonds, which
maintain per capita consumption and increase in- spread the risk between creditors and debtors in
vestment at the same time. They face structural the event of a severe drop in commodity prices.
impediments to growth that will take several years Another is to include provisions for severe output
to overcome. In the meantime, debt service, even price shocks. These provisions might mirror the
after the Toronto reschedulings, will continue to upside recapture clauses in the recent Mexican
cost an average of 5 percent of GDP during the agreement and link lower export prices to lower
1990s. debt repayment.
The Brady initiative is directed at the nineteen Policy reform to encourage investment and
middle-income countries with predominantly growth in export industries is an explicit part of the
commercial debt. In this group the potential for new strategy. Stronger exports would make it eas-
20
Box 1.4 Mexico's economic prospects after Brady
In January 1990 Mexico became the first country to mestic interest rates. A rise in the secondary market
complete negotiations for debt restructuring under the price for sovereign debt, as happened after the an-
Brady initiative. The agreement covers $49 billion of nouncement of the Brady plan for Mexico, may also
commercial loans and, according to some estimates, lower domestic interest rates and so promote invest-
provides $12 billion in debt relief. Net external ment. Second, lower inflation and greater confidence
transfersdebt service minus new loanswill, on av- will encourage the repatriation of flight capital. Be-
erage, be reduced by about $4 billion a year over the tween 1980 and 1988 an estimated $15 billion to $45
next six years. This is equivalent to nearly one-fifth of billion left the country through capital flight by private
merchandise exports in 1989. Mexican investors, and only a small part is estimated to
Success depends on the direct and indirect effects of have returned.
the agreement on domestic investment and growth. Without access to external funds, government invest-
The immediate direct impact is to reduce debt service ment would be severely curtailed, domestic interest
and thus free resources for other uses. But more impor- rates would have to rise much higher to prevent fur-
tant is the indirect effect on investment through the ther capital flight, and private investment would fail to
improvement in confidence and financial stability. recover. The Brady pact brings Mexico's goal of 5 per-
The indirect effects can come about in two ways. cent annual growth by 1996 within reach. Achieving it
First, lower debt service implies less dependence on will depend on stringent macroeconomic management,
printing money to cover the deficit and hence lower continued progress with the adjustment program, and
inflation. This should ease expectations of future ex- increasing private investment.
change rate depreciation and help to reduce real do-
ier to finance imports and service debt and would especially on more sophisticated manufactured
thereby help to restore creditworthiness. Clearly, goods, have proliferated. More than 120 such re-
however, such export-based growth would be hin- straints affected the exports of developing coun-
dered by greater protectionism in the industrial tries in 1988.
countries. Efforts to improve global trade relations The Uruguay Round offers a chance to create a
should be seen as an important part of the broader truly global trade regime under the GATT. The ex-
strategy for reducing debt and restoring growth in pansion of country and product coverage suggests
the middle-income debtor countries. that new ground may be broken. More countries
than ever before are active members of the GATT,
Changes in the world trade system
and several developing countries that had been
During the 1990s developing countries' export vol- nominal signatories are now full partners in the
umes are expected to grow at 6 percent a yearthe negotiations. The talks are covering virtually every
same as during the 1980s. Domestic policy is criti- kind of tradenot just conventional merchandise
cal for such growth, but just as essential will be the but also agriculture, services, irtvestment mea-
strengthening of the General Agreement on Tariffs sures affecting trade, and intellectual property
and Trade (GATT) through the Uruguay Round, rights.
which is scheduled for completion by the end of Open trade relations are ultimately in every-
1990. body's interest. Protection in the industrial coun-
Although successive GATT rounds have cut in- tries preserves only a small number of jobs, and at
dustrial country tariffs significantly, the use of great cost to consumers. In the United States, for
nontariff barriers has been on the rise lately. In the example, the cost of protecting each job in the tex-
United States nontariff barriers on steel, automo- tile industry is roughly four times 1:he annual wage
biles, and textiles are estimated to be equivalent to of the average textile employee. Protection in de-
an additional tariff of about 25 percent, raising pro- veloping countries burdens consumers and indus-
tection to the level of the early postwar years. In tries that need imported inputs, and it creates an
1989 the United States warned Brazil, India, and environment that rewards inefficiency. Competi-
Japan that it might take unilateral steps to protect tive industriesthe automobile industry in Korea,
trade by using its "Super 301" legislation. Trade for instance, and the production of commuter air-
frictions with the United States have been reduced planes in Brazilhave sometimes been built be-
through bilateral discussions, but the threat of uni- hind protectionist walls, but such success is rare,
lateral action remains. Voluntary export restraints, and failure is all too common.
21
Box 1.5 Going bananas in the European Community, 1992
World trade in bananas totals about $2 billion a year, ment will take, but it will have important effects on the
and 30 percent of this is sold in the European Commu- division of export earnings among banana producers.
nity. Although these imports are regulated by the EC, Box table 1.5 reports the simulated effects on the ba-
I
exceptions to existing rules protect the rights of some nana market of a move to free trade.
countries to grant preferential arrangements to tradi- Moving to free trade would lower the price that pro-
tional suppliers. For example, Italy imports bananas tected exporters receive by 49 percent, and they would
from Somalia and Britain from Jamaica and the Wind- suffer a loss of $209 million. Developing country ex-
ward Islands. Belgium, Denmark, and Luxembourg porters in the dollar banana countries would gain an
have set a flat 20 percent tariff on bananas from other estimated $61 million from the move to free trade. The
than African and Caribbean countries. Germany, the largest net gains ($386 million) would go to the Europe-
only country with essentially unrestricted trade, im- ans themselves, mainly as a result of the 24 percent
ports mainly from the efficient "dollar banana" coun- drop in import prices. Non-European importers (prin-
tries in Central and South America. cipally the United States) would lose an estimated $46
After 1992 a common agreement on banana imports million owing to the rise in world prices.
will be in place. It is still unclear what form this agree-
Box table 1.5 The impact on the banana market of moving to free trade in the European Community
Price effects Volume effects Revenue effects
Group (percent change)' (percent change) (millions of dollars)
Exporters
Protected1' 49 46 -209
Other suppliers 4 12 61
Importers
European protected markets' 24 15 394
European unprotected markets 1 1 8
Rest of world 2 1 -46
Note: Prices are based on 1987 data.
Average retail prices are used for importers and f.o.b. export prices for exporters.
Canary Islands, Guadaloupe, Jamaica, Madeira, Martinique, Somalia, the Windward Islands, and other African, Caribbean, and Pacific (ACP)
countries. (European imports of bananas from the Pacific states are, however, negligible.)
France, Greece, Italy, Portugal, Spain, and the United Kingdom. French prices are used to represent the market.
German prices represent the market.
Source: Borrell and Yang 1990.
ii
"Project 1992," the effort to create a single Euro- ten years, leading to an increase of about $4 billion
pean market, is likely to have a great effect on in imports of primary commodities from develop-
world trade. The European Community (EC) will ing countries. The EC Commission, however, has
become the largest market in the world, with a estimated that the removal of internal barriers may
population of 320 million and a GDP of about $6 reduce Europe's manufactured imports by 10 per-
trillion. It will account for about 30 percent of de- cent, and the developing countries would bear
veloping country exports. And this huge market part of that loss of trade. The possibility that trade
may well expand during the next decade to include diversion could outweigh trade creation is greater
several members of the European Free Trade Asso- for manufactures, especially if the gains in EC out-
ciation (EFTA) and some Eastern European coun- put come about through trade diversion rather
tries. (Indeed, the Gerthan Democratic Republic is than as a result of higher productivity. The net
already likely to be included.) impact on the main developing country exporters
The effect on the developing countries will de- to Europe might then be severe. This risk would
pend on whether the trade-creating effects of increase still further if the EC adopted special trade
greater efficiency and growth outweigh the trade- arrangements with Eastern Europe; other develop-
diverting effects of external barriers to entry. It has ing countries compete with Eastern Europe in sup-
been estimated that the 1992 program could boost plying manufactured goods to European markets.
European GDP by as much as 5 percent in five to Much depends on what form Europe's common
22
external trade barriers take. Will Europe remove its gional performances projected fo: the 1990s and
trade restrictions and reduce tariffsperhaps as asks how they will affect the outlook for reducing
part of the Uruguay Round (Box 1.5)? Or will it global poverty at the start of the next century.
replace trade restrictions with tariffs or adopt other
nontariff barriers, such as uniform quality stan-
dards and content requirements? All developing
country exportstropical and temperate agricul-
tural goods, manufactures, and serviceswill be Figure 1.7 Real per capita growth in developing
affected directly and indirectly by these decisions. countries in the 1980s and forecast values,
The greatest threat is that new trade barriers will 1990 to 2000
proliferate, encouraging retaliation elsewhere. (percent)
What does this mean for the poor?
7
The aggregate statistics examined in this chapter
6
are important for understanding events in the
world economy, but they do not show what is hap- 5
pening to people, especially the poor. The rest of
the Report addresses this question directly. How 4
the poor earn their living, the adequacy of their
3
health care, and their access to education and
other public services will be examined in detail. 2
But this chapter's findings concerning the differ-
ences in regional economic performance are imme- 1
diately relevant. Although growth in average per
capita incomes does not automatically improve the
well-being of the poor, it is a crucial factor. Figure
1.7 shows sharp regional differences in recent and
forecast income growth. What do these differences -2
mean for the poor?
-3
The next two chapters examine the current ex-
tent of poverty and the consequences for the poor o,
of recent economic performance. Has rapid per
co
capita growth in East and South Asia really
reached the poor? Who suffers most from falling 4?
incomes in Sub-Saharan Africa? Chapters 2 and 3
lay the foundation for the discussion of public poli- 1980.89 0 1990-2000
cies in Chapters 4 through 8. The final chapter re-
examines the implications of the divergent re-
23
What do we know
about the poor?
Reducing poverty is the fundamental objective of mud bricks, with earthen floors. They have little
economic development. It is estimated that in 1985 furniture and no toilet, electricity, or running wa-
more than one billion people in the developing ter. Water is obtained from a stream a fifteen-
world lived in absolute poverty. Clearly, economic minute walk away. The family has few posses-
development has a long way to go. Knowledge sions, apart from three acres of unirrigated land
about the poor is essential if governments are to and one cow, and virtually no savings.
adopt sound development strategies and more ef- The family raises sorghum, vegetables, and
fective policies for attacking poverty. How many groundnuts on its land. The work is seasonal and
poor are there? Where do they live? What are their physically demanding. At peak periods of tilling,
precise economic circumstances? Answering these sowing, and harvesting, all family members are
questions is the first step toward understanding involved, including the husband's parents, who
the impact of economic policies on the poor. This are sixty and seventy years old. The soil is very low
chapter draws on a number of detailed household in quality, but the family lacks access to fertilizer
surveys done over the past ten years or so, includ- and other modern inputs. Moreover, the region is
ing some conducted by the World Bank, to esti- susceptible to drought; the rains fail two years out
mate the number of poor people and to establish of every five. In addition to her farm work, the
what is known about them. wife has to fetch water, collect firewood, and feed
the family. The market town where the husband
Three poor families sells their meager cash crops and buys essentials is
five miles away and is reached by dirt tracks and
We begin by focusing on the people this Report is an unsealed road that is washed away every time
intended to helpby telling the stories of three the rains come.
poor families living in three different countries. None of the older family members ever attended
These families have much in common. For them, school, but the eight-year-old son is now in the
the difference between a tolerable quality of life first grade. The family hopes that he will be able to
and mere survival depends on their capacity to stay in schol, although there is pressure to keep
work and on their opportunities to work. Lack of him at home to help with the farm in the busy
education, landlessness, and acute vulnerability to periods. He and his two younger sisters have
illness and seasonal hard times affect all of them to never had any vaccinations and have never seen a
varying degrees. Problems such as these are at the doctor.
core of poverty.
A poor urban household in Peru
A poor subsistence farmer's household in Ghana
In a shantytown on the outskirts of Lima a shack
In Ghana's Savannah region a typical family of made of scraps of wood, iron, and cardboard
seven lives in three one-room huts made from houses a family of six. Inside there is a bed, a table,
24
a radio, and two benches. The kitchen consists of a A poor landless laborer's household in Bangladesh
small kerosene stove and some tins in one corner.
There is no toilet or electricity. The shantytown is In a rural community in a drought-prone region of
provided with some public services, but these tend Bangladesh a landless laborer and his family at-
to be intermittent. Garbage is collected twice a tempt to get through another lean season.
week. Water is delivered to those who have a ce- Their house consists of a packed mud floor and a
ment tank, but this family has been unable to save straw roof held up by bamboo poles from which
enough for the cement. In the meantime, the dry palm leaves are tied to serve as walls. Inside
mother and eldest daughter fill buckets at the pub- there is straw to sleep on and burlap bags for
lic standpipe 500 yards away. warmth. The laborer and his wife, three children,
Husband and wife are Indians from the same and niece do not own the land on which the shack
mountain village in the Sierra. Neither completed is built. They are lucky, however, to have a kindly
primary school. They came to Lima with two chil- neighbor who has indefinitely lent them the plot
dren almost four years ago, hoping to find work and a little extra on which they are able to grow
and schools. Although they have jobs, the eco- turmeric and ginger and have planted a jackfruit
nomic recession of the past few years has hit them tree.
hard. Better-off neighbors who arrived in Lima The father is an agricultural day laborer and
three to six years before they did say that it was tends to be underemployed most of the year. Dur-
easier to get ahead then. Still, husband and wife ing slow agricultural periods in the past he could
are hopeful that they will soon be able to rebuild sometimes find nonagricultural wage laborfor
their house with bricks and cement and, in time, example, in construction in a nearby townbut he
install electricity, running water, and a toilet like lost the strength to do much strerLuous work after
their neighbors. They now have four children, af- a bout of paratyphoid. He therefore engages in
ter losing one infant, and the two oldest attend the petty services around the village for very low pay.
local community school, recently built with funds The wife typically spends her dLay cooking, car-
and assistance from a nongovernmental organiza- ing for the children, husking rice, and fetching wa-
tion (NGO). All the children were given polio and ter from the well. She is helped in these tasks by
diphtheria-pertussis-tetanus (DPT) inoculations her thirteen-year-old niece, whose parents died in
when a mobile clinic came to the shantytown. a cholera epidemic some years ago. The woman
Community solidarity is strong, and a community and her niece are always on the lookout for ways
center is active in the shantytown. to earn a little extra. Such work as husking rice,
The father works in construction as a casual la- weeding fields, and chopping wood is sometimes
borer. The work is uncertain, and there are periods available from better-off neighbors. The nine-year-
when he must take any odd job he can find. When old son attends school a few mornings a week in a
he is hired on a construction site, however, it is town an hour's walk away. The rest of the day he
frequently for a month or so. His wife worries that and his seven-year-old sister gather fuel and edible
he will be injured on the job like some of his fellow roots and weeds. The sister also looks after the
workers, who can no longer work and yet receive baby when her mother or cousin cannot.
no compensation. She earns some income doing The household spends about 85 percent of its
laundry at a wealthy person's house twice a week. meager income on foodpredominantly rice. Fam-
To get there she must take a long bus ride, but the ily members are used to having only two meals a
job does enable her to look after her one- and day. They hope to struggle through to the rice har-
three-year-old children. She is also in charge of all vest without having to cut down and sell their
domestic chores at home. When she is away from jackfruit tree or the bamboo poles supporting their
the house for long periods, the two oldest children roof.
take morning and afternoon turns at school so as
not to leave the house unattended. There have Measuring poverty
been many burglaries in the neighborhood re-
cently, and although the family has few posses- These are the people behind the statistics. Lifting
sions, radios and kerosene stoves are much in de- them out of poverty will depend to a large extent
mand. The family lives on rice, bread, and on a better understanding of how many poor there
vegetable oil (all subsidized by the government), are, where they live, and, above all, why they are
supplemented with vegetables and, occasionally, poor. None of these questions turns out to be
some fish. straightforward. To begin with, it is necessary to
25
be more precise about what "poverty" really eracy, and access to public goods or common-
means. property resources. Being able to get clean
Poverty is not the same as inequality. The dis- drinking water, for example, matters to one's stan-
tinction needs to be stressed. Whereas poverty is dard of living, but it is not reflected in consump-
concerned with the absolute standard of living of a tion or income as usually measured. Households
part of societythe poorinequality refers to rela- with access to free public services are better off
tive living standards across the whole society. At than those without, even though their incomes
maximum inequality one person has everything and expenditures may be the same. Because of
and, clearly, poverty is high. But minimum in- these drawbacks, this Report supplements a
equality (where all are equal) is possible with zero consumption-based poverty measure with others,
poverty (where no one is poor) as well as with such as nutrition, life expectancy, under 5 mortal-
maximum poverty (where all are poor). ity, and school enrollment rates.
This Report defines poverty as the inability to
attain a minimal standard of living. To make this The poverty line
definition useful, three questions must be an-
swered. How do we measure the standard of liv- All the measures described above are judged in
ing? What do we mean by a minimal standard of relation to some norm. For example, we deem life
living? And, having thus identified the poor, how expectancies in some countries to be low in rela-
do we express the overall severity of poverty in a tion to those attained by other countries at a given
single measure or index? date. The choice of the norm is particularly impor-
tant in the case of the consumption-based mea-
Measuring the standard of living sures of poverty.
A consumption-based poverty line can be
Household incomes and expenditures per capita thought of as comprising two elements: the expen-
are adequate yardsticks for the standard of living diture necessary to buy a minimum standard of
as long as they include own production, which is nutrition and other basic necessities and a further
very important for most of the world's poor (Box amount that varies from country to country, re-
2.1). Neither measure, however, captures such di- flecting the cost of participating in the everyday
mensions of welfare as health, life expectancy, lit- life of society. The first part is relatively straightfor-
Box 2.1 How should we measure living standards?
Current consumption (including consumption from variability made little difference to estimates of pov-
own production) reflects households' ability to buffer erty. In bigger countries, however, rural-urban differ-
their standard of living through saving and borrowing, ences are a more significant problem. In Indonesia, for
despite income fluctuations. To that extent, consump- example, regional price differentials are large and affect
tion is a better measure of well-being than income. A regional comparisons of poverty.
study of Sri Lanka, however, found that both con- Household size and composition are also relevant.
sumption and income indicators identified the same Researchers estimate "equivalent adult scales" based
people, by and large, as poor. Furthermore, current on the consumption needs of individuals according to
consumption may not be a good measure of a house- age, sex, and activity. Allowing for household compo-
hold's typical standard of living (although it is proba- sition in the Sri Lanka estimates, however, yielded a
bly better than current income). In another study, poverty estimate similar to that based on the per capita
which looked at a set of Indian households over nine measures; the overlap in classifying people as poor was
years, 54 percent of households, on average, were 90 percent.
deemed poor on the basis of current-year consump- Such corrections typically assume that the distribu-
tion. Slightly fewer, 50 percent, were deemed poor on tion of consumption within the household is equal.
the basis of their nine-year mean consumption. What if this is not the case? A recent study of data for
Another problem is that different households may the Philippines concluded that allowing for inequality
face different prices. In developing countries prices of- in distribution might change the figures for the overall
ten vary between urban and rural areas. A study of Sri incidence of poverty but not the pattern across socio-
Lanka that allowed for this difference found that price economic groups.
26
Figure 2.1 Poverty and average standards of living, developed and developing countries
(1985 PPP dollars per capita a year)
Enlarged view
Poverty line 600
4,800
[
__- Upper poverty line ($370)
300
4,200
Lower poverty line ($275)
3,600
0
0 300 600 900 1,200 S
3,000 S S
2,400
.
1,800
S
1,200
S
S
600 S
.
0
0 1,200 2,400 3,600 4,800 6,000 7,200 8,400 9,600 10,800
Mean consumption
Note: PPP purchasing power parity.
ward. The cost of minimum adequate caloric in- chapter and in Chapter 9, however, a universal
takes and other necessities can be calculated by poverty line is needed to permit cross-country
looking at the prices of the foods that make up the comparison and aggregation. This global poverty
diets of the poor. The second part is far more sub- line is inevitably somewhat arbitrary. Rather than
jective; in some countries indoor plumbing is a lux- settle for a single number, this chapter employs
ury, but in others it is a "necessity." two: $275 and $370 per person a year. (The
The perception of poverty has evolved histori- amounts are in constant 1985 PPP prices.) This
cally and varies tremendously from culture to cul- range was chosen to span the poverty lines esti-
ture. Criteria for distinguishing poor from non- mated in recent studies for a number of countries
poor tend to reflect specific national priorities and with low average incomesBangiLadesh, the Arab
normative concepts of welfare and rights. In gen- Republic of Egypt, India, Indonesia, Kenya, Mo-
eral, as countries become wealthier, their percep- rocco, and Tanzania (see the inset in Figure 2.1).
tion of the acceptable minimum level of The lower limit of the range coincides with a pov-
consumptionthe poverty linechanges. Figure erty line commonly used for India.
2.1 plots country-specific poverty lines against per
capita consumption (both in 1985 purchasing How much poverty is there?
power parityPPPdollars) for thirty-four devel-
oping and industrial countries. The poverty Once the poor have been distinguished from the
threshold rises slowly at low levels of average con- nonpoor, the simplest way to measure poverty is
sumption but more sharply at higher levels. to express the number of poor as a proportion of
When discussing poverty within countries, this the population. This headcount index is a useful
Report uses country-specific poverty lines. In this measure, although it is often criticized because it
27
Box 2.2 Does how we measure poverty really matter?
The poverty gap and the headcount index are insensi- the price of rice goes upexactly the opposite of the
tive to the extent of inequality among the poor. If in- message conveyed by the headcount index.
come is transferred from a poor person to someone Consider next the design of a scheme for transferring
who is poorer, neither measure changes. Distribution- income to the poor. If success is judged solely by the
ally sensitive measures are designed to capture such headcount index, it is plain that the money should go
effects. first to the least poor because a given transfer will push
Suppose we want to know how an increase in the more of them over the poverty line. Small transfers to
price of basic food staples affects poverty. A recent the poorest of the poor will have absolutely no effect on
study examined this question for Java, Indonesia, us- the headcount measure of poverty. A similar issue
ing data for 1981. Households close to the poverty line arises in designing public employment schemes. For a
were found to be, on average, net producers of rice, the given budget, a scheme can aim for wide coverage at
main food staple. Thus, the headcount index of pov- low wages, leaving many participants still in poverty
erty will fall when the price of rice increasesassuming (albeit better off), or it can ration participation at a wage
that this price change is passed on to producers. But rate sufficient to allow more people to rise above the
the study found that the poorest of the poormany of threshold.
them landless agricultural laborers or farmers with lit- So, although simple poverty measures often give a
tie land but with some other source of incomeare net good indication of what has happened to poverty over
consumers of rice. They are worse off, at least in the time, for many other purposesincluding evaluations
short run, when the price of rice increases. The study of the effects of policy on povertyit is necessary to
showed that measures that take the severity of poverty look carefully at the distribution of income below the
into account tend to show an increase in poverty when poverty line.
ignores the extent to which the poor fall below the Although care has been taken to make the table
poverty line. The income shortfall, or poverty gup, as precise as possible, the margins of error are in-
avoids this drawback. It measures the transfer that evitably wide, and the figures, it must be stressed,
would bring the income of every poor person ex- are only estimates. The quality of the underlying
actly up to the poverty line, thereby eliminating data varies. Reputable household income and ex-
poverty. This Report relies on both the headcount penditure surveys have been used where avail-
index and the poverty gap. Box 2.2 discusses some able. These surveys encompass 2.5 billion people,
other measures. or almost 75 percent of the total. For other coun-
The use of the upper poverty line$370gives tries, including most of Sub-Saharan Africa, ex-
an estimate of 1,115 million people in the develop- trapolations have been made on the basis of indica-
ing countries in poverty in 1985. That is roughly tors that are strongly correlated with the measures
one-third of the total population of the developing of poverty derived from the household surveys.
world. Of these, 630 million-18 percent of the to- The notes to Table 2.1 give calculations of the po-
tal population of the developing worldwere ex- tential imprecision of the estimates as a result of
tremely poor: their annual consumption was less inadequacies in the data.
than $275, the lower poverty line. Despite these Nearly half of the developing world's poor, and
massive numbers, the aggregate poverty gapthe nearly half of those in extreme poverty, live in
transfer needed to lift everybody above the pov- South Asia. Sub-Saharan Africa has about one-
erty linewas only 3 percent of developing coun- third as many poor, although in relation to the re-
tries' total consumption. The transfer needed to lift gion's overall population, its poverty is roughly as
everybody out of extreme poverty was, of course, high. Table 2.1 also shows that both South Asia
even smallerjust 1 percent of developing coun- and Sub-Saharan Africa have low scores on several
tries' consumption. Mortality for children under 5 other social indicators; in Sub-Saharan Africa, in
averaged 121 per thousand for all developing particular, life expectancy and primary school en-
countries, aggregate life expectancy was 62 years, rollment rates are alarmingly low, and under 5
and the overall net primary school enrollment rate mortality rates are alarmingly high. The Middle
was 83 percent. These figures hide considerable Eastern and North African countries have the next
variation within and among countries. Table 2.1 highest poverty, according to all the indicators.
sets out a detailed regional breakdown of these They are followed by Latin America and the Carib-
estimates. bean and by East Asia. China's overall perfor-
28
mance is impressive, although the size of its popu- and the infant mortality rate (in 1:he lower panel)
lation means that a relatively low headcount index against average monthly consumption per capita
still translates into large numbers of poor. for urban and rural areas in India. At any given
level of consumption, the headcount index and
The chacacteristics of the poor (especially) the infant mortality rate can vary
widely. This underlines the need to look beyond
If governments are to reduce poverty or to judge average incomes to the distribution of income and
how their economic policies affect poverty, they the provision of social services.
need to know a lot about the poor. For example, Poverty as measured by low income tends to be
information on how the poor derive and spend at its worst in rural areas, even allowing for the
their incomes can help policymakers assess how often substantial differences in cost of living be-
changes in relative prices will affect real income. tween town and countryside. The problems of
Policies targeted directly to the poor can hardly malnutrition, lack of education, low life expect-
succeed unless governments know who the poor ancy, and substandard housing are also, as a rule,
are and how they respond to policies and to their more severe in rural areas. This is still true in Latin
environment. Unfortunately, gathering this sort of America, despite high urbanization rates. The im-
information is not always easy. The poor are heter- portance of rural poverty is no always under-
ogeneous, and data about their characteristics are stood, partly because the urban poor are more visi-
patchy. The following discussion looks at where ble and more vocal than their rural counterparts.
the poor live, the size and composition of their In 1980 El Salvador's infant mortality rate was 81
households, what they do for a living, what they per thousand live births in rural areas and 48 in the
own and purchase, what risks they face, and how towns; the incidence of malnutrition was five
they fit into the society around them. times higher in Peru's Sierra than in Lima. Table
2.2 confirms that in many countries rural poverty
Rural and urban poverty is a critical factor in the overall incidence and depth
of poverty.
In many countries poverty has a significant re- The extent of poverty can vary greatly among
gional dimension. In general, it is more common in rural areas within the same country. The acute
areas with low average incomes, but the link is deprivation in Brazil's Northeasi: Region, which
sometimes surprisingly weak. Figure 2.2 plots the has more than 50 percent of the country's poor but
headcount index of poverty (in the upper panel) only 27 percent of its total population, is well
Table 2.1 How much poverty is there in the developing countries? The situation in 1985
Social indicators
Extremely poor Poor (including extremely poor) Under 5 Net primary
Headcount Headcount mortality Life enrollment
Number index Poverty Number index Poverty (per expectancy rate
Region (millions) (percent) gap (millions) (percent) gap thousand) (years) (percent)
Sub-Saharan Africa 120 30 4 180 47 11 196 50 56
East Asia 120 9 0.4 280 20 1 96 67 96
China 80 8 1 210 20 3 58 69 93
South Asia 300 29 3 520 51 10 172 56 74
India 250 33 4 420 55 12 199 57 81
Eastern Europe 3 4 0.2 6 8 0.5 23 71 90
Middle East and North
Africa 40 21 1 60 31 2 148 61 75
Latin America and the
Caribbean 50 12 1 70 19 1 75 66 92
All developing
countries 633 18 1 1,116 33 3 121 62 83
Note: The poverty line in 1985 PPP dollars is $275 per capita a year for the extremely poor and $370 per capita a year for the poor.
The headcount index is defined as the percentage of the population below the poverty line. The 95 percent confidence intervals around the
point estimates for the headcount indices are Sub-Saharan Africa, 19, 76; East Asia, 21, 22; South Asia, 50, 53; Eastern Eutope, 7, 10; Middle East
and North Africa, 13, 51; Latin America and the Caribbean, 14, 30; and all developing countries, 28, 39.
The poverty gap is defined as the aggregate income shortfall of the poor as a percentage of aggregate consumption. Under 5 mortality rates are
for 1980-85, except for China and South Asia, where the period is 1975-80.
Source: Hill and Pebley 1988, Ravallion and others (background paper), and United Nations and World Bank data 1989.
29
known. Regional disparities are equally stark in Tenggara, the Andean highlands of Bolivia, Ecua-
many other countries. Thailand's Northeast Re- dor, Guatemala, and Peru, rural Gansu Province in
gion, Côte d'Ivoire's Savannah, Indonesia's Nusa China, parts of the central Asian region of the
U.S.S.R., and Appalachia in the United States are
all areas of concentrated poverty.
Many of the poor are located in regions where
arable land is scarce, agricultural productivity is
Figure 2.2 Poverty and infant mortality rates, low, and drought, floods, and environmental deg-
India, 1983, by urban and rural areas of states radation are common. In Latin America, for exam-
ple, the worst poverty occurs predominantly in
arid zones or in steep hill-slope areas that are eco-
Prcentage of population logically vulnerable. Such areas are often isolated
below poverty line
70
in every sense. Opportunities for nonfarm em-
ployment are few, and the demand for labor tends
60 to be highly seasonal. Others among the poor live
in regions that have a more promising endowment
of natural resources but lack access to social ser-
50
vices (education and health) and infrastructure (ir-
40
rigation, information and technical assistance,
transport, and market centers).
Although urban incomes are generally higher
30
and urban services and facilities more accessible,
poor town-dwellers may suffer more than rural
20
households from certain aspects of poverty. The
urban poor, typically housed in slums or squatter
10 settlements, often have to contend with appalling
overcrowding, bad sanitation, and contaminated
0
water. The sites are often illegal and dangerous.
80 100 120 140 160 180 200 Forcible eviction, floods and landslides, and chem-
Average consumption per capita ical pollution are constant threats. Some of these
(rupees a month)
people are migrants from the countryside who are
seeking better-paid work. For many, particularly in
Infant mortality rate Latin America, migration is permanent. For oth-
(deaths per thousand live births)
ers, as in East Africa and parts of Southeast Asia, it
200
may be temporary, reflecting (for example) season-
ality in agriculture. The effect that migration to the
towns has on poverty depends crucially on
150 whether urban employment opportunities are bet-
ter or worse than in rural areas. The evidence sug-
gests that urban areas do offer more opportunities
for higher-paid work, and this implies that, on bal-
100 ance, urbanization helps to reduce poverty.
What are the demographic characteristics of the poor?
50
Households with the lowest income per person
tend to be large, with many children or other eco-
nomically dependent members. In Pakistan in 1984
0 the poorest 10 percent of households had an aver-
80 100 120 140 160 180 200 age of 7.7 members, of whom 3.3 were children
Average consumption per capita under age 9. The corresponding national averages
(rupees a month)
were 6.1 and 2.0. Lack of a fit male adult can be
Source: Daft and Ravailion 1990 (upper panel); crucial, especially if women have small children to
India 1987 (lower panel).
care for or are culturally discouraged from taking
paid employment.
30
Table 2.2 Rural and urban poverty in the 1980s
Rural Rural Infant Access to
population poor mortality safe water
as as (per thousand (percentage of
percentage percentage live births) population)
Region and country of total of total Rural Urban Rural Urban
Sub-Saha ran Africa
Côte d'Ivoire 57 86 121 70 10 30
Ghana 65 80 87 67 39 93
Kenya 80 96 59 57 21 61
Asia
India 77 79 105 57 50 76
Indonesia 73 91 74 57 36 43
Malaysia 62 80 . . 76 96
Philippines 60 67 55 42 54 49
Thailand 70 80 43 28 66 56
Latin America
Guatemala 59 66 85 65 26 89
Mexico 31 37 79 29 51 79
Panama 50 59 28 22 63 100
Peru 44 52 101 54 17 73
Venezuela 15 20 .. 80 80
Does family size determine living standards, or lower in various parts of the world. Some of the
is it the other way around? The decision to have work that children do is highly exploitative; cases
many children can be a sensible response to pov- of debt bondage and of long hours worked in un-
erty. Mortality is high for children in destitute fam- healthy conditions for low wages are widely docu-
ilies, but it is essential to ensure that some children mented.
survive to support the household in the parents' Are women poorer than men? The data on in-
old age, if not sooner. Even before they can earn comes are too weak to give a clear answer. But the
income, children can free adults from various do- available figures on health, nutrition, education,
mestic tasks. Stifi, many poor parents report that and labor force participation show that women are
they want no more children and that their last- often severely disadvantaged. For example, data
born child had been unwanted. These couples of- for 1980 indicate that the literacy rate for women
ten lack access to modern family planning services. was only 61 percent of that for men in Africa; the
In the rural areas of many developing countries figures were 52 percent in South Asia, 57 percent
the aged often rely on the extended familya in the Middle East, 82 percent in Southeast Asia,
structure that tends to be stable over time. In ur- and 94 percent in Latin America. Women face all
ban areas multigenerational households are more manner of cultural, social, legal, and economic ob-
likely to break up, and the elderly are becoming stacles that meneven poor mendo not. They
more vulnerable. Moreover, traditions such as kin- typically work longer hours and, when they are
ship in Africa and the duty of sons in India and paid at all, for lower wages. A study in Nepal
Bangladesh to care for widowed mothers may be found that, on average, poor women worked
in decline. In India widows without an adult son eleven hours a day, men seven-and-a-half. In
are already a particularly underprivileged group. many developed countries the poorest include
Poverty and hunger among children is of partic- large numbers of single-mother households. Poor
ular concern. The very young are highly suscepti- female-headed households are also increasingly
ble to disease, and malnutrition and poverty- common in southern Africa and Latin America. In
related illnesses can cause permanent harm. Child Brazil female-headed households account for 10
poverty is strongly self-perpetuating. Child labor percent of all households but for 15 percent of the
is common; many households depend on it, and poor.
much of the work has the social purpose of engag-
ing the child in family activities. But work is often Assets
at the expense of schooling. For many poor people
the opportunity costs of sending children to school The poor usually lack assets as well as income. In
outweigh the future benefitsespecially for girls, local economies in which wealth and status come
whose economic value is often reckoned to be from the land, disadvantaged households are typi-
31
cally land poor or landless. Poverty is highly corre-
lated with landlessness in South Asia, southern
Africa, and much of Latin America. (See the data Figure 2.3 Poverty and illiteracy in India, 1981
in Table 2.3 for Bangladesh.) When the poor do
own land, it is often unproductive and frequently Percentage of illiterates age 15+
lies outside irrigated areas. The poor are usually 100
unable to improve such plots, since they lack in-
come and access to credit.
Many of the poor have access to land without
80
having ownership rights. Tenancy is common
although the poorest are often locked out of these 1-----II
arrangements because they lack the other re-
sources needed for farming. Tenancy does not pro- 60
vide collateral or a secure hedge against risk, and
access to the land from one year to the next is often
uncertain. 40
In other cases the poor have access to land that is
owned by the community or is common property.
Such arrangements are increasingly jeopardized
by population pressure, privatization, overex- 20
ploitation of resources, and deterioration of the en-
vironment. Studies of Rajasthani villages in India
over twenty years found that the income earned 0
I
from common-property resources has declined for
all these reasons. In the Sahel region of Africa 40-50 70-85 125-150 250-300
common ownership of the (generally unproduc- Household per capita expenditure
tive) land is considered a principal cause of over- (rupees a month)
grazing and deforestation. Environmental degra- o Female/urban Female/rural
dation of common-property resources can badly o Male/urban 0 Male/rural
hurt the assetless poor. Improving their access to
income-earning opportunities while protecting the Source: Sarvekshana 1988.
environment is an important policy issue.
The poor are also lacking in human capital.
Everywhere, they have a lower level of educational
achievement than the population at large. (Figure close together, to the detriment of their health. The
2.3 illustrates this with data from India.) Poor poor frequently suffer from hunger and malnutri-
women often have too many children, spaced too tion and from related illnesses. This undermines
their capacity for laboroften their main or only
asset.
The ownership of assets directly affects income
Table 2.3 Poverty and landholding opportunities (Box 2.3). Without assets such as
in Bangladesh, 1978-79
land, the poor must hire out their labor. Without
Headcount adequate human capital, they are limited to un-
Landholding Percentage Mean index
class of total income Mean land- (percentage skilled work. The elderly and the incapacitated
(acres of households (taka a holdings of may not even be able to offer their labor and may
land owned) in class month) (acres) population)
be forced to rely on charity. The importance of as-
Landless 7.1 508 0 93 sets, broadly defined, suggests that policies should
0-0.5 36.1 560 0.1 93
0.5-1.0 10.5 711 0.7 84
seek to increase the assets owned by the poor
1.0-1.5 8.9 783 1.2 78 especially skills, health, and other aspects of hu-
1.5-2.5 12.1 912 2.0 68 man capital and, in agricultural economies, land.
2.5-5.0 13.8 1,163 3.5 45
5.0-7.5 5.7 1,516 6.0 23
Sources of income
7.5+ 5.8 2,155 14.0 10
Besides having lower incomes and fewer assets
Total 100.0 865 2.1 70
than the nonpoor, the poor often have distinctive
Source: Ravallion 1989b. sources of livelihood. Most of the destitute mix
32
Box 2.3 Village-level perspectives on asset poverty
Much has been learned about asset poverty from vil- portant; it enables family members to bring in urban
lage studies. In Palanpur, a well-studied village in In- wages that can then be invested in farm innovation and
dia's Uttar Pradesh State, the most disadvantaged higher productivity.
groups were invariably landless casual laborers for A study of rural Tanzanian households in 1980 found
whom work was not available on a regular basis and that the poorest in the twenty sampled villages did not
households without an able-bodied male. In 1983-84 all possess significantly less land or labor resources than
households with both these characteristics were found others. Differences in living standards were largely at-
to be poor. They had very few opportunities for raising tributable to differences in human capital and in own-
their incomes. Self-employment was either restricted ership of nonlabor resources such as livestock. The
to certain castes or required skills and physical capital poorer households were less likely to participate in
that poor households lacked. As entry to regular wage market transactions than the rionpoor, since they
employment was also limited, poor men had no choice lacked the resources to grow cash crops and could not
but to take occasional agricultural work. Local tradition take the chance of a bad harvest that would leave them
denied even this option to most women. dependent on the market for their food needs. The
In Kenya, where population pressure on the land has poor also had much lower rates of return on work away
been rising, landholding was found to be an important from the family farm. Weak skills restricted them to
determinant of welfare. Another study, however, sug- marginal pursuits such as handicrafts, while the better-
gests that among smaliholders education is more im- off captured more lucrative wage employment.
many different earning activities. It is common for households were headed by self-employed cultiva-
the poor to work as cultivators, hunters and gath- tors. (This group also had the lowest literacy rate
erers, small artisans, petty traders, and wage la- and the largest poverty gap.) In these African
borers at various times of the year. The poor are countries agricultural wage earners are still rela-
rarely self-sufficient. They need cash to buy small tively unimportant among the poor. The situation
household items such as soap, clothes, salt, and is somewhat different in southern Africa; there
cooking oil, and they have to pay taxes and medi- many households hire out their labor, although not
cal and school costs. So they need to sell at least necessarily in farming.
some of their produce or obtain some paid work. Agricultural self-employment is also important
Agriculture is still the main source of income for in Southeast Asia. In Thailand a study of 1981 data
the world's poor. We have seen that the greatest found that 75 percent of poor rural households
numbers of the poor, including the very poorest, were self-employed in agriculture; for nonpoor ru-
are found overwhelmingly in rural areas. Their ral households the figure was 64 percent. Seven-
livelihoods are linked to farming, whether or not teen percent of all households, poor and nonpoor,
they earn their incomes directly from it. The de- depended primarily on agriculturaL wage labor. Re-
mand for nonfarm goods and services often de- cent data for Indonesia indicate that in 1987 house-
pends on the health of the farm economy. To help holds that were self-employed in farming (as tabu-
the rural poor, policies should strive to raise agri- lated by the principal activity of the head of
cultural productivitythrough investment in in- household) accounted for 58 percent of the poor
frastructure and through appropriate pricing, for and 41 percent of the total populaiion. Farm wage
example. laborers, in contrast, accounted for only 14 percent
Within agriculture, there are two groups of poor: of the poor and 9 percent of the population.
the self-employed and wage laborers. The distinc- The picture is more varied in South Asia. In In-
tion is often imprecise. Many casual farm laborers dia households self-employed in agriculture ac-
also own tiny plots of land. These plots are not counted for 35 percent of poor rural households in
sufficient to ensure family survival, but if they are 1977-78 and for 46 percent of all rural households.
productive, they can provide extra income and Households engaged in agricultural labor ac-
some collateral against risk. Most of the poor in counted for a further 44 percent of poor rural
Botswana, Côte d'Ivoire, Ghana, Kenya, Nigeria, households but for only 30 percent of all rural
and Tanzania are small-scale agriculturalists or households. Figures for Pakistan suggest that
pastoralists. In 1981-82 the poorest tenth of Ken- among the poor there are fewer farm laborers than
ya's population earned two-thirds of its income own-account cultivators; the opposite is true for
from farming. In Côte d'Ivoire in 1985 most poor Bangladesh.
33
In Latin America most of the poor are small-scale ers, prostitutes, or pickpockets. Incomes are low
farmers, but few derive adequate subsistence from and insecure. In Bombay poverty was found to be
their plots. In the plantation economies of Central more common among casual workers than among
America, especially in Costa Rica, wage labor in regular employees, and there was little mobility
agriculture is important. Peru's poor are mainly between the two forms of employment. About half
small farmers and herders. In a survey conducted the urban poor in Pakistan are self-employed,
in 1985-86, 78 percent of the heads of poor house- mostly in trade and manufacturing. They are gen-
holds said they were self-employed and 71 percent erally less skilled than people who work for wages.
said they worked in farming; the corresponding Transfers can be an important source of income
figures for all Peruvians were 60 and 40 percent. In for some of the poor. In most developing countries
the Sierra, Peru's poorest region, the numbers are transfers are made by relatives and friends or
even higher. In Mexico, too, poverty is concen- through vifiage support systems rather than by the
trated among cultivator families. government. Transfers accounted for 9 percent of
Rural nonfarm employment mainly consists of cot- the incomes of poor Kenyan smallholders in 1974-
tage industries, services, and commerce. In Asia, 75. In rural Java transfers are targeted toward the
Sub-Saharan Africa, and Latin America rural non- sick and elderly; in urban areas the unemployed
farm work tends to be highly seasonal or part- receive assistance. For the most disadvantaged
time. In Asia wages in these jobs are generally households, transfers can be crucial.
lower than wages in agriculture. The poor are con-
centrated in traditional industries with low skifi How does income variability affect the poor?
and capital requirements and very low labor pro-
ductivity. Their products are normally intended for Because incomes fluctuate, a static picture of pov-
home consumption or for the local market. De- erty can be deceptive. Evidence indicates that
mand is an important constraint on nonfarm eco- some people move in and out of poverty, whereas
nomic activities, which depend heavily on the pri- others never cross the poverty threshold (Box 2.4).
mary farm sector. In areas in which agricultural Households deliberately diversify their sources
incomes have grown, nonfarm employment has of income to reduce the risk that adverse circum-
flourished and wages have risen. Nonfarm em- stances will cause a sharp drop in income. But liv-
ployment is particularly important in providing ing standards still fluctuate considerablywith the
work in slack seasons for landless laborers and weather, with the time of year, because of the
women from poor households. death of the family breadwinner, and so on. If
In some of the poorest villages in Thailand, for households are unable to cushion a fall (through,
example, woven bamboo baskets and other forest for example, borrowing or insurance), an unfavor-
by-products provide extra income for rice farmers. able turn of events, especially an unexpected one,
Forest protection laws make this a risky and un- can be catastrophic. The poorest households are
profitable businessanother ifiustration of how the most vulnerable and the least able to protect
the loss of common-property resources harms the themselves from contingencies. One study tells
poor in the short term. Studies have found that how a young household in Guinea fell into destitu-
virtually all poor farmers in western Guatemala tion and eventually disintegrated after the father
and the northernmost region of the Peruvian Si- contracted river blindness. As he became progres-
erra supplement their farming income with arti- sively incapacitated, the mother's work load multi-
sanal production and petty trade. In the poorest plied and her health deteriorated, along with the
households much of this additional work is done children's. Soon the family was unable to feed it-
by women. self. In time, various members of the household
Informal sector jobs of one sort or another succumbed to illness or left the vifiage. After fif-
generally the jobs that pay leastare the main teen years what remained of the household was
source of livelihood for the urban poor. In Brazil in entirely dependent on village charity.
1985 an estimated 75 percent of heads of poor fami- Often, unanticipated events affect the poor not
lies worked in the informal sector, compared with merely in certain localities but nationwide. For in-
35 percent of the population as a whole. Disadvan- stance, a sudden decline in the terms of trade, fol-
taged urban groups are largely self-employed. lowed perhaps by policy changes intended to cope
They sell services and engage in trade or work on a with that decline, may change relative prices in
casual basis in construction, manufacturing, and ways that hurt the poor. Others may gain from the
transport. Some are full-time beggars, garbage sift- same events. Real currency devaluations, for ex-
34
Box 2.4 Moving in and out of poverty in rural India
How much of the poverty that we observe at any one
date is persistent and how much is transient, reflecting Box figure 2.4b Number of years in poverty sample
variability in individual incomes over time? Data col- villages, India
lected by the International Crops Research Institute for
the Semi-Arid Tropics (ICRISAT), based in Hyderabad,
India, track the incomes and consumption of 211 agri- Percentage of households
cultural households in central India between 1975 and 50
1983. Drought conditions are common in this region.
In Box figure 2.4a the percentage of households
40
deemed to be poor fluctuates over time around a gener-
I ally downward trend, from 64 percent at the beginning
of the period to a low of 41 percent in 1982; the average 30
proportion of poor was 50 percent. For each year the
poor are divided into those who were poor in the pre-
vious period and those who were not, and the same is 20
I done for the nonpoor. On average, 84 percent of the
poor at each date had been poor in the previous period, 10
and 16 percent were newcomers to poverty. Of the
I
0 2 3 4 5 hor
more
Box figure 2.4a Dynamics of rural poverty, Number of years in poverty
sample villages, India
100 0
nonpoor, 75 percent had been nonpoor in the previous
period, and 25 percent had moved out of poverty. The
80 20 income fluctuations are large: more than half the
households that moved into poverty did so with in-
comes of less than 80 percent of the poverty line, and
more than half of those who moved above the poverty
60 40
V line moved at least 20 percent above it.
Percentage Percentage Box figure 2.4b shows the number of years (not nec-
poor nonpoor essarily consecutive) that were spent in poverty. Only
A
40 60 about 12 percent of households were never poor dur-
ing the nine years. At the other extreme, 44 percent
were poor for six or more years, and 19 percent were
poor in every year. Thus, most households in these
20 80 villages do experience poverty at one time or another.
Whereas 50 percent are poor in a typical year, nearly 90
percent of the households are poor for at least one of
the nine years. The transient component is large. Yet it
0 100 is also clear that there is a substantial core of persistent,
1976 1977 1978 1979 1980 1981 1982 1983 chronic poverty in these villagespoverty experienced
by the same households year after year.
I
ample, usually benefit farmers who are net suppli- Iraq, Lebanon, southern Africa, and Sri Lanka
ers of internationally traded produce while hurting many of the poor are uprooted people fleeing from
the urban poor, who are typically net consumers of wars, persecution, famine, and natural disasters.
tradable goods. In some parts of the worldsuch Previously, they may not have been poor. In An-
as Afghanistan, Central America, Indochina, Iran, gola and Mozambique about 8.5 million people
35
ups and downs. Figure 2.4 shows consumption
and income over the course of a year for 178 poor
Figure 2.4 Seasonality in rural Bangladesh: poor
households in six villages, July 1984 to June 1985 households in Bangladesh. Consumption is clearly
smoother than income, which varies markedly
with the crop cycle. For these households, earn-
Taka a week ings from labor are less than consumption in most
265 periods, and the remaining sources of income
245
transfers, asset sales, and borrowingsmall as
they are, can be crucial to survival. Widespread
225 indebtedness to local moneylenders and shop-
Income keepers has been reported for various countries.
205
Friends and family are often a preferable source of
185 small amounts of credit, but they are likely to have
little to spare. (One study of the Philippines de-
165
Consumption
scribed extensive short-term borrowing and lend-
145 ing among poor families as a form of shared pov-
erty.) Borrowing may often be the only way to
125 maintain a minimum level of consumption.
July January June Credit is rarely available for accumulating pro-
Source: Hossain 1987.
ductive assets. Savings and borrowings often have
to be held in unproductive forms such as cash or
grain to hedge against future calamities. The poor
have few opportunities to obtain insurance, and
they are not able to use credit and savings to bol-
have become war refugees. Poverty in these coun- ster their capacity for bearing risk or to become
tries reflects a massive destruction of rural infra- entrepreneurs. Those on the edge of survival can
structure and productive assets. not afford to gamble.
Incomes in rural households vary substantially Another way of coping with risk is to diversify
according to the season. For example, wage work income sources. A study of rural Java found that
is readily available only at certain times of the crop the poorest landless households coordinate their
year, and it often depends on the weather. In many members' activities much more carefully than do
African countries the dry season puts an extra bur- asset-owning households. The main purpose of
den on women, who may have to walk miles to this coordination is to establish a steady stream of
find water. In some busy seasons heavy agricul- income. The household may adjust its overall sup-
tural work coincides with depleted food stocks and ply of labor, either through hours worked or
higher prices. Undernutrition and illnesses are through changes in labor force participation. Mem-
more common at certain times of the year. The bers of the household may migrate in search of
rains typically increase water contamination and work. Farmers are known to intercrop and to
the incidence of waterborne diseases. Acute choose crops that are quick to mature or are more
weight loss during the "hungry season" has been resilient, even though they may have lower yields
documented among farmers in The Gambia; adult and be less valuable.
weight fluctuated as much as 4.5 kilograms within
one year. In northeast Ghana losses of 6 percent of How do the poor spend their incomes?
body weight were recorded. Among women
farmers in Lesotho the figure was 7 percent, and Over a typical year the poor spend nearly all their
for pastoralists in Niger it was 5 percent. For vul- incomes on consumption of one sort or another,
nerable groups such as children, the aged, and and at least half of this consumption is likely to be
others whose biological defenses are already weak- in the form of food. Data for both Côte d'Ivoire
ened, seasonal weight change can be extremely and Peru in 1985 indicate that about 70 percent of
damaging. the expenditure of poor households goes for food;
Most of the rural poor, however, can cope rea- the corresponding figure for all households is
sonably well with normal seasonality. Saving and about 50 percent. A high proportion of the food
dissaving help to smooth consumption over the budget-60 percent in the case of the Indonesian
36
poor, for instanceis devoted to the local food mates, it is only slightly more equal than the distri-
staple. bution of income, which is known to be among the
The relative prices of food staples can be crucial most unequal in the world. The primary school
to the welfare of the poor. The number of rural enrollment rate for the wealthiest 20 percent in
poor who rely on the market for their consumption Côte d'Ivoire is twice that of the poorest 20 per-
depends on the distribution of access to productive cent. In both India and Nepal enrollment rates for
land. It is typically the "not so poor" who are net the top 10 percent of families are 50 to 100 percent
suppliers of farm produce and so benefit from higher than for the poor. Morbidity and mortality
higher prices. The poorest of the poor consume figures tell a similar story. Access is not the only
more of the food staple than they produce, and issue; use of services by the poor can also be low.
they rely on agricultural wages, which may be But the fact remains that fewer social services are
slow to respond to changing prices. available to the poor.
Poverty is often the fundamental cause of mal- There are exceptions. In Eastern Europe and in a
nutrition. Yet nutrition need not be very respon- few low-income and lower-middle-income
sive to changes in income. The poor may spend a countriesincluding Chile, China, Costa Rica,
rise in income on "better" food (rather than on Cuba, Mauritius, and Sri Lanka as well as India's
more nutritious food) or on other things alto- Kerala Stategovernments have tried hard to pro-
gether. In Indonesia one study found that when vide basic services widely and have largely suc-
income rose by 10 percent, calorie intake for the ceeded. For example, in Sri Lanka 93 percent of the
average household increased only 1.5 percent. For population has access to health services. Chile's
the poorest tenth of households the increase was 4 social services are carefully targeted toward the
percent. neediest. Through its commune organization
Within the household the distribution of con- China was able to meet certain basic needs such as
sumption often favors males and income-earning health care, education, and family planning ser-
adults. This finding has been documented in vari- vices for most of its people.
ous countries, including Bangladesh, Ghana, Gua- The poor are often set apart by cultural and edu-
temala, India, and Papua New Guinea. Poor cational barriers. Illiterate people may be intimi-
households are more likely to invest in education dated by officials or may simply lack information
for boys than for girls. A series of studies in Brazil, about programs. Sometimes the design of the ser-
India, Kenya, Malawi, and other countries indi- vices unintentionally adds to the problem. Agricul-
cates that a larger proportion of women's than of tural extension programs, for example, are usually
men's income tends to be spent on household nu- geared toward men even where--as is often the
trition and basic welfare. Of the studies conducted casemany, if not most, cultivators are women.
in India, one covering twenty villages in the south The requirement that birth certificates be produced
found that 80 to 100 percent of women's wage in- for admission to school in urban aras of India pre-
come was devoted to family maintenance, whereas vents poor migrant women from enrolling children
men committed between 40 and 90 percent of their and taking needed work. The poor play little part
earnings. This suggests that raising women's in- in politics and are often, in effect, cLisenfranchised.
comes directly is a good way to reach children as In many countries poverty is correlated with race
well as to strengthen women's status and bargain- and ethnic background. Indigenous peoples in Bo-
ing power within the household. livia, Ecuador, Guatemala, Mexico, and Peru are
disproportionately represented among the poor.
What is the position of the poor in the society Scheduled castes and tribal peoples are among
around them? those most at risk of poverty in India and Bangla-
desh. In Australia the aborigines are prominent
In general, the poor have less access to publicly among the poor. In the United Stal:es 45 percent of
provided goods and infrastructure than do other all black children were poor in 1984, compared
groups. On the whole, governments fail to reach with 17 percent of white children. In South Africa
the rural poor. Even in urban areas poor neighbor- the mortality rate for white infants averaged 12 per
hoods are less well supplied with services than thousand live births between 1981 and 1985; for
well-to-do ones. Data for Latin America show that black Africans it was estimated to fall between 94
the pattern of social expenditures is regressive in and 124. The risk of contracting tuberculosisa
most countries. In Brazil, according to some esti- disease closely associated with povertywas
37
twenty-two times greater for blacks than for whites skilled, educated, and healthy, minimal standards
in South Africa, excluding the homelands, and of living are ensured and poverty is eliminated.
fifty-five times greater in Transkei. When such opportunities are lacking and access to
social services is limited, living standards are unac-
From diagnosis to treatment ceptably low. The living standards of many in the
developing world are also highly vulnerable to a
This chapter's survey of what we know about the variety of misfortunes, ranging from illness to
poor points to two overwhelmingly important de- drought. Inability to cope with shocks can render
terminants of poverty: access to income-earning relatively well-off households poor and lead to
opportunities and the capacity to respond. When starvation and death for those already impover-
households have secure opportunities to use their ished. These facts set the agenda for the rest of this
labor to good purpose and household members are Report.
38
Progress on poverty: lessons
for the future
Over the past twenty-five years the developing nesses in the data make it impossible to be precise.
countries have made tremendous progress. Con- The evidence points to considerable progress in
sumption per capita has increased by nearly 70 reducing poverty, especially in the 1960s and
percent in real terms, average life expectancy has 1970s; the picture for the 1980s is mixed. In some
risen from 51 to 62 years, and primary enrollment regions the poor have suffered serious setbacks,
rates have reached 84 percent (Table 3.1). These whereas in others the progress of previous de-
remarkable gains, if evenly spread, would have cades has continued and has even accelerated.
gone a long way toward eliminating povertybut,
of course, they have not been evenly spread. Some General indicators
countries have done much better than others. An
examination of why this is soof the reasons for Since there is no universally accepted indicator of
successes and failurescan suggest practical solu- welfare, it is necessary to look at several different
tions to the problem of poverty. measures. We begin with a review of changes in
This chapter draws on past trends to identify the consumption per capita, life expectancy, and edu-
key factors behind rapid and sustained improve- cational attainment. Figure 3.1 reveals that al-
ment in the living standards of the poor. The ele- though all regions have made good progress in life
ments of the broad strategy that emerge from this expectancy and primary enrollment, the gains in
analysis are then discussed in more detail in Chap- consumption per capita have differed consider-
ters 4 through 7. ably. South Asia, for example, recorded an im-
provement of only 30 percent from a very low
Changes in poverty base, and consumption per capita in Sub-Saharan
Africa stagnated. There are two implications. First,
If history is to guide future policy toward poverty, improvements in social indicators such as life ex-
it is important to be as accurate as possible about pectancy and primary enrollment can occur inde-
what has actually happened. Unfortunately, weak- pendent of increases in consumption. And second,
since many of the world's poor live in South Asia
and Sub-Saharan Africa, the substantial increase
Table 3.1 Twenty years of progress in consumption per capita in the developing world
in developing countries
as a whole has not led to an equally impressive
Indicator 1965 1975 1985 reduction in poverty.
Consumption per capita Focusing on regional averages, however, pro-
(1985 PPP dollars) 590 780 985 vides only indirect evidence about the poor. How
Life expectancy much, if at all, have the poor participated in the
(years) 51 57 62 general improvement shown in Figure 3.1? To an-
Primary net enrollment rate swer this question it is necessary to move from
(percent) 73 74 84 aggregate data to survey results.
39
Figure 3.1 Progress and stagnation in the developing world
Net primary enrollment rates Life expectancy at birth Consumption per capita
Percentage of age group Years 1985 PPP dollars
100 75 3,000
ni
75 65 2,250
50 55 1,500
25 45 750
I) 35 ()
.
C,
L2 1965 0 1975 0 1985
What has happened to the incomes of the poor? lowed inegalitarian paths of development, such as
Brazil and Pakistan, have succeeded in reducing
Surveys that cover periods of at least ten years are the headcount index. More important, India, the
available for eleven countries, which together ac- country with the world's greatest concentration of
count for 40 percent of the total population of the poor, cut the incidence of poverty by 11 percentage
developing world and for 50 percent of the poor points in just eleven years.
(Table 3.2). The surveys for earlier years are gener- In some countries rapid population growth has
ally less reliable than more recent ones, and the increased the absolute number of poor. Neverthe-
results should therefore be interpreted as indica- less, as the changes in the average income shortfall
ting trends rather than precise magnitudes. Note show, the poor are better off even though they
also that, in contrast to the exercise in Chapter 2, may have remained poor. In Morocco, for exam-
this chapter uses country-specific poverty lines. ple, the number of people below the poverty line
Comparisons between countries should be made increased by almost a million, but their average
with this difference in mind. expenditure rose from 54 percent of the poverty
Table 3.2 reveals considerable progress in reduc- line to 64 percent. In sum, therefore, the evidence
ing the incidence of poverty. Indonesia, for exam- suggests that there has been considerable progress
ple, cut its headcount index by 41 percentage in reducing the incidence of poverty, a more mod-
points in just seventeen years. (Box 3.1 describes est reduction in the number of poor, and achieve-
this change from the perspective of one village.) ment of somewhat better living standards for those
Even countries that are often thought to have fol- who have remained in poverty.
40
A weakness in this assessment is the lack of reli- suggest that progress in reducing poverty has
able data for Sub-Saharan Africa. The discussion probably been slowest in that region. Even assum-
of diverging trends in Chapter 1 and the evidence ing that the distribution of income did not worsen
of stagnant consumption per capita in Figure 3.1 between 1965 and 1985, the number of Africans in
Table 3.2 Changes in selected indicators of poverty
Average
Number income
Headcount of poor shortfall
Length index (millions) (percent)
of period First Last First Last First Last
Country and period (years) year year year year year year
Brazil (1960_80)a.b 20 50 21 36.1 25.4 46 41
Colombia (1971_88)' 17 41 25 8.9 7.5 41 38
Costa Rica (1971_86)a 15 45 24 0.8 0.6 40 44
India (1972-83) 11 54 43 311.4 315.0 31 28
Indonesia (1970-87) 17 58 17 67.9 30.0 37 17
Malaysia (1973-87)' 14 37 15 4.1 2.2 40 24
Morocco (1970-84) 14 43 34 6.6 7.4 46 36
Pakistan (1962-84)'." 22 54 23 26.5 21.3 39 26
Singapore (1972-82) 10 31 10 0.7 0.2 37 33
Sri Lanka (1963-82)' 19 37 27 3.9 4.1 35 29
Thailand (196286)a.b 24 59 26 16.7 13.6 35
Note: This table uses country-specific poverty lines. Official or commonly used poverty lines have been used when available. In other cases the
poverty line has been set at 30 percent of mean income or expenditure. The range of poverty lines, expressed in terms of expenditure per
household member and in PPP dollars, is approximately $300-$700 a year in 1985 except for Costa Rica ($960), Malaysia ($1,420), and Singapore
($860). Unless otherwise indicated, the table is based on expenditure per household member. The headcount index is the percentage of the
population below the poverty line. The average income shortfall is the mean distance of consumption or income of the poor below the poverty
line, as a proportion of the poverty line.
Measures for this entry use income rather than expenditure.
Measures for this entry are by household rather than by household member.
Box 3.1 Development in a Javanese village
The story of Balearjo, an East Javanese village of almost shoes were commonplace. Most villagers had radios,
4,000 people, shows what declining poverty means for and some even had television sets. More than 90 per-
individuals. The village is about eight kilometers from cent of the houses were made of colorfully painted
the town of Gondanglegi and is connected to the out- brick and stucco, with partial cement floors. Furnish-
side world by bumpy but passable dirt roads. Although ings were more extensive and included chairs and ta-
Balearjo is still somewhat poorer than its neighbors, bles bought from stores. Literacy had improved dra-
research conducted in 1953 and 1985 shows that the matically thanks to two primary schools, one financed
lives of its inhabitants improved greatly in the inter- by the village and the other by the central government.
vening years. Rice yields increased dramatically, from 2 Travel outside the village was common, and knowl-
tons to 6 tons of paddy per hectare for the wet season edge of national events, provided through hourly radio
crop, and the wage for a day's work increased from 2 broadcasts, was widespread. In 1953 villagers relied on
kilograms of rice in 1953 to nearly 4 kilograms in 1985. homemade kerosene lamps that provided little illumi-
In 1953 the village would have been considered poor nation, but by 1982 electric power lines had reached
by most definitions. Rice was available for only four Balearjo, and by 1985 many households had electric
months; the diet for the rest of the year consisted of light.
corn and, when that ran out, cassava. Clothes were Such burdensome activities as rice pounding and
worn until they were in tatters, and few people had shoulder transport had disappeared, relieving women
shoes. A typical house was made of thatch and bam- of some of their most exhausting tasks. Higher incomes
boo, with an earthen floor. Furnishings were sparse had led to demands for new products and services and
and uncomfortable. Few villagers could read, and few hence to more productive work, such as construction,
had traveled any distance from the village. A daily pa- trade, and small manufacturing. Growing specializa-
per brought from a nearby town supplied the only out- tion was also evident: houses in 1953 were constructed
side news. by the owners with the help of neighbors, but by 1985
By 1985 things had changed. Rice was available most of the work was done (and done better) by full-
throughout the year. Clothing was much better, and time carpenters.
41
Box 3.2 Poverty in Africa
The absence of reliable intertemporal statistics on in- In Nigeria trends in poverty have followed a some-
come distribution in most Sub-Saharan African coun- what different pattern. The rise in world oil prices and
tries makes any comprehensive account of trends in in Nigerian oil production increased per capita con-
poverty there impossible. The Social Dimensions of sumption and income throughout most of the 1970s,
Adjustment project is beginning to address this prob- but the economic reversal of the early 1980s has had a
lem, and surveys have been conducted in three coun- severe effect on the country's poor. Consumption has
tries. For now, however, an assessment of past trends plummeted by 7 percent a year, and standards of living
in poverty has to rely on a few household surveys, were lower in the mid-1980s than in the 1950s. Analy-
supplemented by village studies, and on aggregate sta- ses of caloric intake show no improvement between
tistics for income, consumption, and population. 1952 and 1985. The economic crisis of the 1980s was so
Tanzania is one of the few Sub-Saharan African severe that it more than canceled out the progress of
countries for which large-scale urban and rural house- the previous twenty years.
hold surveys have been conducted over a period of Ghana has only recently begun to see rising living
several years. The surveys show that real rural living standards after two decades of negative growth. In
standards declined at an average annual rate of 2.5 1985 nearly 60 percent of the population lived on less
percent between 1969 and 1983. The decline in urban than $370 a year. Botswana has been one of the few
areas was even more dramatic; real wages fell by 65 African countries to achieve rapid growth (nearly 9 per-
percent over the period. Real private consumption per cent a year since 1965), but even there almost 50 per-
capita has fallen by 43 percent since 1973, and food cent of the population had incomes of less than $370 a
purchases have moved away from meat, dairy prod- year in 1985-86. With few exceptions, the evidence
ucts, and vegetables toward cheap starches and beans. supports the conclusion that poverty in Sub-Saharan
Small-scale village surveys have revealed worsening Africa is severe and has been getting worse.
social indicators despite government efforts to provide
services.
poverty would have increased by 55 million. Data Social progress and the poor
from small local surveys and other evidence sup-
port this general conclusion (Box 3.2). In principle, a similar analysis for the social indica-
Table 3.2 deals with periods of at least ten years tors shown in Figure 3.1 would be desirable, but
and may therefore mask the effect of short-run re- distributions of social indicators are not usually
cessions. Many observers have argued that the re- available. We do know, however, that the nonpoor
cession and adjustment of the 1980s were particu- usually have access to social services before the
larly harmful to the poor. Table 3.3 presents poor do. Thus, only if the percentage of the popu-
evidence on changes in poverty in the 1980s. The lation with access to social services exceeds the
data reveal continued progress in several East percentage of nonpoor can we conclude that some
Asian and South Asian countries. India, Indone- of the poor are being reached.
sia, Malaysia, and Pakistanwhich accounted for Table 3.4 compares the percentage of nonpoor in
more than 40 percent of the world's poor in 1985 the population with primary enrollment rates and
managed to reduce the incidence of poverty in the the percentage of children immunized. In East
1980s. China, although it suffered a reversal after Asia universal primary enrollment had almost
1985, also saw a decline in poverty during the early been reached by 1985, implying that most of the
1980s. In the regions most severely affected by the poor had access to primary education. In Sub-
recession, however, poverty has increased. In all Saharan Africa the enrollment rate is only slightly
the Latin American countries in the table, includ- higher than the percentage of nonpoor, and it is
ing Brazil, the incidence of poverty increased for at therefore probable that few of the poor are being
least some part of the 1980s. In Sub-Saharan Africa reached. The other regions occupy an intermediate
the only data available, those for Côte d'Ivoire, position. Health coverage, as measured by the per-
display a slight increase in the mid-1980s. Finally, centage of children immunized, follows a similar
the problems of Eastern Europe are clear: Poland regional pattern, but at lower levels. The evidence
and Yugoslavia experienced a sharp rise in suggests that many poor people have benefited
poverty. from the expansion in education (except in Sub-
42
Saharan Africa) but that less progress had been countries that have done much worse than re-
made up to 1985 in extending health care to the gional averages indicate. In Pakistan the net enroll-
poor. Further expansion in coverage will mainly ment ratio has hardly improved in the past twenty
benefit the poor. years-it was only 43 percent in 1985-and an esti-
Regional averages mask the tremendous mated 36 percent of the population lacks access to
achievements that some countries have made in health care.
providing social services to their populations. Co- Recently, concern has centered on the effect of
lombia, where mortality for children under 5 fell the recession of the early 1980s on the provision of
from 135 per thousand in 1965 to 42 per thousand social services to the poor. In Sub-Saharan Africa
by 1985, and Costa Rica, where 95 percent of the and Latin America, the two regions worst affected
population has access to primary health care, show by recession, roughly half the countries for which
what can be done. Even in regions with poor over- information is available experienced substantial
all performance, some countries have managed to declines in real per capita spending on education
make great strides. Botswana, for example, has and health. The social indicators for the early
achieved universal primary enrollment, and its un- 1980s, however, tell a somewhat less gloomy story,
der 5 mortality rate fell from 165 to 70 per thousand at least in Latin America.
during the past two decades. The sheer scale of the Figure 3.2 shows that progress in under 5 mor-
improvements in these countries suggests that the tality and primary school enrollment rates contin-
poor must have participated in the overall ued into the 1980s in most of the developing
progress. At the other end of the spectrum are world. Progress was least in the region with the
Table 3.3 Changes in poverty in the 1980s
Average
Number income
Headcount of poor shortfall
index (millions) (percent)
Length
of period First Last First Last First Last
Count nj and period (years) year year year year year year
Brazil (1981-87)' 6 19 24 23.1 33.2
China (l985_88)ab 3 10 14 79.2 101.3 25 24
Colombia (1978-88)' 10 24 25 6.0 7.5 36 38
Costa Rica (1977-83)' 6 29 36 0.6 0.9 44 39
Costa Rica (1983-86)' 3 36 24 0.9 0.6 39 44
Côte d'Ivoire (1985-86) 1 30 31 3.1 3.3 33 26
India (1977-83) 6 50 43 324.9 315.0 29 28
Indonesia (1984-87) 3 28 17 45.4 30.0 24 17
Malaysia (1984-87)' 3 15 14 2.3 2.2 26 24
Pakistan (1979-84) 5 21 20 17.1 18.7 19 19
Poland (1978-87)' 9 9 23 3.3 8.6
Thailand (1981-86y' 5 20 26 9.5 13.6 27 35
Venezuela (1982-87)' 5 12 16 1.9 3.0 26 31
Yugoslavia (1978-87)' 9 17 25 3.8 5.7
Note: See note to Table 3.2 for definitions. The range of poverty lines, expressed in terms of expenditure per household member and in PPP
dollars, is approximately $300-$700 a year in 1985 except for Costa Rica ($960) and Malaysia ($1,420).
Measures for this entry use income rather than expenditure.
Rural only.
Measures for this entry are by household rather than by household member.
Table 3.4 Have social services reached the poor in developing countries?
Latin
Sub- Europe, Middle America All
Saha ran East South East, and and the developing
Indicator (1985) Africa Asia Asia North Africa Caribbean countries
Percentage of nonpoor in population 53 79 49 75 81 67
Primary net enrollment rate (percent) 56 96 74 88 92 84
Percentage of children immunized 47 73 43 63 65 58
43
Figure 3.2 The impact of recession on social indicators in developing countries
Under 5 mortality rates
Deaths per thousand children under 5, 1975 - 80 Percentage reduction between 1975 - 80 and 1980 - 85
East Asia
Latin America
and the
Caribbean
Middle East
and
North Africa
South Asia
Sub-Saharan
Africa
200 150 100 50 0 0 5 10 15 20 25
Primary enrollment rates
Net enrollments as percentage of children age 6- 11, 1980 Percentage increase between 1980 and 1985
East Asia
Latin America
and the
Caribbean
Europe,
Middle East,
North Africa
r-
South Asia
Sub-Saharan
Africa
100 90 80 70 60 50 -10 -5 5 10
greatest needs, Sub-Saharan Africa. That region example, recorded higher rates of decline in the
had the highest mortality for children under 5 and late 1970s and early 1980s than in the 1960s and
the lowest enrollment rate at the start of the pe- 1970s.
riod; it saw the smallest improvement in under 5 Changes in nutritional status are another mea-
mortality of all the regions, and its enrollment rate sure of the impact of recession. Data on nutrition
actually declined. By contrast, irs several Latin are not always national in coverage, but they are
American countries under 5 mortality declined at nonetheless of special interest because they are of-
an increasing rate. Brazil, Chile, and Colombia, for ten drawn from areas in difficulty. Evidence from
44
the regions most affected by recession echoes the
patterns already noted. In Latin America malnutri- Figure 3.3 Trends in malnutrition in the 1980s:
tion has continued to decline steadily in Chile, Co- Ghana and Lesotho
lombia, and Costa Rica, but progress has halted in
Uruguay and Venezuela, and there are signs of Percentage of children underweight
worsening in Guatemala and Peru. In Sub-
Saharan Africa ma'nutrition is on the rise, and the
number of underweight children has grown sub-
stantially. Two broad patterns can be discerned. In Ghana
Benin, Burkina Faso, Ghana, and Togo malnutri- (age 7-48 months)
45
tion rose sharply in the mid-1980s, then declined
until 1986, and is now rising again. Ethiopia, Le-
sotho, Madagascar, Niger, and Rwanda have seen
a more persistent trend of rising malnutrition
around marked seasonal fluctuations. (The pat-
terns for Ghana and Lesotho are shown in Figure
3.3.) Lesotho
(age 0-5 years)
25
An overall assessment
Although circumstances vary enormously from
country to country, this review shows that there 15
has been a significant long-run improvement in
under 5 mortality, life expectancy, and primary en- 1980 1982 1984 1986 1988
rollment. In these respects the situation in the
early to mid-1980s was substantially better in every Note: The percentage of children underweight (less than
80 percent of standard weight-for-age) is used here as an
region than in the 1960s and 1970s. Except in Sub- indicator of malnutrition.
Saharan Africa, the incomes of the poor have also
risen, in some cases substantially.
In recent years, however, much depended on
where the poor lived. Most of the poor became Table 3.5 Variation in country experience
better off on every measure, but that is because Average annual reduction
most of them live in South Asia and East Asia. In Under 5
many countries of Latin America and Eastern Eu- Headcount index mortality
Country and period (percentage points)' (percent)
rope the incomes of the poor have declined
although social indicators have proved more resil- Indonesia (1970-87) 2.34 (58) 3.3 (146)
Malaysia (1973-87) 1.66 (37) 3.7 (46)
ient and in some countries have improved faster Brazil (1960-80) 1.45 (50) 2.8 (107)
than before the recession of the 1980s. Sub- Pakistan (1962-84) 1.43 (54) 1.8 (200)
Saharan Africa's poor fell further behind in the Costa Rica (1971-86) 1.41 (45) 9.3 (35)
1980s; the decline that started in the 1970s has con- Thailand (1962-86) 1.40 (59) 4.4 (70)
India (1972-83) 1.04 (54) 1.8 (199)
tinued and in some cases has accelerated.
Colombia (1971-88) 0.91 (41) 7.2 (64)
Regional differences aside, the review also Morocco (1970-84) 0.64 (43) 5.6 (136)
showed that some countries have been much more Sri Lanka (1963-82) 0.51 (37) 2.8 (66)
successful than others in reducing poverty. Table Note: Use of the income shortfall rather than the headcount index
3.5 shows this for two key indicatorsthe head- yields essentially the same ranking.
Initial level in parentheses.
count index and the under 5 mortality ratefor ten 1975-80 rate in parentheses.
countries. The next part of this chapter draws on 1977-81.
country experience to identify the factors underly-
ing these differences.
the rate and pattern of economic growth. It has a
Factors influencing country performance more direct effect on social indicators, mainly
through the government's expenditure program.
Public policy is critical both in reducing poverty, as The two issueshigher incomes and improved so-
measured by income, and in improving social indi- cial indicatorsare clearly linked. Each supports
cators. Policy affects incomes indirectly, through the other in a variety of ways. But since the direct
45
Table 3.6 Social sector spending as a percentage cial services. Within the sample of ten countries
of GNP identified in Table 3.5, public spending is an im-
portant part of the explanation for variations in
Sector and country 1975 1985 mortality of children under 5. The best performers
Primary education are Sri Lanka among low-income countries and
High enrollment rate Costa Rica in the middle-income group. Both
Botswana 2.1 2.8 countries have had a long-standing commitment,
Tunisia 1.9 2.3 dating to the first half of the century, to providing
Chile 1.4 2.2 social services.
Thailand 1.6 2.0
Low enrollment rate
Raising incomes
Nepal 0.7 1.0
Ghana 1.1 0.7
Pakistan 0.6 0.7 Isolating the factors that influence the incomes of
Haiti 0.6 0.6 the poor is more complex than is the case with
Health social indicators. As a preliminary step, changes in
Low under 5 mortality the incomes of the poor can be "explained" by
Chile 2.5 2.1 decomposing them into the part attributable to
Mauritius 2.0 2.0
Malaysia 1.9 1.8
overall economic growth and the part attributable
Sri Lanka 1.7 1.3 to changes in the inequality of income (Box 3.3). It
High under 5 mortality
has been argued that inequality worsens at first
Ethiopia 0.8 1.2 with development and improves only later. This
Burkina Faso 0.8 0.9 idea was encapsulated in the "Kuznets curve,"
India 0.3 0.3 which posits an inverted-U relationship between
Pakistan 0.3 0.2 inequality and average income. The curve implies
the possibility of a conflict between growth and
poverty. In the extreme, the inequality of income
may worsen fast enough at the outset of economic
effect of policy on social indicators is easier to mea- growth for poverty to increase; growth would be
sure, that is a good place to start. "immiserizing."
Table 3.7 presents a simple test of this view. It
Improving social indicators shows the change in poverty, as measured by the
headcount index of Tables 3.2 and 3.3, and com-
The countries that have succeeded in providing pares it with a simulated change in poverty. This is
primary education and health care to the poor are the change that would have occurred if inequality
those that have made adequate provision for the had remained constant-that is, if everyone had
purpose in their budgets (Table 3.6). In 1985 received the same percentage increase in income.
spending on primary education as a percentage of If the actual decrease in poverty is less than the
GNP was more than four times higher in Bo- simulated decrease, growth has increased inequal-
tswana, where the enrollment rate was 99 percent, ity, and the poor have gained less than the non-
than in Haiti, where it was only 55 percent. Simi- poor. If the actual decrease is bigger than the simu-
larly, in countries that have achieved broad provi- lated decrease, the opposite is true.
sion of health care, such as Chile and Mauritius, Table 3.7 looks both at periods of long-term
spending as a percentage of GNP is several times growth and at short-term recessions. In general,
greater than in countries such as India and Paki- long-term growth has had only a limited effect on
stan, where under 5 mortality remains exception- inequality, but it has tended to reduce it. In India,
ally high and the percentage of children immu- for example, the actual and simulated changes in
nized is low. poverty are almost exactly the same. This implies
But increased government spending is not al- that the poor enjoyed approximately the same per-
ways the answer to improving the well-being of centage increase as everyone else and that the in-
poor people. Better allocation of expenditures come attributable to growth was distributed in the
within the sector and more efficient use of funds same way as the initial income.
are often more important. It is clear, however, that In some countries-Colombia, for example-
through well-directed public expenditures even long-term changes in inequality have been more
low-income countries can dramatically improve so- important. But these and other cases lend no sup-
46
Box 3.3 The mechanics of changes in poverty
For any given increase in the incomes of the poor, the
reduction in poverty depends on where the poor are in Box figure 3.3 Different initial conditions: the impact
relation to the poverty line. If they are concentrated on poverty reduction
just below the line, the increase in their incomes will
have a bigger effect on poverty than if they are spread
Heavy concentration of poor
more evenly. just below poverty line
Box figure 3.3 shows the cumulative distribution
functionthat is, the percentage of persons who re- Percentage of population
ceive no more than a particular income, expressed as a 100
function of that income. For example, with the poverty
line set at 30, the curve on the left in each figure shows Initial
that 50 percent of the population is poor. A 50 percent distribution
75
increase in income will shift the distribution function to
the right. The reduction in the incidence of poverty is
37 percentage points in the upper panel of the figure Final
but only 27 percentage points in the lower panel. 50 distribution
The difference in outcome arises from differences in
the slope of the distribution function at the poverty 37
line. If the slope is very steep (implying less inequality 25
I
in the region of the poverty line), as in the upper panel, Poverty
a large number of people is concentrated just below the line
line. An increase in income moves many people above I I
the line, and the incidence of poverty falls substan- 0 10 20 30 40 50 60 70 80
tially. If the slope is less steep (implying greater in- Income per capita
equality around the poverty line), as in the lower
panel, few people are located immediately below the
Lower concentration of poor
poverty line. In this case the same increase in income just below poverty line
moves only a few of the poor above the line, and the
reduction in the incidence of poverty will be much Percentage of population
smaller. 100
/
For example, starting from the latest available distri-
Initial
butions, a 10 percent increase in the incomes of the distribul on
poor in Bangladesh and India would reduce the inci- 75
dence of poverty by about 7 percentage points. Where
the distribution of income is more unequal, as in Vene-
zuela and Brazil, the corresponding figure would be 50 Final
only 3 percentage points. distribution
27
2
Poverty
line
I I I
0 10 20 30 40 50 60 70 80
Income per capita
I
port to either the Kuznets or the immiserizing- sis). In short, growth reduces poverty. Even so,
growth hypothesis. In the low-income countries where inequality has worsened, as in Brazil, the
inequality consistently improves (contrary to the implications for poverty are significant. If inequal-
Kuznets hypothesis), and there is no case in which ity had declined as in Malaysia, Brazilian poverty
the effect of growth is offset by changes in inequal- would have fallen by 43 percentage points be-
ity (contrary to the immiserizing-growth hypothe- tween 1960 and 1980 rather than by 29. The pattern
47
Table 3.7 Poverty, economic growth, and recession
Observed Simulated Annual growth
Length reduction reduction of mean income
of period in poverty in poverty or expenditure
Country and period (years) (percentage points)a (percentage points)b (percent)
Long-run growth
Indonesia (1970-87) 17 41 35 3.4
Thailand (1962-86) 24 33 30 2.7
Pakistan (1962-84) 22 31 26 2.2
Brazil (1960-80) 20 29 34 5.1
Malaysia (1973-87) 14 23 19 4.0
Singapore (1972-82) 10 21 19 6.4
Costa Rica (1971-86) 15 21 22 3.5
Colombia (1971-88) 17 16 8 1.1
India (1972-83) 11 11 10 1.0
Sri Lanka (1963-82) 19 10 8 0.9
Morocco (1970-84) 14 9 1 0.2
Short-run recession
Costa Rica (1983-86) 3 12 13 10.9
Indonesia (1984-87) 3 11 9 5.0
India (1977-83) 6 7 2 0.8
Malaysia (1984-87) 3 1 1 0.7
Pakistan (1979-84) 5 1 4 1.2
Colombia (1978-88) 10 1 1 1.2
Côte d'Ivoire (1985-86) 1 1 5 5.4
China (1985_88)' 3 4 5 6.7
Brazil (1981-87) 6 5 1 0.9
Venezuela (1982-87) 5 5 6 4.5
Thailand (1981-86) 5 6 0 0.0
Costa Rica (1977-83) 6 7 8 3.4
Yugoslavia (1978-87) 9 7 12 2.9
Poland (1978-87) 9 14 17 1.2
Absolute change in the headcount index on the basis of the definition of absolute poverty in the specific country.
The simulation assumes that the inequality of income remains unchanged.
Rural only.
of growth as well as its rate is thus an important average percentage increase in income in all coun-
determinant of changes in poverty. tries except Brazil and Costa Rica, where inequality
As the lower part of Table 3.7 shows, in the 1980s worsened. In Brazil the poor nevertheless enjoyed
the link between growth and poverty reduction is a substantial increase in income; in Costa Rica they
still there, but it is weaker than before. By and suffered a loss. In general, therefore, the poorest
large, economic growth reduces poverty and eco- of the poor participated fully in economic growth.
nomic decline increases it. Fluctuations in inequal- No simple pattern emerged during the 1980s. In
ity, however, were larger in the 1980s. In Malaysia, about half the countries the very poor suffered
for example, poverty decreased even though mean more, or advanced less, than the average citizen.
income also declined. This suggests that external In Colombia, for example, mean income for the
shocks or important policy changes can alter the entire population fell by 11 percent between 1978
incidence of poverty by way of changes in the in- and 1988, whereas for the poorest tenth it fell by
equality of income, whereas in more stable periods more than 20 percent. In contrast, the poorest
economic growth is the dominant influence on households in other countries did much better
poverty. than the rest of the population. In Malaysia aver-
Would the conclusion that growth reduces pov- age incomes fell by 2 percent between 1984 and
erty change if attention were shifted from the poor 1987, but the mean incomes of the poorest actually
to the very poor? The country-specific poverty increased by 9 percent.
lines used in this analysis define 20 to 50 percent of These conclusions shed some light on the differ-
the population as poor. If we turn to the poorest mg experience of the countries identified in Table
tenth of the population, we find that in periods of 3.5. Economic growth was clearly importantin
stable growth this group enjoyed a larger-than- the two countries that experienced the fastest re-
48
duction in poverty, per capita income grew by
more than 3 percent a year, whereas the two coun- Figure 3.4 Inequality and wages,
tries that reduced poverty least had growth rates of Colombia, 1965 to 1986
less than 1 percent. In egalitarian economies (Indo-
nesia, for example) economic growth that pre- Index (1975 = 100) Gini coefficient
served the existing degree of inequality was suffi-
175 0.65
cient to reduce poverty quickly. Where the initial
distribution of income was less equal, as in Colom-
bia, changes in inequality were an important com-
150 0.60
plement to overall growth.
Economic growth and changes in inequality,
however, are not instruments of policy; they are Gim
125 coefficient
consequences. What were the policies that, in egal- / 0.55
itarian economies such as Indonesia, spurred
growth and even slightly reduced the initial degree
of inequality? And what were the policies that, in 100 0.50
inegalitarian economies such as Colombia, enabled
the poor to benefit more than proportionately from
growth? The answer lies in the factors that deter- 75 0.45
mine the incomes of the poor. Real
unskilled wages
OWNERSHIP OF ASSETS. Redistributing the exist- 50 0.40
ing stock of assets to the poor has sometimes
proved successful, but this is the exception rather 1964 1967 1970 1973 1976 1979 1982 1985 1988
than the rule. Large redistributions of land have
been associated with rapid reductions in poverty,
but they have occurred only in times of great politi-
cal upheaval. Distribution of new capital (that is,
investment) in favor of the poor has been more
common. In particular, investment in human capi- place at a time when agricultural purchasing
tal through primary health care and education has power was growing at 1.5 percent a year. Econo-
been an important part of the approach adopted by metric analysis confirms that rural poverty in India
several countries. Whether this raises incomes, has a strong negative link to farm incomes.
however, depends on the opportunities that are In India and Indonesia the initial distribution of
available for using the new skills. income was fairly equal, and growth was able to
reduce poverty even though the degree of inequal-
RETURNS TO ASSETS. Since poverty is largely a ity remained the same. Where the initial distribu-
rural phenomenon and since many of the poor de- tion is less equal, growth must shift the pattern of
pend, directly or indirectly, on the farm sector for returns in favor of the poor if it is to have much
their incomes, growth that raises agricultural pro- effect. Colombia is one of the few Latin American
ductivity and the return to farm labor ought to be countries that has managed to improve an initially
particularly effective in reducing poverty. The con- very unequal distribution of income. Its Cmi coef-
trast between Indonesia and India illustrates this ficient (a measure of inequality in which 1 repre-
point. Between 1970 and 1987 poverty in Indonesia sents complete inequality and 0 perfect equality)
declined by 41 percentage points; over the same fell from 0.54 to 0.48 between 1971 and 1978, and
period the purchasing power of agricultural value its incidence of poverty fell by 2.9 percentage
added rose by 2.6 percent a year per rural dweller. points a year. As Figure 3.4 shows, the return to
Between 1984 and 1987, a period of especially rapid poor people's most important assetunskilled
decline in poverty, purchasing power grew by 5.0 laborhad to rise sharply for this to come about.
percent a year. In contrast, in India poverty de- Between 1971 and 1978 real unskilled wages in-
creased by 11 percentage points, and agricultural creased at 6.6 percent a year, as against 3.9 percent
purchasing power grew by less than 0.4 percent a for GNP per capita. This shifted the distribution of
year. Most of the decline in poverty in India-7 income in favor of the poor. Box 3.4 illustrates
percentage points between 1977 and 1983took what happened to wages and poverty in a period
49
Box 3.4 Growth, real wages, and poverty: the United Kingdom and the United States,
1770 to 1920
The history of the industrial revolution in the United by 1.3 percent a year for the next forty years.
Kingdom and the United States suggests links among In both countries technological advances initially fa-
growth, real wages, and poverty. In both countries de- vored capital-intensive and skill-intensive industry
velopment in the early phase of the revolution was over labor-intensive agriculture. Slow growth in labor
capital-intensive. Since at the same time the labor sup- demand coincided with dramatic population growth to
ply was increasing, the real wages of unskilled workers restrict the growth of real wages. Several decades after
grew slowly, and economic growth had only a small the start of the industrial revolution, technological
effect on poverty. After about 1820 in the United King- progress in farming led to a more balanced pattern of
dom and 1880 in the United States, however, real growth, and the labor-saving bias of early industrializa-
wages began to rise, and poverty began to decline. tion gave way to a neutral or labor-intensive bias.
Britain's industrial revolution began around 1770, Lower birthrates and stricter immigration laws slowed
but until 1820 real wages barely increased. In the first population growth, and real wages increased at a faster
twenty years of the nineteenth century the earnings of rate.
adult male unskilled workers grew at just 0.2 percent a In the United Kingdom pauperism declined after
year. The next fifty years saw a much faster and stead- 1840. The most reliable data for the United States, from
ier increase at 1.7 percent a year. After about 1840 the records in New York State, suggest that poverty in-
GDP of the United States grew significantly faster than creased up to 1865, when 8 percent of the population
that of the United Kingdom at a comparable stage, but was receiving local relief. After that, poverty declined
real wages for urban unskilled labor increased by less until the end of the century. In both countries growth
than 0.2 percent a year between 1845 and 1880. Then, in the real wages of unskilled labor reduced the inci-
as in the United Kingdom, they accelerated and grew dence of poverty.
of economic growththe industrial revolutions in Another way of increasing the productivity of
the United Kingdom and the United States. the assets of the poor is to provide better physical
Returns to the assets of the poor can also be in- infrastructure. For example, Indonesia used its oil
creased through higher productivity. In Malaysia receipts to improve and extend infrastructure
investing in the poor contributed to rapid growth, throughout rural Java and thus provide access to
with some improvement in a fairly unequal distri- markets for most of the country's poor. In contrast,
bution of income and a decrease in poverty of 23 the inadequate infrastructure in much of Sub-
percentage points between 1973 and 1987. During Saharan Africa continues to deteriorate.
this period average labor earnings for rural work-
ers rose by almost 75 percent. Studies suggest that TRANSFERS. All the countries listed in Table 3.5
better education accounted for roughly one- have used transfers, usually in the form of food
quarter of the rise in farm earnings and about subsidies. In countries such as Indonesia and Thai-
three-quarters of the rise in nonfarm earnings. By land moderate transfersabout 0.2 percent of
investing in the human capital of the poor and GDPhave proved to be consistent with rapid
providing an environment in which new skills growth. Other countries have gone much further.
could be productively used, Malaysia improved In the late 1970s Egypt's transfers (which, al-
both the short-run welfare of the poor and the though they reached the poor, went mainly to the
prospects for raising incomes in the medium term. nonpoor) amounted to 7 percent of GDP. Such
Sri Lanka has been less successful in realizing heavy use of transfers can cause severe macroeco-
the potential increase in productivity made possi- nomic difficulties, and by reducing growth, it con-
ble by social spending. Between 1970 and 1985 it demns future generations to poverty.
allocated about 10 percent of GDP to social expen- Transfers are unlikely to be the answer to
ditures, and coverage was widespread. Although povertycertainly not in low-income countries
social indicators improved, the incomes of the with large numbers of poor. One reason is the
poor did not rise significantly. Poverty declined by sheer size of the problem. Transfers of at least 15
only 0.5 percentage points a year during the 1960s percent of current GDP would be needed to elimi-
and 1970s. nate poverty in Bangladeshand this assumes that
50
the transfers would not "leak" (as they usually do) in povertythe headcount index remained
to the urban middle class. In some middle-income roughly constant in Colombia and increased in
countries the situation is different; transfers Costa Rica. Still, under 5 mortality continued to
amounting to only 1.1 percent of current GDP fall, demonstrating that efforts to improve social
could eliminate poverty in Brazil. Because of leak- indicators can yield results even during a reces-
age and administrative costs, however, much more sion. Similarly, the experience of Sri Lanka shows
than this would be required in practice. that remarkable social progress can be achieved
even at low levels of income. The benefits of Sri
Explaining successful performance Lanka's long-standing support for the social ser-
vices can be seen in its under 5 mortality rate,
Armed with this account of the factors that have which was 66 per thousand in 1980an impressive
influenced poverty, we can review the perfor- achievement for a low-income country.
mance of the countries listed in Table 3.5. The Yet, as the experiences of India, Morocco, and
three East Asian countriesIndonesia, Malaysia, Sri Lanka show, low GDP growth makes it difficult
and Thailanddemonstrate the benefits of an ap- to reduce poverty. This suggests an important dis-
propriate balance between policies that spur tinction: raising the incomes of the poor (and thus
growth and policies that enable the poor to partici- lifting people above the poverty line) requires
pate in growth. All three achieved and sustained broadly based economic growth, but making the
annual GDP growth rates of more than 6 percent. poor better off in other respectsby reducing child
This growthrelatively labor-intensive, with agri- mortality, for instancecan be brought about
culture to the foregenerated demand for the fac- through specific public actions.
tors of production owned by the poor. These coun- A key conclusion emerges from all this: the
tries also provided for adequate social spending. countries that have been most successful in attack-
As a result, they have achieved universal primary ing poverty have encouraged a pattern of growth
education, and their infant mortality rates are that makes efficient use of labor and have invested
lower than those of many countries with similar in the human capital of the poor. This two-part
incomes. The improvement in the skills and qual- approach is the basic strategy for the reduction of
ity of the labor force enabled the poor to seize the poverty proposed in this Report. Both elements
opportunities provided by economic growth. are essential. The first provides the poor with op-
In other countries the creation of opportunities portunities to use their most abundant asset
for the poor and the development of their capacity labor. The second improves their immediate well-
to respond have not always been as well balanced. being and increases their capacity to take
Brazil's GDP growth exceeded that of every other advantage of the newly created possibilities. To-
country in the sample, and Pakistan equaled the 6 gether, they can improve the lives of most of the
percent annual growth achieved by the East Asian world's poor.
countries. Yet in neither country did social indica- Some among the poorfor example, the infirm,
tors improve rapidly. Brazil has one of the highest the aged, and those in resource-poor areasmay
mortality rates for children under 5 among the not benefit even from successful implementation
middle-income countries, and Pakistan has one of of the two-part approach. Others, although bene-
the lowest rates of primary school enrollment in fiting, wifi remain highly vulnerable to personal
the world. The failure to improve the skills of the disasters such as death of the family breadwinner
labor force has limited poor people's ability to ben- and to national calamities such as drought or eco-
efit from growth. In each case the headcount index nomic recession. A comprehensive approach to
fell, but less quickly than in Indonesia or Malaysia. poverty, therefore, requires that the basic strategy
So, it is possible to have economic growth with- be supplemented by a system of well-targeted
out much social progress. The converse is also transfers and safety nets.
true: social indicators can be improved even in the
absence of rapid economic growth. Between 1971 What is the critical tradeoff?
and 1978 in Colombia and between 1971 and 1977
in Costa Rica poverty declined rapidly, under 5 Discussions of policy toward the poor usually fo-
mortality fell at an extraordinary rate, and GDP cus on the tradeoff between growth and poverty.
grew at 5 to 6 percent a year. The recession of the But the review of country experience suggests that
early 1980s brought stagnant or declining GDP per this is not the critical tradeoff. With appropriate
capita and put an abrupt halt to the improvement policies, the poor can participate in growth and
51
Box 3.5 Political coalitions and the poor
Bringing about reforms that are intended to reduce and organizational capacity to lobby successfully for
poverty is not necessarily a matter of simply pitting the greater investments in the development of human re-
poor against the nonpoor. Although many economic sources. Kenya and Sri Lanka, which spend relatively
policies benefit the rich at the expense of the poor, high amounts on primary education, not surprisingly
others link the fortunes of both groups and can draw have powerful teachers' unions. The expansion of pri-
support from coalitions that cut across the poor- mary education in Peru was largely attributable to the
nonpoor divide. efforts of political parties to win teachers' votes.
Policy coalitions often form across the income spec- In Argentina, Chile, and Peru the success of tax and
trum when sectoral interests are at issue. Tariff, ex- other policy reforms to benefit the poor has generally
change rate, and food pricing policies often have ef- turned on the stance of white-collar workers, profes-
fects that differentiate more between the agricultural sionals, bureaucrats, and small- and medium-size busi-
and industrial sectors than across income levels. In ness interests. Redistributive policies were more likely
many African and Latin American countries the agri- to succeed when these sectors shared in transfers di-
cultural sector has long suffered from policies that fa- rected primarily to the poor. The same is no doubt true
vor industry and cities. For example, food prices are in many other countries. The Maharashtra Employ-
frequently kept low, which benefits the urban poor, ment Guarantee Scheme in India transfers income
industrial workers, and enterprise owners at the ex- from the urban nonpoor to the rural poor, but it never-
pense of the entire rural sector, including the rural theless enjoys wide political support. The urban non-
poor. poor see the reduction of migration to Bombay as a
Coalitions of the poor and nonpoor also form along benefit, and landowners may look favorably on the
geographic linesto promote the flow of resources to scheme because it helps to stabilize the rural labor force
an entire region, such as Brazil's Northeast, or to push and because it creates infrastructure in the countryside.
for interventions such as an irrigation project that will By contrast, when Sri Lanka switched from a general
help a specific locale. Entire districts in India have food subsidy and ration scheme to a food stamp pro-
pressed to be included within irrigation command ar- gram during 1979-80, the fall in implicit transfers to the
eas in the expectation of increased productivity for nonpoor undermined support for the scheme.
both large and small farmers. In sum, the political economy of poverty reduction is
Service providers and recipients may also form coali- complex and varies greatly according to country and
tions. Pressures on governments to finance social ser- historical circumstances. As this brief review illus-
vices often come as much from the middle-income pro- trates, policies to benefit the poor have been proposed
viders of services as from the beneficiaries. Teachers, and successfully implemented under a diversity of con-
medical personnel, social workers, and other middle- ditions. But experience shows that success is often built
and upper-income service providers themselves bene- on objectives and strategies that are shared to some
fit when the government devotes more resources to degree by the poor and at least some groups among the
social services, and they often have the voting power nonpoor.
contribute to it, and when they do, rapid declines strong influence on policy. Giving the poor a
in poverty are consistent with sustained growth. greater say in local and national decisionmaking
If policies can be identified that both reduce pov- would help to restore the balance. But since politi-
erty and contribute to growth, why have more cal power tends to reflect economic power, it is
countries not adopted them? The answer lies in important to design poverty-reducing policies that
political feasibility. The strategic choices that gov- will be supported, or at least not actively resisted,
ernments make reflect both economic and political by the nonpoor. Sometimes it is possible to build
factors. Countries differ enormously in their politi- coalitions that bring together the poor and certain
cal culture, in the nature of their political organiza- nonpoor groups that have an interest in reform
tions, in their leadership, in their bureaucratic pro- (Box 3.5). As a rule, however, avoiding resistance
cesses, and so forth. This section investigates one by the nonpoor will call for policies that put the
aspect of the complex interaction between political least burden on the majority.
and economic factors.
Increasing assets
Policies that help the poor but impose costs on
the nonpoor will encounter resistance whether or Governments that seek to increase the assets of the
not they increase national income. The nonpoor poor have pursued two approachesredistri-
are usually politically powerful, and they exert a bution of existing assets (such as land) and in-
52
creased public investment in the human capital of effects on the distribution of income and thus in
the poor. Both policies are beneficial to the poor. their political feasibility (see the upper panel of
The experience of Japan and Korea provides an Figure 3.5). Land redistribution causes an immedi-
example of significant land redistributions that led ate and permanent loss of consumption for the
to a marked and sustained decline in poverty. Co- nonpoorwhich is why it is resisted so strongly.
lombia and Malaysia illustrate the benefits of in- Extra spending on education can be built up more
creasing human capital. gradually so that the increased burden of taxation
The lower panel of Figure 3.5 uses the frame- on the nonpoor is paid entirely out of increments
work described in Box 3.6 to illustrate the benefits to income. The second approach, therefore, is
to the poor under each approach. Land redistribu- likely to be more feasible. Moreover, some non-
tion brings art immediate and lasting gain; higher poor groupsfor example, industrialists who need
spending on education brings no improvement in a skilled labor forcestand to gain from the human
the short run but a rapid increase later. (The poli- capital policy and would be likely to support it.
cies were calibrated to ensure the same increase in This does not imply that governments should
the incomes of the poor by the tenth year.) reject asset redistribution or that asset redistribu-
Do the two policies' effects on long-term growth tion and investment in the poor are mutually ex-
make one preferable to the other? In principle, clusive. Indeed, a relatively egalitarian distribution
both policies have growth-increasing and growth- of land increases the effectiveness of other policies
reducing effects. The land reform, for instance, aimed at reducing poverty. It does suggest, how-
might reduce saving (since the poor tend to save a ever, that tilting the distribution of new investment
smaller proportion of their income than the rich), in favor of the poor (as advocated in Redistribution
and that would be bad for growth. But it might with Growth, by Chenery and others) is likely to be
also encourage the more efficient use of land, more popular than reshuffling the stock of existing
which would be good for growth. Investment in assets. If redistribution is impossible, the case for
education improves the quality of the work force spending more on education and other forms of
(good for growth) but requires, at least in the short investment in human capital is all the stronger.
term, higher taxes to finance the increase in public
Increasing income
spending (bad for growth). On balance, there may
be little to choose between the two approaches. A similar comparison can be made between a cur-
The policies, however, differ strikingly in their rent transfer of income and growth that raises the
IBox 3.6 Tradeoffs and poverty: a simple simulation model
The simulation model underlying the results of this transferring income from the nonpoor to the poor will
section captures two tradeoffsthat between poverty reduce national saving. This in turn will mean slower
and growth and that between the poor and the non- economic growth. A net transfer of 0.5 percent of GNP
poor. The severity of the tradeoffs depends on two fac- reduces the growth rate of GNP by 0.1 percentage
tors. points and that of the consumption of the nonpoor by
The first is the immediate economic cost of raising 0.12 percentage points. The productivity of capital,
the revenues necessary to finance transfers and social however, is assumed to be independent of ownership.
programs. Higher taxes are likely to reduce economic Thus, increased investment in the human capital of the
efficiency (and hence GNP), at least to some extent. poor, for example, will increase growth to the same
Moreover, not all of the increased public spending will extent as investment elsewhere in the economy.
find its way to the intended beneficiaries. Similarly, The model depicts growth in a one-sector economy.
some of the rise in taxes will fall on the poor. So, to It consists of a single production function that com-
deliver a net benefit of $1 to the poor, more than $1 will bines factors of production and yields national income;
have to be raised in taxes. In the model, a net transfer a submodel that distributes national income to the poor
of 0.5 percent of GNP to the poor reduces GNP by 0.4 and the nonpoor according to their ownership of fac-
percent and the consumption of the nonpoor by 1.0 tors of production; a set of taxes and transfers that
percent. Policies that allow a reduction in taxation, determines disposable income by income class; saving
such as the elimination of subsidies to capital, have the rates by income class; and a "cost" function that mea-
reverse effect. sures the loss of GNP arising from taxation.
Second, if the nonpoor save more than the poor,
53
Figure 3.5 Increasing the assets of the poor Figure 3.6 Increasing current income
140 140
Political costs
Consumption per capita of the nonpoor
130 130 Political costs
Consumption per capita of the nonpoor
120 120
110 110
100 100
,
60 60
/
The impact on poverty
er capita of the poor
50 50 - The impact on poverty
JsPton Consumption per capita of the poor
40 40
30 30
20 20
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
Year Year
Base run - Base run
Increased investment in human capital Labor-intensive growth
Redistribution of land Transfer
Note: GNP per capita in year 0 is set at 100. Note: GNP per capita in year 0 is set at 100.
Consumption per capita in year 0 is equal to 80. Consumption per capita in year 0 is equal to 80.
returns to assets owned by the poor. Some coun- will depend on the interaction of many different
tries, such as Sri Lanka, have relied heavily on factors. By contrast, an increase in the transfer is
subsidies; others, such as Colombia, have in- likely to reduce both the level and the future
creased the return to labor through policies that growth of income. Higher taxes will be needed to
promote its more efficient use. Both approaches finance the increase, and long-term saving will be
can benefit the poor. In the lower panel of Figure lower; both are bad for growth.
3.6 the transfer has been chosen to ensure the This initial preference for the labor-promoting
same increase in income over the simulation pe- approach is greatly strengthened by consider-
riod as with policies that remove biases against the ations of political feasibility. Here, as before, the
use of labor. difference between the two policies is striking (see
As before, both policies affect growth. But in this the upper panel of Figure 3.6). Whereas transfers
case there is a stronger presumptionconfirmed reduce the present consumption of the nonpoor,
by the modelin favor of removing the biases switching to a more efficient use of labor increases
against the use of labor. This policy produces an their consumption initially and leads to only some-
immediate gain in efficiency, and the level of GNP what slower rises in the future. Transfers are there-
rises. Thereafter, the effect of the policy on growth fore far more likely to encounter political resis-
54
tance. Moreover, some among the nonpoor, such growth. Yet the poor are especially vulnerable dur-
as investors in labor-intensive industry, stand to ing adjustment. Protecting them at such times be-
gain from the removal of biases against the use of comes both more urgent and more difficult. Trans-
labor and may therefore support such reforms. fers must be targeted with special care if the poor
Again, however, the two approaches are not mutu- are to be protected at reasonable cost. The case for
ally exclusive. Indeed, some transfers will clearly transfers in the event of a temporary shock is much
be necessary for those who cannot participate in more clear-cut. The value of investment is not af-
growth. fected by a drought, for examplebut the value of
As with the review of country experience, this a transfer to prevent death from famine is certainly
analysis suggests that a more efficient use of labor, increased.
coupled with increased investment in the human The next four chapters turn from a broad view of
capital of the poor, reduces poverty and increases approaches to development to a more detailed dis-
national income. The two-part strategy proposed cussion of specific policies. How are governments
in this Report thus entails no tradeoff between in- to promote efficient, labor-intensive growth to-
come and growth on the one hand and poverty on gether with investment in human capital? Chapte
the other. The tradeoff between the poor and the 4 concentrates on the first element of the strategy
nonpoor remainsbut the strategy achieves a sub- by identifying the policies that have encouraged a
stantial reduction in poverty with a smaller cost to pattern of growth that uses labor more efficiently.
the nonpoor than under other approaches. It may Chapter 5 turns to the second component and ex-
therefore be politically more acceptable. amines the delivery of social services to the poor.
A tradeoff between growth and poverty does Chapter 6 discusses the role of transfers and safety
emerge during the economic restructuring that fol- nets in helping those who may not benefit from
lows, for instance, a permanent fall in the terms of the policies recommended in Chapters 4 and 5 as
trade. The need to change the pattern of produc- well as those who may suffer income-reducing
tion makes investment even more important than shocks. And Chapter 7 asks how all these policies
usual, and as a result, policies that reduce invest- need to be modified to cope with recession and
ment become more "expensive" in terms of future restructuring.
55
Promoting economic
opportunities for the poor
Ensuring that the poor participate in and contrib- Since labor is an abundant resource, encourag-
ute to growth requires the adoption of appropriate ing its use is generally consistent with rapid and
economywide and sectoral policies and of mea- efficient growth. Yet most countries have adopted
sures to help the poor grasp new income-earning policies that are implicitly biased against labor. In
opportunities. This chapter examines policies that particular, although agriculture is the principal
will: labor-intensive sector, almost all developing coun-
Encourage a pattern of growth that increases tries have taxed agricultural output. Governments
the efficient use of the assets owned by the poor have done this for two reasons: to finance public
Expand the access of the poor to land, credit, spending (it is administratively easy to tax agricul-
infrastructure, and productive inputs. tural output) and to protect manufacturing, espe-
cially in the early phases of industrialization.
As noted in Chapter 3, there are strong comple- Country experiences suggest two conclusions.
mentarities among growth, poverty reduction, and
human capital. Investment in human capital, Successful rural development entails avoiding
which is critical for enabling the poor to seize ex- excessive taxation of agriculture, providing strong
support for rural infrastructure, and making tech-
panded opportunities, is discussed in Chapter 5.
nical innovations accessible to small farmers.
The pattern of growth and the incomes A rising urban demand for labor (in industry
of the poor and services) plays an increasing role as incomes
rise. Governments can foster urban job creation by
Most of the poor in developing countries depend avoiding severe distortions in product and factor
on income from laborfrom work on their own markets and by providing suitable urban infra-
structure.
land, from wages, or from other self-employment.
The countries that have succeeded in reducing Government policy and rural development
poverty over the long term have encouraged
broadly based rural development and urban em- The expansion of agriculture is the driving force
ployment, thereby increasing the returns to small- behind effective rural development, which in turn
farm production and wage labor. Successful ap- lays the foundation for broadly based, poverty-
proaches to development, however, have varied reducing growth. The poor benefit directly if they
according to the initial economic conditions. In are farmers, and they benefit indirectly from
Thailand in the 1960s and Kenya in the 1970s the growth in demand for farm labor and for the prod-
rising productivity of small farms spurred growth ucts of the rural nonfarm sector. Especially in low-
in nonfarm incomes. In Malaysia in the 1970s ex- income countries, there is a strong association be-
pansion of urban employment played an equally tween growth in agricultural purchasing power
important role. and rural wagesa key welfare indicator for the
56
sia enjoyed an agricultural growth rate of 5 percent
Figure 4.1 Agricultural purchasing power and a year over that decade. Ghana, in contrast, taxed
rural wages agricultural commodities to the tune of 63 percent
(percent) and spent only 3 percent of value added on sup-
port. Its farm output fell by more than 1 percent a
Annual growth in rural wages
year.
8
In some cases taxes can be offset by effective
public support. Thailand's total taxation on agri-
cultural commodities amounted to 43 percent of
6 output value, with the burden falling mainly on
rice and rubber. But substantial public support for
Sri Lanka
4 China infrastructure and services made up for these out-
flows. Figure 4.3 shows the pattern of gross and
Indonesia
India Mauritius
net flows to the sector. It excludes spending on
2
Bangladesh Kenya
Morocco
0
Figure 4.2 Direct and indirect taxation of
Nigeria agriculture in the 1970s
-2
Ghana Malawi
Percentage of agricultural value added
-4 I I
0 10 20 30 40 50 60 70
10 -5 0 5 10 15
Annual growth in agricultural
purchasing power Malaysia
Note: The period covered varies from country to country
but is early 1970s to mid-1980s except for Bangladesh and Chile
China, 1980-87, and Kenya, 1970-79.
Morocco
Colombia
rural poor (see Figure 4.1). Moreover, agricultural Thailand
growth helps the rest of the economy. Typically (as
Sri Lanka
noted in World Development Report 1986) countries
with rapid agricultural growth have also had rapid
Pakistan
industrial growth.
The main policies that affect agricultural perfor- Egypt
mance are taxation and public support for agricul-
tural development. In quantifying taxation it is im- Cóte d'Ivoire
portant to take account of both direct taxes on
agricultural products and the indirect taxes that are Ghana
implicit in industrial protection and overvaluation
of the exchange rate. As Figure 4.2 shows, these
indirect taxes can add substantially to the overall D Direct taxes 0 Indirect taxes
tax burden.
Note: Direct taxes are calculated as the difference between producer
The range of experience is wide, but countries prices (in relation to the price of nonagricultural goods) and the
that have performed well have usually taxed mod- relative border price at the official exchange rate. Indirect taxes are
the additional reduction in relative prices attributable to the effects
erately and provided strong support. For example, of industrial protection on the equilibrium exchange rate and on
the prices of nonagricultural goods. Both direct and indirect taxes
in the 1970s total taxation of agricultural commodi- are expressed as a ratio of the prices without any distortion and at
ties in Malaysia was a relatively low 19 percent of the equilibrium exchange rate. The results are weighted averages
for those commodities for which taxes are reported in a given
the value of output (at international prices); gov- country.
Source: Schiff and Valdés forthcoming.
ernment spending for direct support of agriculture
was 10 percent of the sector's value added. Malay-
57
sector. By contrast, Nigeria, Senegal, and Tanzania
Figure 4.3 Net resource flows between the
sought capital-intensive industrialization. They
government and the agricultural sector, imposed heavy direct and indirect taxes on farm-
Thailand, 1964 to 1982 ing and provided only weak public support for in-
frastructure and institutions. Like Ghana in the
1970s, they failed dismally to spur growth in agri-
Millions of baht a
cultural output and in nonfarm output and total
8
labor demand.
Policy within the agricultural sector influences
not just the growth of the sector but also the extent
to which this growth reduces poverty. When prod-
uct pricing and support are biased in favor of large
farmers, the poor benefit less. For example, policy
in Malawi contains biases in the form of pricing
that discriminates in favor of large estates and
against small farmers in the tobacco sector. Small-
farm tobacco production has grown only slowly. In
Zimbabwe before independence the entire system
of infrastructure and services was intended to help
large farmers. When this bias was partially re-
dressed in the 1980s, smallholder maize and cotton
production grew rapidly. In many Latin American
countries, too, support services are designed
1964 1970 1976 1982 mainly in the interests of large farmers.
Other aspects of policy may add to the bias
Public expenditure on agriculture against labor. Credit subsidies, for instance, foster
Gross taxes excessive mechanization, and research suggests
D Net resource flow from government to agriculture that they have reduced the demand for farm labor
a. Deflated by rural consumer price index. in India. A highly unequal land distribution also
Source: Siamwalla and Setboonsarng 989. limits the extent to which the poor gain from agri-
cultural growth. Brazil shows that transfers of re-
sources to the sector can fail to have much effect on
poverty if policy within the sector is biased against
rural roads and the effects of the country's indirect the poor and if the poor have little access to land
taxation (which is about 15 percent of value (Box 4.1).
added), but the overall pattern is clear; substantial
public spending led to net flows into the sector INFRASTRUCTURE AND TECHNOLOGY. Public pro-
after the early 1960s. This spending supported ex- grams to provide services, infrastructure, research,
pansion of the land frontier and crop diver- and technology for agriculture have a decisive in-
sificationgenerally toward less heavily taxed fluence on the level and pattern of agricultural
commodities such as cassava. Like Malaysia, Thai- growth and on private investment in the sector. A
land was strikingly successful in reducing poverty. study of fifty-eight countries during 1969-78 found
A World Bank study of six African countries, that a 1 percent increase in irrigation coverage was
"Managing Agricultural Development in Africa," associated with a rise in aggregate crop output of
offers further contrasts between good and bad per- 1.6 percent and that a 1 percent increase in paved
formers. The countries with the most successful roads was associated with a rise in output of 0.3
agricultural sectors (especially Cameroon and percent. Investments in infrastructure also help to
Kenya) were also the best performers in GDP per improve and maintain natural resources. Cost-
capita and manufacturing output. These countries effectiveness, however, is important. An analysis
taxed farming less severely than the others, partly of completed Bank-supported projects to create in-
by avoiding acutely overvalued exchange rates. frastructure in agriculture found that economic
They also invested in an institutional and infras- rates of return averaged 17 percentwell above
tructural environment that supported diversified the 10 percent used to qualify a project as success-
agricultural growth, much of it in the small-farm ful. (In Africa, however, performance is much
58
Box 4.1 Agricultural policies in Brazil favor large farmers
Brazil has promoted agriculture and has achieved rapid income tax. Brazil tried this but failed because of wide-
growth in the sector. But its record on poverty is disap- spread evasion and many exceptions. One such excep-
pointing. One reason is that biases within the agricul- tion, intended to encourage land use, reduces the tax
I tural sector, notably in taxation and subsidized credit, by up to 90 percent if owners use the land to graze
favor large farmers and work to the disadvantage of cattle. This promotes the conversion of forestland to
labor. These biases interact with the highly inequitable uneconomic livestock ranching, reduces the demand
initial land distribution to reduce the gains to the poor for labor, and has harmful environmental conse-
from growth. Brazil vividly demonstrates that promot- quences.
I ing agriculture is not enough. Policies within the sector
also matter. Subsidized credit
Land and agricultural taxation Agricultural credit has been exceptionally distorted in
I Brazil. Until recently, real interest rates on official credit
Brazil's income tax greatly favors agriculture in relation were negative, and real interest rates on loans for agri-
to other sectors, but only the rich can reap the benefits. culture were lower than in the nonagricultural sector.
Through special provisions in the income tax code, cor- The difference in credit terms between sectors has been
porations can exclude up to 80 percent of agricultural capitalized in the price of land. Although subsidies
profits from their taxable income, and individuals can raise profits in agriculture, they have mainly benefited
exclude 90 percent. Fixed investments can be fully de- large farmers and have encouraged excessive mechani-
preciated in the first year and can even be depreciated zation, again reducing the demand for unskilled labor.
two to six times over. This, together with high infla- Poor people who lack land titles have not benefited
tion, encourages corporations and the rich to overin- from credit subsidies.
vest in land. The result has been the accumulation of Agricultural policies in Brazil have reduced labor de-
large landholdings and increases in land prices that mand and have made it almost impossible for a poor
exceed growth in land productivity. The poor do not person to buy land and become a farmer. Opportuni-
benefit from the tax breaks (they pay no income tax), ties for unskilled workers to acquire skills by becoming
and they cannot afford to buy land. Many move to long-term workers have been substantially reduced by
frontier areas in search of unclaimed land. subsidized mechanization.
A progressive land tax could offset the bias in the
lower because of weak institutions and poor can sometimes affect certain subgroups adversely
project implementation.) unkss planners anticipate and prevent such po-
Better infrastructure can lead to increased pro- tential effects. For instance, thirty-nine dam
ductivity, technical change, and strengthened mar- projects approved for financing by the World Bank
ket linkages. Irrigation and water control have in twenty-seven countries during 1979-85 brought
been crucial for higher yields (through the adop- considerable benefits to people in the command
tion of modern varieties and multiple cropping) areas but also entailed relocation of about 750,000
and reduced variability of output in the past inhabitants of reservoir areas. Such displacement
twenty years in South and East Asia. And al- can cause profound distress, disruption of social
though landowners are likely to enjoy the biggest and productive structures, increased poverty, and
absolute gains, the income gains from infrastruc- environmental damage. The issue has not always
ture can be widely dispersed. In Andhra Pradesh, been adequately addressed by governments and
India, for example, the monthly per capita expen- aid agencies. During the 1980s, however, the treat-
diture for small farm units and wage-earning ment of populations displaced by infrastructure
households in 1983 was 35 percent higher in irri- projects improved considerably. Resettlement
gated than in nonirrigated districts. A study of plans are becoming integral components of the
Bangladeshi villages found that infrastructural de- projects, which also include funds for acquiring
velopment is associated with a rise in the mcomes land and providing infrastructure and services for
of all households, including the poor and landless the resettled areas.
(Box 4.2). Much of the increase was a result of Technological change is vital for agricultural
changes in the way labor is used. growth. The record of the past thirty years clearly
Infrastructural projects benefit many, but they supports the argument for public funding for agri-
59
There is also evidence of a link between technol-
ogy and wages. Detailed country studies often find
Box 4.2 Infrastructural development that new technologies give an initial boost to the
and rural incomes in Bangladesh demand for agricultural labor because total labor
use is greater over the course of the year. Rural
A study of sixteen villages in Bangladesh shows wages increased in many states in India and Paki-
how the development of infrastructureroads,
stan between the mid-1960s and the mid-1970s and
electric power, banks, markets, schools, and
health centersaffects the incomes of rural house- in Java, Indonesia, between the mid-1970s and the
holds. The study divided the villages into those early 1980s. The demand for farm labor may have
that had and had not benefited from the provision slowed in India from the late 1970s owing to labor-
of public infrastructure. With other factors con- saving innovations, sometimes encouraged by
trolled, the study found that greater infrastruc- subsidies to mechanization. But slow growth in
tural development was associated with a one- farm employment has been offset by the buoyancy
third increase in average household incomes.
Crop income increased by 24 percent, wage in-
of the nonfarm sector. There has been some con-
come by 92 percent, and income from livestock cern that modern technologies and mechanization
and fisheries by 78 percent. These three changes may have reduced employment opportunities and
largely benefited the poor. Income from nonfarm incomes for women, but the evidence does not, in
businesses increased by 17 percent; this largely general, support this view (Box 4.3).
benefited the nonpoor.
Roads, electricity, and other economic services FARM-NONFARM LINKAGES. Growth in the rural
encouraged the production of new farm products
(including perishable commodities) and higher
nonfarm economy is important in creating rural
output in transport, construction, services, and employment and in raising labor incomes. Small,
small-scale industries. All this had a substantial labor-intensive enterprises are the most common.
effect on the pattern of labor demand. Although Nonf arm activities typically account for 20 to 30
households worked roughly as many days a year percent of rural employment in Asia and Latin
in developed as in undeveloped villages, in the America and for 10 to 20 percent in Africa. If rural
developed ones they spent less time on family towns are included, the proportions rise substan-
labor, which had low implicit returns, and much tially. A study in India found that employment in
more on wage labor, especially in the relatively
high-paying nonfarm sector. Poor households the nonfarm sector grew 35 percent in the 1970s,
with few physical assets, including landless compared with 14 percent for employment in agri-
households, gained substantially. culture; without that growth, rural wages would
have risen much more slowly. In Kenya among
smallholder families per capita income from non-
farm sources climbed 14 percent a year between
cultural research and for dissemination of new 1974-75 and 1981-82; incomes from farm employ-
technologies in small-scale production. Many in ment rose only 3 percent a year.
the 1970s thought that the Green Revolution Although the nonfarm sector has often ex-
would do little or nothing to help the poor, but panded faster than the farm sector, agriculture is
recent appraisals suggest that most of these fears usually the key to the growth of nortfarm activities.
were unfounded. In Asia and Latin America (the An expanding farm economy demands inputs
main areas in which the new varieties were from and supplies raw materials to transport, pro-
adopted) small farmers and laborers alike have cessing, and marketing firms. Rising farm incomes
benefitedalthough most studies have found that lead to greater demand for consumer goods and
small farmers adopt new varieties with a lag of services. A study found that spending on locally
three years or so because they are less willing or produced nonfoods accounted for 33 percent of the
able to take risks. Better access to new varieties as increase in household expenditures in rural areas
well as to water, chemical inputs, and credit has of Malaysia and India and for 15 percent of in-
encouraged small farmers to switch more quickly. creased spending in Sierra Leone and Nigeria.
Extension services designed for small farmers have More generally, the study found that every dollar
also helped. Where policies and support were in- increase in agricultural income led to an income
adequate, as in much of Sub-Saharan Africa, few increase of about eighty cents elsewhere in the
small farmers have adopted the improved technol- economy in the Asian cases and fifty cents in the
ogies, and both overall growth and the incomes of African cases. This difference reflects the lower
the poor have suffered. population densities and poorly integrated mar-
60
Box 4.3 The impact of technological change on women
Modern seed varieties, irrigation, and the increased Other studies for India and Nepal have found that the
commercialization of crops have commonly been ac- overall use of hired female labor rose substantially with
companied by the greater use of hired labor, mostly the introduction of modern varieties.
from landless households. The new technologies have In some cases, however, mechanization has led to
also had important implications for the division of lower female employment. The outcome has often de-
household labor. Wage labor has replaced unpaid la- pended on the tasks mechanized. When predomi-
bor, and in some cases male labor has replaced female nantly female tasks were given over to machinery,
labor. This has raised concerns that technological women were displaced. This happened in Bangladesh,
change has harmed women. Indonesia, and the Philippines with the replacement of
The substitution of hired labor for family labor usu- the finger knife as a harvesting tool and the introduc-
ally improves the household's standard of living. In tion of direct seeding and portable mechanical thresh-
the Philippines, for example, the new technologies ers. In Bangladesh most of the postharvest work had
raised farming incomes, allowing households to hire been done by women using the dheki (a foot-operated
labor and purchase labor-saving farm implements. This mortar and pestle). When the dehusking and polishing
reduced the number of hours worked by family mem- of grain were mechanized, these operations were
bers in low-productivity jobs on the farm and allowed turned over to men, who now operate the modern
them to engage in other, more productive, activities mills. A study in the Indian states of Kerala, Tamil
such as trade or raising livestock. In addition, greater Nadu, and West Bengal found that where chemical fer-
demand for hired labor provided jobs to landless tilizers have replaced cow dung, men rather than
workers. women now apply the fertilizer because women lack
Modern varieties have, in general, raised the de- access to the information provided by extension ser-
mand for hired female labor. They usually require more vices.
labor per acreparticularly in tasks typically done by When women were displaced, the effect on their in-
women, such as weeding, harvesting, and postharvest comes and on household welfare depended on
work. A study of three Indian states concluded that the whether they found more productive jobs elsewhere.
use of hired female labor was greater on farms that had Overall nonfarm employment did increase, but data
adopted modern varieties than on those that had not. classified by gender are scanty.
kets in Africa. Public investments in rural infra- on policy toward the labor market itself. Often,
structure (in electricity, transport, water, banks, industrial protection reduces both the level and
telephones, schools, and so on) can greatly growth of labor use in the formal sector. As a rule,
strengthen these linkages. the greater the degree of protection, the greater the
capital intensity of production; this is illustrated
The pattern of urban and industrial growth with data for India in Table 4.1. A more neutral
trade regime would therefore increase the demand
Growth in urban employment and wages is the for labor. Some of this expansion in demand would
second broad determinant of the pace of poverty probably come from increased exports, but the
reductionthrough its direct influence on the ex- main reason for reducing protection is to use
isting urban poor and through the opportunities it resourcesincluding labormore efficiently.
creates for migration from rural areas (Box 4.4). Greater neutrality in the trade regime can support
Growth of urban employment is especially impor- a more labor-intensive pattern of industrial expan-
tant in middle-income developing countries, but it sion in import-competing as well as exporting sec-
matters increasingly everywhere. Governments tors. A study of ten countries in the 1970s confirms
can affect the urban demand for labor by altering this view. In addition, it found that in Indonesia,
the incentives and regulations that face workers Pakistan, and Tunisia labor demand would have
and their employers and by providing, or failing to increased more from better use of resources within
provide, adequate urban infrastructure. the import-competing sector than from a shift be-
tween the two sectors.
INCENTIVES AND REGULATION. The demand for The contribution of manufacturing to employ-
urban labor depends partly on government policy ment over time differs greatly from country to
toward the markets for goods and capital as well as country (Table 4.2). The growth of manufacturing
61
Box 4.4 Does rural-to-urban migration help or hurt the poor?
Most studies have found that people migrate mainly probability that an educated person will leave the vil- I
for economic reasons. Poverty, both absolute and rela- lage is five times greater than for an uneducated per-
tive, and income variability, which leads to greater vu!- son; in Tanzania 90 percent of the men who left their
nerabiity, cause people to move. In Botswana house- villages had some schooling.
holds "place" different members of the household in The poor who do not migrate may still benefit indi-
different labor markets so that, for example, the effects rectly. Rural-urban migration has tightened the rural
of a drought on rural incomes will be offset by remit- labor market in many countries. In addition, remit-
tances from members working in urban areas. In rural tances help those who stay behind. In Kenya remit-
India households often marry their daughters into dis- tances helped to lift some of the poorest households
tant and dispersed (although still kinship-related) out of the lowest income class and contributed to agri-
households. In almost all cases rural-urban migrants cultural innovation, partly by reducing income variabil-
increase their incomes. Most migrants have jobs wait- ity. There is little evidence to suggest that migration
ing for them or find one within a month or two. In the worsens poverty in the sending areas.
Ludhiana District of Punjab, India, 78 percent of mi- The impact of migration on the urban poor is also of
grants had a job after one month and 94 percent after interest, but information is sparse. Migration, it might
two months. A study of the poorest of the urban poor, seem, will hold down urban wage rates for unskilled
the Calcutta pavement dwellers, found that migrants work in the informal sector. Evidence from Colombia
were better off after moving. In Colombia rural-urban showed that migrants had higher incomes than urban
migrants had higher incomes than comparable people natives of similar education. But there is no evidence
in their rural places of origin; the gains were even that migration causes the incomes of natives to fall.
higher for the better educated. Some countries have tried to reduce migration on the
The poorest are likely to be underrepresented in grounds that it leads to excessive and costly urbaniza-
rural-urban migration. The study of Punjab's Ludhiana tion. With few exceptions, these efforts have failed to
District showed that only 15 percent of rural outmi- stem rural-urban migration significantly or to redirect
grants belonged to the lower classes, although they migrants to secondary towns. An urban bias in policy
accounted for more than 24 percent of the sample. In a could lead to more migration than is socially desirable.
study of forty North Indian villages, only 5 percent of The appropriate policy is to focus on direct remedies
I
the working migrants came from farm-labor house- such as reducing the biases against agriculture in pric-
holds, although such households made up 19 percent ing and improving the social and economic infrastruc-
of the villages' population. The poorest may stay at ture of rural areas.
home because they are less educated. In Kenya the
Table 4.1 Effective protection and labor try opted for an entirely neutral trade regime, but
intensity of manufacturing, India, 1986 incentives were far less distorted in Korea, Malay-
Degree of Share of sia, and (after the mid-1970s) Thailand than in Ar-
effective fixed Share of Fixed capital gentina, Pakistan, or Tanzania.
protection capital employment per worker
of industry (percent) (percent) (thousands of rupees)
FACTOR MARKET INTERVENTIONS. When govern-
High 53 19 93
Medium 4 3 32 ments intervene in the markets for capital and la-
Low 43 78 18 bor, they often exacerbate the antilabor bias of pro-
All industries 100 100 32 tection. Many countries make imports of capital
Source: World Bank 1989b.
goods cheap (through low tariffs and overvalued
exchange rates), offer tax breaks for investment in
capital equipment, and subsidize creditall of
employment depends on both the growth and the which tend to reduce the price of capital. Subsi-
labor intensity of the sector. The middle- and low- dized energy prices often exacerbate this bias and,
income countries of East Asia again stand out. furthermore, have adverse environmental conse-
Many of the countries that saw manufacturing em- quences. In contrast, social security taxes, labor
ployment rise quickly also had thriving farm sec- regulations, and high wages (especially in indus-
tors. Policies that avoid discriminating against ag- tries in which competition among producers is
riculture go hand in hand with a more broadly weak) all tend to raise the cost of labor in the for-
labor-intensive pattern of development. No coun- mal sector. A study of incentive structures in ten
62
Table 4.2 The contribution of manufacturing to The informal sector is very diverse in its income
employment, selected countries, 1970 to 1980 structure and activities. It is dominated by one-
(percent)
person firms and small-scale entrepreneurs that
Manufacturing employment as employ a few apprentices (often relatives) and
share of total labor force Increment, hired laborers. Firms are not covered by govern-
Country 1970 1980 1970-80'
ment labor regulations, and there are no restric-
Korea, Rep. of 7 14 36 tions on entry. But many in the informal sector pay
Malaysia 9 16 33
Indonesia 6 9 23
indirect taxes and feeslicense fees for small re-
Brazil 7 10 19 pair shops and street vendors, for example. Wages
Thailand 5 7 13 are generally lower than in the formal sector, espe-
Colombia 5 6 10 cially for apprentices. But there is also a high de-
India 2 3 5 gree of inequality, and many entrepreneurs do
Zambia 3 3 4
Kenya 2 2 4
much better than workers in the formal sector. Ac-
Tanzania 1 1 2 tivities range from efficient manufacturing, trans-
Pakistan 2 2 1 port, and trading enterprises to marginal work
Argentina 17 13 21 such as collecting and recycling trash. Most of the
Note: The figures are based on registered employment in manufac- sector's production is for consumptionespecially
turing. by low-income householdsand little is exported.
a. The ratio of the increase in manufacturing employment to the
increase in the labor force. As in the rural nonfarm sector, growth in the
Source: United Nations Industrial Development Organization and urban informal sector depends on the rest of the
World Bank data.
economy and, in particular, on the demand for
nontraded goods and services. But preferential
countries found that government intervention treatment for large firms has undermined the in-
raised the relative price of labor in all cases in the formal sector. Subsidized capital for one part of the
early 1970s. The increase was 11 percent in Korea, economy implies fewer resources for the rest.
between 30 and 50 percent in Argentina, Brazil, Some countries have tried to offset these biases.
and Côte d'Ivoire, almost 90 percent in Tunisia, India, for example, has provided cheap credit for
and more than 300 percent in Pakistan. small enterprises and has restricted competition
Labor-market policiesminimum wages, job se- from larger firms. This kind of approach rarely
curity regulations, and social securityare usually works. Protection and constraints on the entry of
intended to raise welfare or reduce exploitation. large firms into labor-intensive activities have
But they actually work to raise the cost of labor in tended to reduce any gains in employment, espe-
the formal sector and reduce labor demand. Stud- cially since credit subsidies have led many small
ies from the 1970s and 1980s found that job secu- firms to adopt relatively capital-intensive tech-
rity regulations reduced the long-term demand for niques. Distortions in product markets, particu-
labor by an estimated 18 percent in India and 25 larly biases in the structure of protection among
percent in Zimbabwe. There is little poverty, in industries, have probably been a more important
any case, in the formal sector. Yet by trying to im- influence on the demand for labor. Indeed, evi-
prove the welfare of workers there, governments dence suggests that small firms in the manufactur-
reduced formal sector employment, increased the ing sector are not necessarily more labor-intensive
supply of labor to the rural and urban informal than large firms in the same line of business. In
sectors, and thus depressed labor incomes where Korea labor intensity within industries seldom var-
most of the poor are found. ies by a factor of more than three, whereas across
industries it varies by a factor of more than a
THE ROLE OF THE INFORMAL SECTOR. In most de- hundred.
veloping countries the informal sector plays a Most countries have further undermined infor-
prominent role in providing employment and in- mal employment through heavy regulation. In In-
comes. It has been estimated to account for 75 per- donesia, where most of the informal sector is thriv-
cent of urban employment in many countries in ing, restrictions on pedicabs have steadily
Sub-Saharan Africa and for 85 percent in Pakistan. increased, and pedicabs are now completely
It also accounts for most of the poverty in urban banned from Jakarta. In Zimbabwe street vendors
areas. In Côte d'Ivoire, for example, the incidence and small-scale enterprises are excluded from
of urban poverty in the informal sector is more many parts of the towns, and small businesses
than three times that in the formal sector. have to struggle to acquire land titles. A study of
63
Peru found that a prospective entrepreneur who Vietnam. The old agrarian order was swept away
wanted to set up a small garment factory had to entirely and no compensation was paid. Else-
spend 289 working days dealing with regulations; where, reform helped existing tenants,
in Lima it took more than two years to register a smallholders, or new settlers by transferring indi-
minibus route. In addition to removing biases that vidual rights; Egypt, India, Iran, Japan, Kenya,
favor larger firms, light regulation together with and the Republic of Korea are examples. In some
the provision of appropriate urban infrastructure is cases compensation was paid, but it was usually
the best way to facilitate growth in the informal based on depressed land prices.
sector. In China, Japan, and the Republic of Korea, land
reforms were extensive. They affected the poor in
URBAN INFRASTRUCTURE AND POVERTY. Urbaniza- two ways. First, rural households benefited from
tion will make increasing demands on infrastruc- the reallocation of rents and from new opportuni-
ture. Investments in infrastructure can make in- ties to use the family's resources. Second, together
roads on poverty both by improving living with favorable incentives for agriculture and
conditions and by promoting employment. Indeed strong local institutions, they provided the basis
the two go together. In many cases roads, water, for broadly based poverty-reducing growth.
sanitation, electric power, and other services for In the absence of a major upheaval, land reforms
low-income areas help small businesses as well as have rarely gone so far, and as a result their impact
households because many informal businesses are on the poor has been modest. In Egypt, India, and
based in the home. Studies on Colombia and Ko- Iran the reforms made owners out of former ten-
rea show that access to public utilities is essential ants but favored the better-off farmers. Redistribu-
for small new firms. In Nigeria, as in many other tions in Bolivia and Mexico attacked the ineffi-
African countries, weak urban infrastructure in- ciency and inequity of the "hacienda" system, but
hibits the growth of small firms that cannot afford, since subsequent policies failed to sustain growth
for example, their own power generators or water of productivity in smallholder agriculture, the ef-
facilities. fect on the poor was muted. In Bolivia, moreover,
most of the land was given to large agricultural
Increasing the participation of the poor companies for livestock raising and forestry. And
in growth
although more than half a million rural people
Growth that creates opportunities for the poor will gained access to four million hectares of land,
have a greater impact on the poor if they have many still lack land titles. This makes transactions
access to land, credit, and public infrastructure difficult and limits the use of land as collateral.
and services. Many countries have adopted pro- In most circumstances, political realities forbid
grams to this end. reform to stray far from the status quo. Where ex-
propriation is not possible, the cost of compensat-
Increasing access to land ing the former owners becomes a major difficulty,
especially when policy distortions have driven up
Policies to redistribute land have deservedly re- the price of land. Land reform can be expensive in
ceived much attention. In addition, policies that other ways, too: there is the cost of mapping and
expand tenancy, provide clear land rights where registering the new owners' holdings and of pro-
traditional systems fail, and improve the manage- viding infrastructure and services to raise produc-
ment of common-property resources can create op- tivity on the new farms.
portunities for many of the rural poor. Such poli- The chances of successful land reform are greater
cies should help reduce poverty and make the land if aid is available, as in the case of Kenya. A project
more productive at the same time. in northeast Brazil also suggests that more modest
land transfers to small farmers are feasible despite
REFORMING PROPERTY RIGHTS. In the twentieth all the difficulties and that, when supported with
century virtually all the major redistributions of adequate investment, they help to raise farm in-
property rights in land were precipitated by social comes (Box 4.5).
revolution, defeat in war, or national liberation
from colonial rule. Many of these upheavals led to IMPROVING TENANCY. Many governments have
large-scale collective forms of organization, as was tried to make tenancy more secure and thus, in
the case in China, Cuba, Ethiopia, the Democratic effect, to transfer ownership rights. Some of these
People's Republic of Korea, the U.S.S.R., and reforms have been successfulin the Philippines,
64
BALANCING INDIVIDUAL AND COMMON PROPERTY.
Africa's traditional forms of land ownership seem
Box 4.5 The merits of selective land to be evolving toward individual property rights.
reform: Northeast Brazil This is mainly because of population growth and
The Piaui Rural Development Project in Brazil
the increasing commercialization of agriculture.
the first World Bank-supported project to focus But this shift toward individual land rights tends
on land tenure reformshows that land transfers to undermine the ability of traditional systems to
to small farmers are feasible and that security of ensure that all members of the extended family
tenure encourages poor farmers to adopt new have access to land. This feature of their land sys-
p technologies. It demonstrates what can be tems has helped some countries in Africa to avoid
achieved in a country with a highly unequal distri- the extremes of poverty and landlessness that are
bution of land and agricultural policies strongly
common in much of Asia and Latin America: tradi-
biased against the poor (as discussed in Box 4.1).
Land in Piaui was cheapit was originally $40 tional systems have provided secure land tenure
per hectare and fell to $15 to $25 per hectare as the and encouraged farmers to invest in their land
project went forward. Over five years the project (Box 4.6). In such cases, encouraging individual
bought 200,000 hectares, distributed land to 3,480 land registration and titling may be undesirable.
families, and regularized land tenure for 1,500 Where traditional systems have failed to provide
families. It also provided physical infrastructure clear land rights, land titles and registration are
mainly roads, water supply, and some irrigation. useful.
The income of 7,760 landholders increased by be-
tween 240 and 293 percent. Overall, the total culti- Common pasture and forest resources are im-
vated area increased by 16 percent and the value portant for poor rural households. They provide
of production by 50 percent. Yields increased by fuelwood, fodder, and employment to those who
10 to 40 percent for rainfed farms and by 30 to 70 otherwise have few land rights. When the popula-
percent for irrigated farms. Clear title had given tion is growing rapidly, or when there is open ac-
the farmers an incentive to invest in the land and cess to these land resources and traditional com-
adopt better technologies. munity management has broken down, the
commons are often overused and degraded. This
hurts those who depend most on the common re-
sources as a safety net. Policies to privatize these
for example, and in West Bengal and Kerala, India, resources have often failed to protect the interests
where tenants' rights became inheritable. But ten- of those with existing rights to the land or to create
ancy reform has usually failed. Landlords have the basis for a viable and equitable system. Small
thwarted the new legislation by evicting tenants or farmers have sometimes been forced, as a result, to
concealing tenancy as wage labor. In Brazil, where sell their individual holdings. Common-property
tenancy laws limit owners' share of output and resources should receive greater attention. They
make it hard to evict tenants and sharecroppers, need to be better protected and better managed.
semiskilled workers have difficulty in gaining ac- Investment in research on how to use these fragile
cess to land. Landowners prefer informal and very resources would be money well spent.
short-term tenancy agreements. Or they may evict
the tenants and use the land for forestry, livestock, Increasing access to credit
or mechanized production or even leave it idle. In
India and Pakistan the area cultivated by pure ten- Credit can help the poor to accumulate assets and
ants has shrunk dramatically over the past thirty to cushion their consumption in hard times. But
years. extending credit to the poor is costly to the lender.
Tenancy law is evidently not the best way to ex- Transaction costs are high, and the risk may be
pand tenancy or to protect tenants' rights and in- great owing to lack of collateral. Moreover, other
comes. Tenants' bargaining power can probably be borrowers, such as large-scale farmers, may have
strengthened most effectively through policies that preferential access for cultural or ethnic reasons.
raise the demand for labor and thus boost un- Many governments have therefore tried to expand
skified wagesthe return to the best alternative for credit for the poor through large-scale subsidized
the potential tenant. Such policies are better suited credit programs and other measures. Several ra-
for improving the chances for the landless to be- tionales have been given for these policies: the
come tenants and enabling their households to in- poor cannot afford market interest rates; formal
crease the returns to their farming skifis. lenders are too cautious; informal lenders are too
65
Box 4.6 Land tenure systems in Sub-Saharan Africa: the case of Rwanda
Some African land tenure systems retain from earlier after independence. The rest obtained most of their
forms of communal ownership features that give the land in the form of transfers (gifts and loans) from
poor guaranteed access to at least a small piece of land within their extended families. These transfers typi-
and at the same time motivate farmers to conserve the cally provide secure rights to the land and hence give
long-term productivity of their holdings. Rwanda of- farmers the incentive to use it efficiently and conserve
fers a good example. It has one of the highest popula- its productivity. This is essential, given Rwanda's
tion densities in Africa (445 inhabitants per square kilo- mountainous terrain and the potential for erosion.
meter of arable land in 1986), and employment The ability of the extended family to exercise some
opportunities outside farming are few (90 percent of control over land transfers is a key feature of the Rwan-
the work force is in agriculture). Yet landlessness is dan land tenure system. Households that borrow land
almost nonexistent. For example, in Ruhengeri Prefec- rarely claim any permanent transfer rights over it, even
ture, a densely farmed area, about 98 percent of rural though they may have the right to farm it on a long-
households cultivate at least 0.1 hectares, and 75 per- term basis. Even land that is inherited or received as a
cent work at least 0.3 hectares. The distribution of land permanent gift cannot always be freely transferred to
is also relatively equitable. This contrasts sharply with others by the recipient. But the lack of full land transfer
similarly populated areas in Asia, where 30 percent or rights in this case does not appear to have affected
more of the rural households may be landless. farmers' investment behavior in improving and con-
Survey data from a World Bank research study in serving land or the level of productivity that is
three prefectures (Butare, Gitarama, and Ruhengeri) achieved. What is crucial seems to be the right to be-
show that 22 percent of rural households did not in- queath land to family members. Without this right, lev-
herit any land. About half the households in this group els of investment and use of modern inputs are signifi-
obtained some land from state allocations at or shortly cantly lower.
exploitative. But this approach turned out to be two-thirds of total credit; in the other village insti-
misguided. tutional credit was a smaller part of the total, and
only 17 percent of borrowers were in arrears. And
SUBSIDIZING CREDIT FOR THE POOR. Despite, or repayment rates were generally lower for large
because of, many years of trying to channel greater farmers than for small ones. So it is not poverty, as
amounts of formal credit to the poor, only 5 per- such, that makes borrowers unreliable.
cent of farms in Africa and 15 percent in Asia and Moreover, experience shows that the poor are
Latin America have had access to it. In Bangla- willing to pay market interest rates. Bangladesh
desh, after more than a decade of subsidies, only found that small farmers continued to demand
15 percent of smallholders and 7 percent of the loans even when the interest rate was about 30
landless households had received institutional percent a year. The poor, in any case, borrow rou-
credit. Cheap credit has become a transfer pro- tinely on the informal market, where rates are fre-
gram for the nonpoor. In Brazil these implicit quently very high. Moneylenders in rural Sri
transfers, at their peak in the early 1980s, were Lanka commonly charge rates of 25 to 50 percent
estimated to be as high as $3 billion to $4 billion a for a growing season.
year (between 1.2 and 1.6 percent of GDP). In ad- Where subsidized credit has reached the poor,
dition, artificially low interest rates and credit reg- other questions arise. By 1988, India's Integrated
ulations distort the allocation of resources and lend Rural Development Program (IRDP) covered 27
themselves to patronage and corruption. They million rural families. A sample study in Uttar Pra-
have damaged the financial sector and have failed desh found that almost 60 percent of investments
to expand credit to the poor. had been retained for four to five years. But in
Studies of formal subsidized credit programs in 1986, 59 percent of IRDP loans were in arrears. The
Sub-Saharan Africa, the Middle East, and Latin study in Uttar Pradesh showed, moreover, that
America have found that loans in arrears range only 7 percent of households that had repaid their
from 30 to 95 percent. Subsidized borrowers are loans were receiving additional credit. Thus, al-
less reliable than unsubsidized ones. A detailed though the IRDP has succeeded in increasing the
study of three Indian villages found that about 60 asset holdings of large numbers of disadvantaged
percent of borrowers were in arrears in the two households, it has not established sustainable fi-
villages in which institutional credit accounted for nancial services for the poor.
66
Box 4.7 Learning from informal finance in Sub-Saharan Africa
Informal finance takes many different forms: saving State Insurance Corporation began a susu-like program
and loan associations, rotating funds, mobile bankers in February 1987. Money Back, as it is called, provides
and moneylenders, financial dealings among family life insurance and investment services primarily to
and friends, and so on. Despite their flexibility, these small- and medium-scale businesses. As with the susu,
arrangements are usually strict and well run. clients deposit an agreed contribution, and staff mem-
Rotating associations, known in some African coun- bers visit markets daily to collect. Money Back works in
tries as susus or tontines, are flexible and creative. Peo- a way that people find familiar, it provides clients with
ple pay an agreed sum into a fund, out of which loans security for their savings, and it attracts funds that the
are then made to members on a rotating basis. In formal banking system, on the whole, would riot. The
Ghana susus have evolved into growing credit and Money Back program is still in its early stages, but it is
saving facilities. Daily contributions range between 10 increasingly popular. it may grow into a regionwide or
and 500 cedis ($0.04.$2.00). Collectors visit markets nationwide program.
daily to accept deposits, no matter how small, mainly In other African countries informal arrangements
from market women. In rural areas collectors make have grown into relatively large financial organiza-
their rounds early in the morning before farmers go to tions. In Cameroon, Côte d'Ivoire, Guinea, Mali, and
the field and again in the evening after they come back. Senegal savings "clubs' 'which include, for example,
Monthly deposits reach millions of cedis. A study in rotating funds and credit unionshave been estab-
Tamale, in Ghana's Northern Region, found that lished. These arrangements are based on personal loy-
monthly deposits by an average collector range from alties, but they also function effectively as financial in-
approximately 1.5 million cedis to 2.0 million cedis. A termediaries that give mutual loan guarantees. The
typical Northern rural bank might hold savings of 10 Cameroon Cooperative Credit Union League provides
million cedis. The informal system evidently mobilizes services to 231 credit unions with about 62,000 mem-
a significant volume of savings. bers and savings of about $33 million.
Seeing the advantages of this approach, Ghana's
Thus, cheap credit programs have not helped to adopt a variety of innovative credit programs
the poor. Public funds are better spent on irifra- targeted to the poor. These vary enormously in
structure and services, such as agricultural exten- coverage (Table 4.3), but successful programs have
sion and market information. A viable, undis- certain features in common. Most have chosen not
torted financial sector will help to make the best
use of this investment.
Table 4.3 Diversity in the coverage of credit
DEVELOPING FINANCIAL INSTITUTIONS FOR THE programs
POOR. Subsidized formal finance has failed. But is Program Year Coverage'
there scope for expanding informal finance, the MicroFund (Philippines) 1989 730
most common source of credit for the poor? Irilor- Production Credit for Rural
mal lenders, operating on low fixed costs, offer Women (Nepal) 1989 6,640
low-income clients small loans on the basis of per- Association for Development of
Microenterprise (ADEMI)
sonal or business acquaintance. Because informal (Dominican Rep.) 1988 19,430
lenders know a lot about their clients, they can be Small-scale Enterprise Program
flexible about collateral and repayment schedules. (Calcutta) 1988 36,000
There are lessons here for formal lenders (Box 4.7). Working Women's Forum
(Madras) 1988 50,000b
Despite its popularity and potential, informal fi-
Small Farmer Development
nance has drawbacks. Its separation from larger Program (Nepal) 1989 78,520
financial markets limits the lenders' access to Saving Development Foundation
funds and reduces competition, and it rarely pro- (Zimbabwe) 1985 250000b
vides term finance or large loans. Some of these Grameen Bank (Bangladesh) 1988 413,000
difficulties may be overcome through links be- Kupedes (Indonesia) 1988 1,300,000'
Badan Kredit Kecamatan (BKK)
tween informal and formal finance, but much re- 2,700,000
(Indonesia) 1982
mains to be learned about these linkages.
a. Cumulative membership.
The limitations of both formal and informal fi- b. Approximate figure.
nance have led governments, donors, and NGOs c. Number of beneficiaries in 1988.
67
Table 4.4 Grameen Bank loans, by purpose Group lending may not always be appropriate or
of loan and sex of borrower, 1985 necessary. Badan Kredit Kecamatan (BKK), an In-
(percentage of current loan amount)
donesian public program, provides individual
Male Female All loans without collateral primarily to low-income
Purpose of loan borrowers borrowers borrowers
women, without relying on groups. It contains its
Crop cultivation 4.0 4.6 4.3 processing costs by making tiny initial loans ($5 is
Livestock, poultry the limit) on the basis of character references from
raising, and
fisheries 18.5 44.6 31.9 local officials and by using one-page loan applica-
Processing and tions that are processed in less than a week. The
manufacturing 18.6 29.9 24.4 program's local units are autonomous. To reach
Trading and the poor, they disburse their loans quickly from
shopkeeping 49.7 18.7 33.8 accessible village outposts. Since the program's
Transport and other
services 9.2 2.2 5.6
loans are small and loan terms have been strict, the
nonpoor tend to look elsewhere for credit. Finally,
Source: Hossain 1988.
the program encourages borrowers to repay by
making repayment a condition for approving new
to subsidize interest rates. Instead, they have tried loans, and it has strong incentives for loan officers
to adapt themselves to the needs of the poor by to expand the client base and maintain high collec-
reducing transaction costs to both lender and bor- tion rates. The BKK has managed both to reach the
rower. Programs with high borrowing rates and poor and to remain financially viable. It serves
strict terms, especially those that link repayment to more than 35 percent of Java's 8,500 villages, and
future lending, stand a better chance of reaching in 1987 it earned profits of $1.4 milliona 14 per-
the poor. cent return on the loan portfolio.
Group lending is one approach for reaching poor Some credit programs have targeted microen-
people. Typically, under such schemes one mem- terprises, often in urban areas, with packages of
ber's failure to repay jeopardizes the group's ac- credit, training, and technical assistance. The pro-
cess to future credit. Joint liability among a group grams that have emphasized credit have been the
of borrowers reduces the risk of default and makes most successful. Schemes such as MicroFund in
it cheaper to reach dispersed clients. The best- Manila, Philippines, and ADEMI in Santo
known example of this approach is the Grameen Domingo, Dominican Republic, have targeted the
Bank in Bangladesh. It has successfully served ex- poorespecially poor women, many of whom
tremely poor people-83 percent of them women work in very small enterprises. Microenterprise
and its loan recovery rates exceed 95 percent. Table lending can have a considerable impact on in-
4.4 shows how the poor use Grameen Bank credit. comes. The average income of new borrowers
The Working Women's Forum in Madras, India, from the Small-Scale Enterprise Credit Program in
has also achieved recovery rates of between 90 and Calcutta rose by 82 percent, and that of borrowers
95 percent. In both cases loan recovery far exceeds from the Kupedes program in Indonesia increased
the national averages for commercial banks. Small, from $74 to $183 after an average of three years.
self-selected groupssuch as those in the Gra- Most microenterprise credit programs receive sub-
meen Bank or in Nepal's Small Farmer Develop- sidies to help cover their initial costs. Experience
ment Program and Production Credit for Rural has shown that the more successful programs,
Women Programusually offer the best base for such as Kupedes, can become financially viable if
such schemes. they charge market-based interest rates and keep
The Zimbabwe Agricultural Finance Corporation operating costs low.
demonstrates that group lending can decrease ad- Experience also highlights the importance of sav-
ministrative costs. Because it lends only to estab- ings. The Savings Development Foundation in
lished groups, its costs are a minuscule 1 percent Zimbabwe generated considerable savings by or-
of loan capital. If lenders must incur the costs of ganizing households into neighborhood groups
setting up the groups, however, the overall costs of and devising a simple system of financial record-
group lending can exceed the costs of lending to keeping that illiterate people could understand. By
individuals. At the Grameen Bank expenses as a 1985, 250,000 members had saved enough to place
share of outstanding loans are 16 to 25 percent for bulk orders for fertilizer and seeds, improve their
new branches, dropping to 6 percent after three housing, and meet other basic needs. Rural
years. women, who account for 97 percent of the pro-
68
gram's participants, acquired new skills while be- ment du Coton du Cameroon, adapted the existing
coming financially more independent. research on cotton to the local environment and
Although successes are none too common, they substantially improved the productivity of
suggest that well-designed programs can give dis- smallholders. Farmers should be more directly in-
advantaged groups access to credit and still remain volved in the selection of research topics. Adaptive
financially viable. Institutions of this kind should research needs to be broadened to include, for ex-
be supported with limited subsidies to help cover ample, soil and livestock management techniques
their initial administrative costs (but not interest in addition to the traditional stress on crop hus-
rates) and to encourage innovation; with time, as bandry. To make all this feasible, countries will
costs decline, the subsidies should be withdrawn. also need well-trained extension workers to act as
Programs such as BKK, Kupedes, the Grameen a bridge between small farmers and researchers.
Bank, and ADEMI have demonstrated the poten- PRODERO in Honduras has improved the pro-
tial for raising the productivity and incomes of the ductivity of small farmers by involving them in
poor. Because of their example, the number of sim- planning and testing new agricultural technolo-
ilar schemes, although still small, has grown rap- gies. The project targeted farmers who owned less
idly in recent years. than five hectares in the isolated western states of
Copan, Lempira, and Ocotepeque. To combat soil
Improving access to infrastructure and technology erosion, it developed low-cost methods of conser-
vation and fertilization. Average maize yields in-
Public investment in technology and infrastructure creased by more than 300 percent.
is critical in raising incomes and reducing poverty. Farming systems research is an approach that
The extent to which the poor benefit, however, listens to farmers and pursues a broad research
depends on the design and effective implementa- agenda. It looks at the entire farming system
tion of the programs. cropping patterns, livestock and irrigation man-
agement, the division of labor among household
ADAPTING TECHNOLOGIES FOR SMALL FARMERS. members, and so on. The results in Zambia, Zim-
Technological improvements in agriculture have babwe, and the Indian state of Bihar (Box 4.8) have
helped small farmers in some regions more than in been encouraging.
others. The Green Revolution benefited many
smallholders in Latin America and Asia, but in PROVIDING RURAL INFRASTRUCTURE. Poor farmers
Sub-Saharan Africa relatively few small farmers typically have less access to public infrastructure
use improved high-yielding varieties. In Malawi, than more prosperous farmers. Rural roads and
after twenty years of agricultural research and ex- electricity, for example, are first extended to the
tension, only 5 percent of farmers have adopted relatively well-off. The supply of water to marginal
hybrid maize. This slow acceptance reflects severe farmers is especially erratic and inadequate. In In-
biases against peasant agriculture. Many countries dia the low quality of canal construction and main-
have encouraged and subsidized large-scale mech- tenance means that surface irrigation reaches only
anized commercial farming, which is beyond the part of the intended area. Small farmers cannot
reach of most smallholders. In addition, supplies afford deepwells, tubewells, or pumpsets.
of inputs are unreliable, and storage facilities are Successful infrastructure projects have usually
lacking. More important, indigenous farm- delegated to the local level as much responsibility
research institutions are weak. As a result, there for administration and maintenance as possible.
has been too little emphasis on developing and This has required some effort in building local in-
disseminating varieties and techniques that are ap- stitutions, involving local people, and designing
propriate for small-scale rainfed farms. infrastructure with these needs in mind. World De-
In addition to removing biases against small velopment Report 1988 showed that delegating pub-
farmers, countries need an indigenous capacity to lic responsibility to local organizations can make
do adaptive agricultural research. International re- infrastructure projects more effective and benefit
search centers initiate much of the scientific work the poor.
necessary for technological change, but countries The National Irrigation Administration (NIA) in
need to be able to identify relevant technologies the Philippines has worked with user groups to
and adapt them to the specific requirements of dif- improve maintenance and distribution. It employs
ferent farm locations. In Cameroon, for example, a community organizers to help form groups of wa-
public sector agency, the Société de Développe- ter users and then negotiates with each group over
69
Box 4.8 Developing appropriate technologies for subsistence farmers in Bihar
The Rural Women's Agricultural Development Project, left fallow. Farmers have also been taught to diversify
which is sponsored by the Birsa Agricultural University to higher-value crops. Farmers, extension workers,
in Bihar, demonstrates how farming systems research and researchers analyzed the problems together. They
can help to improve the incomes of the poor. The then developed a package that combined new and in-
project targets poor tribal farmers in the drought-prone digenous technologies, tested the technologies both
Chota Nagpur Plateau, where illiteracy rates reach 90 on-station and on-farm to adapt them to varying local
percent and 85 percent of the farmers possess less than conditions, and disseminated the package to participat-
five acres of land. Women perform at least half of the ing villages.
preharvest and four-fifths of the postharvest work. The gains are illustrated by the experience of one
They depend mainly on rice-based rainfed farming for farmer. In 1982 she had planted 2.5 acres of rice and
their livelihood. Inappropriate farming practices and millet in the rainy season, 0.5 acres of cauliflower in the
insufficient use of inputs contribute to low winter, and no crops in the summer. In 1987 she
productivity. planted her 2.5 acres with rice and potatoes during the
The local culture makes it difficult for male extension rainy season and with wheat, cauliflower, peas, and
workers to advise female farmers or to be advised by carrots in winter. In the summer she planted 1.25 acres
them. Women with tribal backgrounds have therefore of cabbage and okra.
been hired as extension workers. They work through Diversification and irrigation raised cropping intensi-
mahila mandals (women's groups). ties and incomes. Cropping intensity rose, on average,
Since farmers refused to adopt costly new technolo- from 95 percent in 1981-82 to 145 percent in 1986-87.
gies, researchers worked with them to develop a sim- Gross income per household increased more than four
ple dugwell system that allows vegetables to be grown times in real terms.
during the dry season, when land had previously been
the quantity of water delivered. User groups allo- systems that define clear lines of responsibility for
cate water and costs among themselves and share agency managers and performance criteria that are
construction and maintenance responsibilities. Be- linked to budget allocations and create incentives
tween 1981 and 1984 this allowed a 38 percent drop for greater local responsibility. In the case of the
in the NIA's per hectare spending on operation NIA, farmer groups were required to repay con-
and maintenance. The efficiency of water use and struction costs, which increased their ownership of
distribution improved, and the access of small the program and the accountability of managers.
farmers to irrigation increased. The NIA, which In turn, each regional office was expected to
had started as a pilot project in 1976, covered more achieve financial viability, and managers' perfor-
than 35,000 hectares by 1986. The Gal Oya irriga- mance was evaluated by their actual recovery of
tion project in Sri Lanka was modeled on the Phil- costs from farmers.
ippine program. There too the introduction of user Another illustration of the importance of local
groups almost doubled the efficiency of water use, institutions is the Aga Khan Rural Support Pro-
and the improved irrigation system now reaches gram in northern Pakistan. This NGO-managed
the poorer farmers downstream. project supports commercialization in subsistence
Both schemes were based on pilot projects, and villages. It has created village organizations, built
both embody a flexible design. The designers ex- productive physical infrastructure, developed fi-
perimented extensively with the composition and nancial services, and provided support systems
responsibilities of user groups and with ways to and training for production and marketing. The
integrate these groups with the national adminis- local organizations carry out many tasks and have
tration. Their success has led such countries as Ma- avoided the need for coordination with ministries.
laysia and Thailand to introduce similar systems. The program has recorded household participation
Local organizations are more likely to succeed if rates of up to 97 percent in Gilgit District and has
they are legally recognized by the central govern- achieved average rates of return of 33 percent for
ment and if the government provides support in irrigation schemes and other projects.
the form of guidelines, training, and information More targeted approaches are especially impor-
systems. It is also crucial to establish management tant in remote regions where poor people are iso-
70
lated from markets and services. In order to reach cal groups. Large programs that start as pilot
the most disadvantagedtribal groups, the land- projects seem to have a better chance of success. A
less or near landless, and, in some societies, pilot project may slow the scheme down and use
womenit is essential to work with organizations up valuable management time, but the benefits
such as local groups, NGOs, and private opera- usually outweigh these costs.
tors, that know their needs. The International
Fund for Agricultural Development (IFAD) has car- Reaching resource-poor areas
ried out many projects that target the poorest.
IFAD uses pilot programs and flexible designs and Increasing numbers of poor people live in areas
usually relies on forming groups to reach the target that have little agroclimatic potential and are envi-
population. A rural development project in Cam- ronmentally fragile. Examples include the Loess
eroon, designed to improve coffee production by Plateau in China, the highlands of Bolivia and Ne-
farmers who own less than two hectares, is one pal, the desertic African Sahel, and much of the
example. A similar approach has been used by humid tropics. Population pressure in these areas
NGOs in Bangladesh, where landless groups oper- has decreased the productivity of the land and in-
ate irrigation equipment and sell water to farmers creased its vulnerability to flooding and soil ero-
(Box 4.9). sion. This raises the question of the links between
The lessons of recent experience in rural infra- poverty and environmental degradation.
structure and technology programs are broadly as These regions need a special development strat-
follows. Local institutions can mobilize resources egy, for three reasons. First, their potential for
such as savings and labor. They can help to ensure growth is limited. Second, they are increasingly
that project benefits reach the poor, that specific occupied by poor people with the fewest skills and
local needs are met, and that the projects remain the least access to infrastructure and supplies.
financially viable. Successful programs have not Third, environmental degradation in these regions
relied exclusively on government agencies, which adversely affects both the immediate area and re-
can provide effective central support but often lack gions downstream or downhill.
both the field staff and the flexibility to work at the The causes of these growing pressures on natu-
local level. Instead, they have employed a mixture ral resources are complex and interconnected. In
of institutionsNGOs, private operators, and lo- many countries poor farmers are being margina-
Box 4.9 Groundwater irrigation in South Asia: reaching the landless
I
p In the Gangetic Plain groundwater is one of the few repayment rates achieved by Proshika (75 percent) and
remaining natural resources that can be exploited to the Grameen Bank (more than 75 percent) are much
reduce rural poverty. Recent efforts to help the landless higher than is usual for agricultural loans.
poor benefit directly from groundwater irrigation have The biggest drawback is the demands this approach
I been based on groups. Proshika and the Grameen makes on management. Most groups require costly
Bank have organized new groups and have called on managerial and technical assistance, and they often
existing ones to buy and operate irrigation equipment need help in mediation with water users. Problems
and sell water to farmers. such as technical deficiencies in pump operation and
Five points stand out in the groups' performance to maintenance, difficulties in obtaining satisfactory
date. (1) In certain technical respects (for example, the pump locations, and struggles over contracts with wa-
area irrigated with equipment of a given capacity), the ter users have led the Grameen Bank to assume direct
I groups are at least as effective as private management. management of some groups.
(2) Some groups have successfully promoted the use of After more than a decade Proshika has reached only
high-yielding varieties by their customers. (3) The about 300 groups covering 6,000 irrigated acres. It is
groups invest some of their higher incomes in other unlikely that such groups can become a significant
activities, such as fisheries, poultry raising, or cottage force in groundwater management. But the groups
industries. (4) They help farmers to spread the risks have managed to reach the poorest of the poor, who
associated with irrigation and to reduce the costs of are usually excluded from rural development initia-
organizing farmer cooperatives, settling conflicts, and tives. The reasons for their success are just beginning
so on, which can be high for minor irrigation. (5) The to be understood and merit further study.
71
Box 4.10 Protecting the environment and reducing poverty: China's experience in the
Loess Plateau
In China's Loess Plateau, programs combining erosion and reclaimed flatland and by converting unterraced
control with improved crop- and animal-raising prac- hills to fodder and tree-crop production. Specific mea-
tices have reduced chronic soil erosion and increased sures include constructing new terraces to reduce ero-
rural incomes. This success follows several earlier fail- sion and increase crop yields, introducing improved
ures. It suggests that economic and sustainable agricul- pasture species, eliminating uncontrolled grazing on
ture in such areas requires a mixture of technical, so- unterraced hillsides, planting trees and shrubs on the
cial, and policy actions. It also confirms the importance steepest hillsides, and constructing soil dams to create
of research in pilot stations prior to broad implementa- cultivable flatlands by impounding bess sediment.
tion. These measures require that much of the steeply
The Loess Plateau extends over about 630,000 square sloped land that is now being cultivated be taken out of
kilometers of northwestern China. Rainfall is barely crop production. Higher yields on existing and newly
sufficient for agriculture and is often concentrated in created flatland and terraces are therefore essential for
sporadic, heavy falls in the summer. Agricultural ex- raising production and farm incomes.
pansion and exploitation have progressively removed The initial results have been favorable. The per capita
the plateau's vegetative cover, contributing to erosion gross value of output has doubled, output has been
and degradation of the soil on most of the plateau. diversified, and soil erosion has been reduced. Despite
Streams carry silt to the Yellow River, creating prob- a decrease in the area sown to grain, greater yields on
lems for downstream irrigation works and causing improved terraces and intensively cultivated flatland
dangerous increases in the height of the riverbed. have helped to increase total per capita grain produc-
The situation worsened during the 1960s and 1970s. tion by more than 30 percent. Expanded pastureland
As part of the national policy of self-sufficiency in supports more sheep, and the volume of animal prod-
grain, farmers on the plateau were encouraged to ucts has increased. Cost-benefit analysis indicates that
switch from pasture-based livestock and limited grain the financial and economic rates of return range from
production to extensive grain cultivation. This exacer- 13 percent for construction of soil dams to 25 percent
bated land degradation and led to declines in agricul- for the terracing of gently sloped land for crop produc-
tural productivity and income. By the mid-1980s more tion.
than five million people on the plateau's rainfed up- The government is now encouraging adoption of the
lands were surviving on incomes of less than $50 a erosion control program throughout the Loess Plateau
year. by limiting cultivation on steep, unterraced land and
In the late 1970s the government initiated research by providing credits for constructing terraces and
and development to improve erosion control and agri- planting pasture species. Participating households re-
cultural production in the area. Erosion control experi- ceive full rights to land use and rights to the output of
ment stations were established in Mizhi County trees and fodder from hillsides they have converted.
(Shaanxi Province) and Dingxi Prefecture (Cansu Prov- The agricultural development component of the Gansu
ince). The strategy is designed to reverse environmen- project, which is assisted by the World Bank, is sup-
tal degradation by replacing the existing extensive porting implementation of the strategy over about
cropping system with intensive cropping on terraces 200,000 hectares in the Guanchuan Basin in Dingxi.
lized and pushed to frontier areas. In addition, still worse. For example, land policies have di-
population growth and the commercialization of rected population movement away from the most
agriculture have forced farmers who once relied on productive land by giving a few large estates pref-
environmentally sustainable forms of cultivation to erential access (as in Malawi) or by limiting migra-
use their land more intensively. That might be de- tion (as in Tanzania). Insecure land tenure and en-
sirable under certain systems of land and livestock croachment on common and state lands encourage
management, but the intensification of traditional soil-mining practices that diminish the longer-term
farming methods, such as slash-and-burn agricul- productivity of the land. Policies that promote in-
ture, has damaged the productivity of these mar- dustrial extraction from these areas (such as pulp
ginal areas. Overgrazing, unmanaged irrigation, subsidies to Indian rayon mills) can add to the
and an ever widening search for fuelwood all ac- pressures on the resource base.
celerate the decline. Providing infrastructure to develop these
Policies that discriminate against smallholders in resource-poor regions may be neither cost-
granting access to land and forests make matters effective nor viable. A better strategy would start
72
with investment in education and training to spur ternational research institutes have recognized the
outmigration to areas of better potential. Spending urgency of the problem and have increased their
to meet basic needs such as health care and drink- funding for research in marginal areas. The Inter-
ing water will also be required. Outmigration national Rice Research Institute in the Philippines,
would be most effective in countries where labor for example, now devotes 12 percent of its budget
demand is growing strongly in other regions. to upland rice cultivation alone. In many countries
Experience, however, shows that migration is government subsidies to develop and improve
only a partial solution. Growth in the areas of low-return farming activities may be the only way
greater potential is usually not high enough; many to reduce poverty in these regions.
resource-poor regions have rising populations de-
spite outmigration. Therefore, additional invest- Policies for poverty-reducing growth
ments wifi be necessary. These wifi involve train-
ing poor farmers in better techniques for farming, Achieving a pattern of development that success-
animal husbandry, and soil and moisture conser- fully reduces poverty requires policies that provide
vation, increasing the opportunities for diversifica- opportunities to the poor and enable them to par-
tion and off-farm employment, and providing local ticipate in growth. To accomplish this, policies
user groups (such as the village forestry associa- must be attuned to three broad tasks.
tions in Korea and the rangeland management as- First and foremost, economywide and sectoral
sociations in Botswana) with rights to manage de- policies must encourage rural development and
graded communal lands. In addition, policies to urban employment. Experience indicates that this
make land tenure more secure in areas in which requires moderate taxation of agriculture and rela-
traditional tenure systems have broken down wifi tively undistorted product and factor markets. It
discourage farmers from mining the soil for short- also requires public provision of infrastructure and
term gains. an environment that makes technical change ac-
In many of these areas farmers can adopt low- cessible to small farmers and the urban poor.
cost, low-input technologies that would increase Second, specific policies are needed to improve
and stabilize yields, diversify production, and the participation of the poor in growth by increas-
maintain the resource base. One such technology, ing their access to land, credit, and public infra-
contour cultivation, has raised yields sub- structure and services. Land transfers can reduce
stantiallyby between 6 and 66 percent on slopes poverty, but they succeed only in special circum-
of up to 32 percent. When contour cultivation is stances. Other policies to increase and secure ac-
supported by a vegetative barriervetiver grass, cess to land can also reduce poverty. Subsidized
for examplethe benefits are even greater. Vetiver credit programs have failed to reach the poor, but
grass has been used for many years in the Carib- approaches such as group lending offer a promis-
bean, Kenya, and South Asia. It holds soil while ing alternative. Flexible programs that involve the
allowing for water filtration, and it is cheap to es- intended beneficiaries, build institutions, employ
tablish, manage, and maintain. But it is not appro- NGOs and local groups, and respond to local
priate everywhere; it is ineffective in parts of the needs are the best way of molding infrastructure,
Sahel where soil and moisture are insufficient. services, and technology to the needs of the poor.
In some regions more substantial interventions Third, resource-poor regions, where poverty
will be required. Programs that target only a few and environmental degradation are interrelated,
households or villages cannot prevent soil erosion require a different approach. Since the potential
or protect watersheds. Box 4.10 describes how im- for growth in these regions is limited and the pop-
proved technology for terraced lands and more ap- ulation is increasing, policies that facilitate outmi-
propriate land tenure policies have decreased land gration are essential. But in many of these regions
degradation and poverty in China's Loess Plateau. additional investments, which are likely to require
Reaching the poor in these regions will be diffi- government subsidies, will stifi be necessary to
cult. Without effective measures, the situation is meet basic needs, maintain or increase yields, and
likely to grow worse for the foreseeable future. In- preserve natural resources.
73
Delivering social services
to the poor
Efforts to reduce poverty are unlikely to succeed in owes as much to safe drinking water, improved
the long run unless there is greater investment in sewage disposal, and other sanitation measures as
the human capital of the poor. Improvements in to broad immunization coverage and mass cam-
education, health, and nutrition directly address paigns against parasitic diseases. It has much to do
the worst consequences of being poor. But there is with the provision of basic health care and afford-
ample evidence that investing in human capital, able drugs to even the most remote parts of the
especially in education, also attacks some of the country. It reflects the successful drive to reduce
most important causes of poverty. Social services fertility and to increase, through legislation, the
are therefore an essential part of any long-term age of first delivery, as well as great efforts to pro-
strategy for reducing poverty. vide education on health and nutrition. And it
would probably have been impossible without a
Social sector policies safety net that, among other things, guaranteed
In the developing world as a whole, the past three
minimum food rations to even the poorest rural
people.
decades have seen tremendous progress in social
China's performance is exceptional. But it
indicators. But the advances have been uneven
among regions and among socioeconomic groups
teaches an important general lesson: large im-
provements in the health of the population can be
within countries. Progress in the social sectors re-
quires a long-term commitment to making educa-
achieved if there is a broad and lasting political
commitment, with a consistent emphasis on pre-
tion, health care, and other social services accessi-
ventive measures and basic curative care. In other
ble to all levels of society. Where that commitment
words, social progress is not merely a by-product
is lacking and where government policies have
of economic development. Policies matter. Where
failed to reach the poor, social needs remain stag-
progress has been better than average, strong in-
geringly high.
terventions based on a long-term commitment to
Progress in health social progress lay behind it.
Costa Rica provides another example. Its mortal-
In the past few decades the developing countries ity rate for children under 5 fell from 112 per thou-
have made great strides in health. In China over sand in 1960-65 to 24 in 1980-85. Since the 1960s
the past twenty-five to thirty years, for instance, the proportion of underweight children under 6
the total fertility rate fell from 6.4 to 2.4 children has dropped from 14 to 2 percent. Life expectancy
per woman, and the infant mortality rate dropped is comparable to that in the industrial world. In the
from 90 to 32 per thousand live births; life expec- 1970s the country embarked on several far-
tancy increased from 52.7 to 69.5 years. Here, as in reaching health and nutrition initiatives, including
other countries, no single factor accounts for the the first National Health Plan (1971), the Universal
improvement. China's remarkable performance Social Security Law (1971), and rural and commu-
74
nity health programs (1973 and 1976). A family developing world. Just fifteen years later, the Ex-
planning program was introduced in 1968; by us- panded Programme on Immunization (EPI) had
ing the available rural health infrastructure, it has extended its polio and DPT coverage to about 60
succeeded in reaching people in the countryside as percent (as measured by the number of children
well as the urban poor. The total marital fertility receiving a third dose of vaccine before age 1).
rate declined from 7.6 in 1960 to 3.4 in 1980. These Many countriesfor example, Egypt, Mexico,
achievements did not come cheaply; by 1985 Costa Peru, Thailand, and Zairehave made good
Rica was spending 23 percent of its government progress in training workers to use ORT. Much
budget on health. Such high spending was sus- remains to be done, but after less than a decade
tainable only because the country devoted just 3.2 one-third of all families in the developing world
percent of its budget to military outlays between now have access to this treatment.
1975 and 1985.
Other countries, too, have been able to raise Progress in education
health standards to well above the level that might
be expected on the basis of their per capita GNP. In As in health, the developing countries have made
Cuba, for instance, life expectancy is high, and Sri great advances in education over the past two de-
Lanka reduced its under 5 mortality rate from 101 cades. Between 1965 and 1985 the number of pri-
per thousand in 1960 to 35 in 1985. In all these mary schools in low-income countries (excluding
cases the government has taken the leading role. China) increased by almost 60 percent, to more
Government spending on health usually rises than 775,000. In middle-income countries the
with economic development. Through training number more than quadrupled, to nearly 950,000.
medical personnel, investing in clinics and hospi- (During the same period the number of school-age
tals, licensing practitioners, testing drugs, and children grew 69 percent in low-income countries
running subsidy and insurance schemes, in addi- and 48 percent in middle-income countries.) Be-
tion to directly providing medical care, the public tween 1965 and 1985 the number of teachers went
sector is the dominant player in health. Govern- up 55 percent in low-income countries, 175 percent
ment efforts have helped bring about an increase in lower-middle-income countries, and 120 percent
in the number of doctors, nurses, and hospital in upper-middle-income countries.
beds per thousand population throughout the de- Most of the schools in developing countries are
veloping world. Such summary health measures public schools that are built, financed, and staffed
as life expectancy and under 5 mortality have im- by the government. The share of education spend-
proved dramatically (Figure 5.1). It is widely be- ing allocated to capital expendituresessentially,
lieved that many of the economic problems in the school constructionreflects the big push of the
developing world (as well as in the industrial 1960s to improve access to schooling. In 1965 low-
world) can be attributed to excessive or inappro- and middle-income countries spent 21 and 14 per-
priate government intervention. Yet much of the cent, respectively, of their total education budgets
social progress observed during the past two de- on building schools; twenty years later the share
cades is clearly a direct result of government had fallen to 12 percent in both cases. Most teacher
action. training programs, both for primary and for sec-
Another factor behind the improvement in ondary school teachers, are run by governments.
world health has been the effort of the interna- As in health care, spending on education rises
tional health community to focus on affordable, with the level of development, and so do stan-
low-technology child survival techniques such as dards of attainment (Figure 5.2).
immunization and the use of oral rehydration ther- As a result of this commitment to education, and
apy (ORT) for diarrheal dehydration. In this field it especially primary education, net enrollment rates
is no exaggeration to talk of a public health revolu- grew, on average, 5.7 percent a year between 1965
tion. This progress has been possible because of a and 1975 in low-income countries, 5 percent in
unique coalition of bilateral development agencies lower-middle-income countries, and 2.3 percent in
and United Nations agencies, notably the World upper-middle-income countries. During the late
Health Organization (WHO) and the United Na- 1970s enrollment continued to grow faster than
tions Children's Fund (UNICEF). When in 1974 population. By 1985 there was enough primary
the World Health Assembly announced a new im- school capacity in most middle-income countries
munization drive, immunization services were to enroll virtually all children; in the low-income
reaching less than 5 percent of the children in the countries capacity still fell short of that goal.
75
Figure 5.1 Trends in health expenditure, number of physicians, and health outcome
GDP and public expenditure on health, Physicians per thousand population
dollars per capita, selected countries, 1985
Expenditure
80 1.0
S
0.8
60
40
S
20
0.2
..
1)
GDI' 0 500 1,000 1,500 2,000 2,500 3,000 3,500 1960 1965 1970 1175 1980 1985
Under 5 mortality rate Life expectancy at birth
Deaths per thousand Years
300 70
250
200
150
100
50
0 30
1960-65 1965-70 1970-75 1975-80 1980-85 1961) 1965 1970 1975 1980 1985
Sub-Saharan Africa East Asia ,.. South Asia
Middle East and North Africa Latin America and the Caribbean
Note: For life expectancy, data for East Asia 1960 include the average life expectancy for China for 1963 - 67.
Sources: Hill and Pebley 1988 and World Bank data.
Continuing needs Governments deserve credit for the achieve-
ments to date. But their justifiably heavy outlays
Despite great progress in both health and educa- on health and education would do more good
tion, an immense task remains. About 30 million and would go a long way toward meeting the
children under 5 die every year from causes that problems that remainif they were better spent.
would not usually be fatal in developed countries. Government spending on social services is ineffi-
About 110 million children worldwide (almost 20 cient in many ways. Above all, it tends to be
percent of the relevant age group) receive no pri- skewed away from the people who need it most
mary education. More than 90 percent of them live the poor.
in low- and lower-middle-income countries. On average, an estimated 70 to 85 percent of the
76
developing world's total health spending, both vented altogether or treated earlier at less cost. The
public and private, goes for curative care. Between heavy financial demands of the curative sector also
10 and 20 percent is spent on preventive care and lead to internal inefficiencieshospitals without
the remaining 5 to 10 percent on community ser- equipment and clinics without drugsbecause sal-
vices such as mosquito control and health educa- aries crowd out other spending.
tion. Within the curative sector, hospitals often ac- The result is that the poor often lack access to
count for more than 80 percent of the cost. Yet it is even the most basic services. The proclaimed goal
well known that preventive and community ser- of free health care for all often means low-quality
vices are far more effective in reducing morbidity care in the cities and none at all for the rural poor.
and mortality. If public resources tied up in hospi- Even in China in 1981 government spending per
tals were redirected to the lower levels of the capita on health care was more than three times
health care system, many diseases could be pre- higher for urban dwellers than for rural people.
Figure 5.2 Trends in education expenditure, student-teacher ratios, net enrollment rates, and completion rates
GDP and public expenditure on education, Primary school student-teacher ratios
dollars per capita, selected countries, 1985
Expenditure
160 45
40
30
25
0 20
GDP 0 500 1,000 1,500 2,000 2,500 3,000 3,500 1965 1970 1975 1980 1985
Primary school net enrollment rates Primary school completion rates*
Percentage of age group
100
80
60
40 40
20 20
0 0 Does not include South Asia,
1965 1970 1975 1980 1985 1970 1975 1980 1985
Sub-Saharan Africa East Asia --_ South Asia
Middle East and North Africa Latin America and the Caribbean
77
Seventy percent of Senegal's physicians and phar- In short, the overall expansion of the health care
macists, 60 percent of its midwives, and 40 percent system has shortchanged the most important ser-
of its nurses are concentrated in the Dakar-Cap vices: preventive care and basic curative care. The
Vert region, where less than 30 percent of the pop- poor have not benefited proportionately from the
ulation lives. In Peru two-thirds of all doctors live larger numbers of doctors, nurses, and health care
in the capital, serving just 27 percent of the pop- facilities, from the increased availability of afford-
ulation; in most rural areas, where the majority of able drugs, or from public health services such as
the poor live, there is only one doctor for every water and sanitation. As a result, the link between
10,000 or more inhabitants. poverty and illness remains strong both for chil-
Box 5.1 Adult health: a neglected issue of growing importance
Adult health is a relatively new issue in public health Policies that succeed in reducing poverty in develop-
policy in developing countries. Most public health ef- ing countries will usually improve adult health too.
forts in developing countries have emphasized child Improvements in drinking water, sanitation, housing,
and infant health. Yet half the population in the devel- and nutrition check the spread of tropical diseases.
oping world is between ages 15 and 60. Children de- Less crowding and better housing prevent the trans-
pend on these adults, and so does the economy. More mission of tuberculosis and other viral respiratory dis-
than anybody else, the poor depend on good health to eases. Specific preventive measures are needed for a
maintain the productivity of their only assetlabor. range of other illnesses: vaccination for tetanus, educa-
The loss of the family breadwinner to death or debili- tion and taxation for diseases linked to alcohol and
tating disease may force a whole household into pov- cigarettes, and, especially in Africa, sex information
erty. campaigns for AIDS.
Each year in developing countries ten million young
adults die. Respiratory diseases, malaria, diarrhea, car-
diovascular diseases, cancers, chronic obstructive lung
diseases, sexually transmitted diseases (including ac-
Box figure 5.1 Adult mortality in the developing world
quired immune deficiency syndromeAIDS), diabe-
tes, tuberculosis, and injuries are widespread. In 0 Female fl Male
Guinea-Bissau and Sierra Leone, for example, the
chance at age 15 of surviving to 60 is less than 50 per-
cent; in most developing countries it is between 50 and 35.9
75 percent (as against 85 percent in developed coun-
tries). Diseases such as onchocerciasis, epilepsy, and
leprosywhich raise morbidity rather than mortality
are not captured in those figures. 26.5
As Box figure 5.1 indicates, adult mortality is highest
in low-income countries. Life expectancy at 15 in the
developing world is seven years less than in industrial 9,7
countries. Fertility and mortality rates are falling in
most developing countries, but the adult population is
still growing. This demographic transition is happen-
ing alongside a health transition: chronic diseases are
becoming more important.
In Colombia high rates of parasitic infections in
adults are associated with less education and lower in-
come. In Peru and Côte d'Ivoire poor rural households
lose half as many more working days because of illness
as households in poor urban areas. In Bahrain disabil-
ity and low economic status go together; according to a
1981 census people with disabilities were three-and-a-
half times more likely to be illiterate, twelve times less Low income Lower middle Upper middle
income income
likely to have a college degree, and three-and-a-half
times less likely to be employed. In Kenya bad health
among adults is strongly linked to low economic sta- Note: Adult mortality is the probability at age 15 of dying
tus, poor household sanitation, and inadequate nutri- before age 60, expressed as a percentage.
tion.
78
dren (as stressed in Chapter 2) and for adults (Box books by the early 1980s, in the Dominican Repub-
5.1). lic fewer than 20 percent of eighth-grade students
In education, too, government policy has fa- in public schools had mathematics textbooks, and
vored higher-level training over services that in Botswana fewer than 20 percent of primary
would benefit the poor. It is well known that ter- school students had access to science or social
tiary education yields the lowest social rate of re- studies textbooks. A 1983 study in the Philippines
turn, but many countries nonetheless spend a dis- reported that only 32 percent of fifth-grade science
proportionate share of their education budgets on teachers used textbooks frequently; another in Bo-
that level. tswana showed that teachers used textbooks only
In West Africa, in addition to free tuition and 12 percent of the time. Teachers' guides are seldom
regular government subsidies to the colleges, stu- available.
dents in higher education often receive living al- So it is hardly surprising that repetition rates are
lowances. It is estimated that these living allow- high and that many of those who complete pri-
ances account for nearly half of the total spent on mary school have learned very little. In various
higher education. Only 2 percent of the relevant tests conducted in about forty countries students
age group in Sub-Saharan Africa goes on to higher in low- and lower-middle-income countries an-
education, but that level accounts for 22 percent of swered only 40 percent of the questions correctly.
the region's public education budget. Brazil In particular, reading comprehension appeared to
spends 9 percent of its public education budget on be weak.
secondary education and 23 percent on higher ed- To make matters worse, there is a recent trend,
ucation. In Chile, Costa Rica, the Dominican Re- mainly in low-income countries, toward lower pri-
public, and Uruguay people in the top one-fifth of mary enrollment rates. In the face of tight budget-
the income distribution receive more than 50 per- ary constraints, many African countries are failing
cent of the subsidies for higher education; the to expand their education systems to keep up with
poorest one-fifth receives less than 10 percent.
population growth. (Ghana, Liberia, Mali, and
In India education is heavily biased in favor of Tanzania are but a few examples.) The sharpest
urban dwellers and males. Urban literacy rates are declines in enrollment are observed in low-income
twice as high as rural rates, and females in both countries that have suffered from war or internal
rural and urban India have lower literacy rates strife. Afghanistan's enrollment rate fell 40 percent
than males. In Pakistan 63 percent of the boys in between 1980 and 1985; during the same period
the relevant age group are enrolled in elementary
Somalia's also fell 40 percent. In Ethiopia and Mo-
school, but only 32 percent of the girls are. In rural zambique, after considerable progress in earlier
areas enrollment rates for girls are as low as 20 years, growth in enrollment has ceased.
percent.
But the problems in education go well beyond Extra spending on social services in general will
enrollment rates. Low enrollment figures are often not automatically help the poor. The existing pat-
accompanied by high dropout rates. In low- tern of provision needs to be tilted in their favor in
income countries more than 40 percent of those terms of both the quantity and the quality of ser-
who enter primary school fail to finish, and even in vices. The most important measures in the social
upper-middle-income countries completion rates sectors for improving the living conditions of the
are only 85 percent. As a result, illiteracy remains poor are also the most basic: expand and improve
widespread in the developing world. primary education and primary health care.
The low quality of education goes a long way
toward explaining the weak performance of chil- Investing in people
dren in developing countries. Governments have
stressed quantity over quality. Many countries There is overwhelming evidence that human capi-
spend less each year on textbooks and other mate- tal is one of the keys to reducing poverty. More-
rials; in low-income countries teaching materials over, improvements in health, education, and nu-
account for less than 3 percent of recurrent expen- trition reinforce each other. But the poor generally
ditures. lack access to basic social services. There is too
Until the 1980s textbooks were often not pro- little investment in their human capital, and this
vided to students at all. In the Central African Re- increases the probability that they and their chil-
public the national student-to-textbook ratio was dren will remain poor. To break this vicious circle,
between ten and twenty to one. In Brazil only 23 governments must make reaching the poor a prior-
percent of all schools had received first-grade text- ity in its own right.
79
Box 5.2 Education and economic growth
Many studies of farm productivity, family enterprises, benefiting fully from their greater skills. The results
and wage earners have demonstrated the effects of ed- suggest that there may be thresholds in the returns to
ucation on output and productivity. As Box table 5.2 education. For example, about four years of education
shows, the returns to education are substantial. Al- seem to be needed to attain functional literacy. Invest-
though the impact of education on aggregate real out- ments may yield substantial returns only when they
put has been less well documented, a recent study of are big enough to push the economy over such a
the determinants of real GDP covering fifty-eight coun- threshold.
Box table 5.2 Average social returns to education
(percent)
Box figure 5.2 The effect on real GDP of one additional
Primary Secondary Higher year of education per person
Region education education education
Sub-Saharan Africa 26 17 13
Asia 27 15 13 Percentage change in real GDP
Latin America 4
and the Caribbean 26 18 16
Source: Psacharopoulos 1985. East Asia
tries during 1960-85 strongly suggests that education
can contribute significantly to aggregate output. An in- Latin America and
crease of one year in average years of education may the Caribbean
lead to a 3 percent rise in GDP.
The results of this research differ from region to re- Middle East and North Africa
gion (Box figure 5.2). For most groups of developing
countries the effect of education on CDP is clearly posi- South Asia
tive, and (within the range examined) the higher the
initial level of education, the greater the benefits from
increasing it. This underscores the importance of in-
vesting in education. Sub-Saharan Africa is an excep- Sub-Saharan Africa
tion; there, increasing average years of education by
one year, from 3.25 (in 1985) to 4.25, is expected to have
a negligible effect on output. This reflects many factors: 3 4 5 6
unfavorable local conditions, the lack of complemen-
Average years of education
tary inputs, inadequate institutional capability, and
other economic obstacles that prevent people from
Education and productivity technologies, and virtually all studies on agricul-
tural productivity show that better-educated
The principal asset of the poor is labor time. Edu- farmers get a higher return on their land. One
cation increases the productivity of this asset. The study on Africa found that farmers who have com-
result at the individual level, as many studies pleted four years of educationthe minimum for
show, is higher income. More recent research also achieving literacyproduce, on average, about 8
points to a strong link between education and eco- percent more than farmers who have not gone to
nomic growth (Box 5.2). school. Numeracy and literacy were identified as
In the wage sector the individual returns to edu- the essential skifis. Studies in Korea, Malaysia,
cation are consistently above returns to conven- and Thailand confirm that schooling substantially
tional investments. It is true that most people in raises farm productivity.
the developing world do not work for wages and A growing number of the poor in the developing
that many of the poor are self-employed in agricul- world find employment in the nonagricultural
ture or in small family enterprises. But this does nonwage sectormostly in the informal economy.
not weaken the case for investing in education. Not all these workers are poor, but many of the
Educated farmers are more likely to adopt new self-employed do belong to the poorest segments
80
of society, and employment in the informal sector the poor are engaged. Among sugarcane workers
makes up the main part of their family income. in Guatemala productivity increased with better
Studies on how education influences informal nutrition. The productivity of workers in Indone-
sector productivity are scarce, but the few that do sia who received iron supplements for two months
exist point to benefits. More education gives the rose by between 15 and 25 percent. Increased calo-
worker a wider range of self-employment options rie intake in Sierra Leone substantially raised
and allows him or her to choose more profitable farmers' productivity. A study for India shows a
alternatives (for example, modern trade rather significant link between wages and weight-for-
than traditional handicrafts). Within most subsec- height (a measure of short-term nutritional status)
tors earnings per hour rise with years of education, among casual agricultural laborers. Another study
just as in the wage sector. The precise effects vary found that the effect was especially marked in the
by type of activity and sometimes differ between peak agricultural season, when more energy is re-
men and women. In Peru returns to an extra year quired for harvesting. Estimates for Sri Lanka
of primary education are estimated to be as high as show a significant positive effect of energy intake
33 percent for women self-employed in the retail on real wages.
textile sector. Postprimary education appears to
have a relatively high payoff-14 percentfor men Nutrition and learning
in the service sector. A study of small and
medium-size enterprises in Colombia showed that Just as the education of parents has a positive ef-
the entrepreneur's backgroundskills, education, fect on child nutrition, better nutrition improves
and experience in previous jobsstrongly influ- the child's capacity to learn. Studies in many de-
ences both the technical efficiency and the profit- veloping countries (China, India, and Kenya,
ability of the enterprise. among others) consistently show that protein-
There can be little doubt that educating the chil- energy malnutrition is related to lower cognitive
dren of the poor greatly improves their chances of test scores and worse school performance. Micro-
escaping poverty. Since labor is the one scarce re- nutrient deficiencies are also linked to school per-
source on which all able-bodied poor can rely, in- formance. A study of Indonesia found that iodine
creasing the productivity of this labor is clearly the deficiency reduced cognitive performance among
most effective way to combat poverty. nine-to-twelve-year-old children. Iron deficiency
decreases the child's alertness, which in turn af-
Other benefits of education fects learning. In Thailand providing iron supple-
ments to schoolchildren has improved test scores.
The effectiveness of education as a weapon in the Vitamin A deficiency has long been associated
fight against poverty goes well beyond productiv- with blindness and the severity of measles; a
ity in the labor market. One year of mother's edu- milder deficiency affects growth, including brain
cation has been associated with a 9 percent de- development.
crease in under 5 mortality. The children of The health benefits of adequate energy, protein,
better-educated mothers, other things being equal, and micronutrients are important by themselves.
tend to be healthier. The effects of education on Because of the positive effect of better health and
fertility appear to be more complicated. At low lev- nutrition on productivity, investment in these ar-
els of education a few extra years of schooling may eas also contributes to reducing poverty. And if the
actually lead to increased fertility, but after that full benefits of education are to be realized, pupils
there is a strong negative effect. In what World De- must be adequately nourished. There are many in-
velopment Report 1980 called the "seamless web of expensive things that governments can do. Provid-
interrelations" among social services, education ing vitamin A capsules twice a year and giving
plays the central role. doses of iodine once every three years would go a
long way toward eliminating nutritional disorders
Health, nutrition, and productivity among school-age children.
The effect of better health and nutrition on produc- Poverty and population growth
tivity is less well documented than the effect of
education. An increasing number of studies, how- Poverty and rapid population growth reinforce
ever, show a positive effect on agricultural produc- each other in a number of ways. Low wages (espe-
tivity, especially for the activities in which most of cially for women), inadequate education, and high
81
infant mortalityall linked to povertycontribute health suffers from closely spaced pregnancies,
to high fertility rates and thus to rapid population and her health in turn affects the child's health at
growth. Increased education for girls and women birth and in the critical early years. Providing fam-
is one of the most powerful ways of reducing fertil- ily planning services is one of the most effective
ity. Greater opportunities for women in the wage policies for better health.
sector have an independent negative effect on fer-
Family planning
tility and offer other benefits as well. When
women work, parents may decide they need fewer Having too many children puts mothers at risk and
boys to support them in old age, and if mothers compounds the problems of poverty (Box 5.3).
work for reasonable wages, the opportunity cost of Most developing countries recognize the impor-
bearing an additional child is relatively high. tance of reducing population growth. Lower fertil-
High infant mortality is consistently associated ity is not an end in itself, but it makes other inter-
with high fertility. Studies from places as diverse ventions more effective in improving overall
as Colombia, Egypt, and India show that parents welfare. Providing family planning services is the
who have lost a child expect to have more children most direct way to reduce fertility, although lower-
than do similar parents who have not. Reducing ing infant and child mortality and creating better
mortality among children is a necessary step to- education and work opportunities for women also
ward reducing fertility. help. Again, appropriate policies reinforce each
Thus, the link leading from poverty (low wages, other: better-educated women make more use of
low income, and high mortality) to high fertility is modern contraceptives (when they are available)
well established. How does high fertility or, more than do the uneducated.
generally, rapid population growth affect poverty? In many countries family planning services are
At the national level, the relationship is not sim- virtually nonexistent, despite the apparent de-
ple. In the short run an increase in population will mand. In Ghana, for example, 20 percent of
result, almost by definition, in lower per capita in- women in rural areas and 28 percent in the cities
come growth, but in the longer run the larger say they want no more children, yet modern fam-
number of productive workers may accelerate ily planning services reach less than 7 percent of
growth. It can even be argued that some countries the women. Ghana's fertility rate is 6.4. In Indone-
particularly in the Westneed faster population sia, where about half of the women want no more
growth even to sustain their current economic children, family planning services reach 44 percent
performance. of all women. Indonesia's fertility rate is 3.3.
On the negative side, in many high-fertility Family planning programs, where they have
countries about 45 percent of the population is un- been implemented, have brought birthrates down.
der 15. This puts enormous pressure on the In Costa Rica, Korea, and Singapore, for instance,
schools. Spending on education in the developing birthrates fell by 35 to 48 percent between 1965 and
world is much lower than in the industrial coun- 1985. Other factors in reducing birthrates, how-
tries to begin with, and the scarce resources have ever, should not be overlooked. Rapidly develop-
to be spread over a much larger proportion of the ing economies can bring about some modest de-
population. As a rule, the poor are last in line. The clines, even with weak family planning programs.
same holds for other sectors, such as health care But in the poorest countries (such as India and
and water supply. So, if investing in human capital Indonesia) strong family planning programs are
is necessary for reducing poverty, rapid popula- necessary to slow population growth.
tion growth will make the task much more difficult
Providing access for the poor
in many poor countries.
Population growth also affects the labor market. Providing the poor with access to social services
Rapid growth of the labor force does not necessar- requires a clear commitment. This must be re-
ily lead to unemployment and poverty: if invest- flected in the infrastructure and organization of the
ment in capital is adequate, an expanding econ- social sectors and in the way they are financed.
omy could absorb the additional labor and may Much can be learned from decades of experience in
indeed depend on it. But most countries invest too countries at different levels of development and
little to absorb a population that is growing at 3 to 4 with varying needs.
percent a year. Low wages and growing poverty
Infrastructure and organization
are the likely results.
At the household level, high fertility can damage The biggest obstacle for the poor in gaining access
the health of both mother and child. The mother's to health and education services is the lack of
82
not yet fully grown. Women could help themselves if
Box 5.3 Safe motherhood they had basic information about nutrition and health.
Improving women's income, education, health, and
In many developing countries pregnancy and child- nutrition could greatly reduce maternal mortality and
birth account for more than a quarter of all deaths of morbidity.
women of childbearing age. About half a million Family planning information and services can also
women, 99 percent of them in the developing world, improve maternal health by enabling women to time
die in childbirth each year (see Box figure 5.3). Of every and space pregnancies. In many countries 25 to 40 per-
100,000 women who give birth in Africa, from 200 to cent of maternal deaths could be averted by avoiding
1,500 may die, compared with fewer than 10 in most unwanted pregnancies. The three essential elements
developed countries. By contrast, of every 100,000 Afri- are prevention of complications, routine care, and
can women who take the contraceptive pill for a year, backup for high-risk emergency cases. Existing pro-
perhaps one will die. About three-quarters of maternal grams could be modified to stress:
deaths are from one of five causes: hemorrhage, infec- Stronger community-based health care, relying on
tion, toxemia, obstructed labor, and abortion (particu- nonphysician health workers, to screen pregnant
larly unsafe abortions, performed by untrained person- women, identify those at high risk, and refer them for
nel in unhygienic conditions). Because women in many help; to provide good prenatal care and ensure safe
developing countries, and especially in the poorer ar- delivery for women at less risk; to provide family life
eas, tend to have many pregnancies, the cumulative education and family planning services; and, in gen-
lifetime risk of dying in pregnancy may reach one in eral, to promote better family health and nutrition
twenty. Most of these maternal deaths could be pre- Stronger referral facilitieshospitals and health cen-
vented by relatively cheap and simple measures. ters with bedsto act as a backup network, to take care
A woman's health and nutritional status substan- of complicated deliveries and obstetrical emergencies,
tially affect her capacity to withstand difficulties during and to provide clinical and surgical methods of family
pregnancy, childbirth, and the postpartum period, to planning
produce a strong, healthy baby, and to nurse and care An "alarm" and transport system to transfer women
for her baby. Most pregnant women in developing with high-risk pregnancies and emergencies from the
countries are anemic, and many teenage mothers are community to the referral facilities.
I Box figure 5.3 Estimated number of annual maternal deaths, by region
Every year half a million women die in childbirth. Maternal deaths per 100,000 live births
0 100 200 300 400 500 600 700
''I!!
11111 140,000
III 308,000
Ii
'I
34,000
6,000
Sub-Saharan Africa 0 Asia 0 Latin America and the Caribbean 0 Developed countries
$ Equals 30,000 deaths
83
physical infrastructure, especially in rural areas. sector organizations in the hope of making the
The urban bias in the provision of services affects schools more responsive to labor market demands.
both quantity and quality. The sheer lack of facili- The Ministry of Education allocates resources to
ties makes necessary a continued expansion of ap- the public schools on the basis of enrollment and
propriate services, especially rural health clinics dropout rates. Private schools that provide tuition-
and primary schools. This will almost certainly free education receive the same subsidies. Imple-
benefit the poor. Improving quality will require menting the reforms was difficultthe monetary
more funds, better use of the available resources, transfers had to be suspended during the eco-
and greater accountability in administration. nomic crisis of 1982-83but enrollment in subsi-
dized private elementary schools doubled between
EDUCATION. Governments in middle-income 1980 and 1987, and the proportion of all education
countries have by and large been able to build subsidies received by the poorest 30 percent of stu-
enough schools and train enough teachers to make dents rose from 29 percent in 1974 to 38 percent in
universal primary education possible. Low-income 1986.
countries, however, still have a long way to go. For Simply expanding the education system is not
example, in Bhutan and Mali net primary school enough. In many countries the curriculum needs
capacity is only about 20 percent, approximately to be improved, the schools need more textbooks,
the same as net enrollment rates. Since these coun- and the teachers need better training, more time
tries are faced with the prospect of sluggish eco- for teaching basic skills, and better pay and work-
nomic growth and rapid population growth, it wifi ing conditions. School managers need more au-
take an extraordinary effort to provide the neces- thority to control resources, and they must be
sary infrastructure for achieving universal primary made more accountable to the local community.
education in the foreseeable future. Given the ex- Narrower interventions can sometimes be suc-
tremely low incomes of most households in these cessful in reaching disadvantaged groups. In
countries, public provision of education will be Bangladesh, for instance, a scholarship project for
needed. secondary school girls in the Sharasti Upazila re-
In some countries the private sector can help to gion, funded by the U.S. Agency for International
make schooling more widely available Pakistan in Development (USAID), has been in place since
1979 reversed its nationalization decision of 1970 1982. Before the project started, 27 percent of the
and again allowed private schools to provide edu- secondary school students in the area were girls.
cation. The education system has expanded By 1987 the percentage had risen to 44 percent.
quickly, especially at the lower levels. In the Kara- The project has also greatly reduced the dropout
chi region, for instance, private institutions in 1985 level for female students, from 15 percent before
accounted for 25 percent of primary enrollment. the program started to 4 percent in 1987. In Nepal
Because of the relatively large fees charged, these in the 1970s the United Nations Educational, Sci-
private institutions are unlikely to serve the poor. entific, and Cultural Organization (Unesco) spon-
But where demand exceeds supply and budget sored a project that included a teacher training
constraints are tight, greater reliance on the private component designed to increase the number of
sector wifi allow the government to direct more of women teachers. During the project period the
its scarce resources toward education for the poor. number of girls enrolled in school rose from 13 to
In general, the public sector will remain the 29 percent of the age group.
dominant supplier of primary education in middle-
income countries as well, although here there is HEALTH. Although the government is usually
more scope for private provision. In 1980-81 Chile the dominant provider of health care, in many
reformed its education system and embarked on a countries the private sector also provides some
policy of developing private education, partly sup- (mostly curative) services. The case for the state's
ported by public finance, with the aim of improv- being sole provider of curative care is weak. There
ing quality and increasing enrollment. The govern- is undoubtedly a need for heavy intervention, but
ment encouraged the private sector to provide this need not mean putting every doctor and nurse
education through a per student payment system on the government payroll. Although different
that forced the public schools to compete for stu- countries have different needs, several broad prin-
dents. It transferred responsibility for primary and ciples hold. First, the state should take responsibil-
secondary public schools to the municipalities and ity for health interventions that have a public-good
placed vocational schools under nonprofit private character (for example, clean air and traffic safety)
84
or that generate benefits to the community in addi- 1980 a USAID project in Nicaragua used surveys to
tion to private benefits (for example, immunization establish local needs and involved village health
against communicable diseases). Second, in cura- committees in building wells, latrines, and health
tive care the main role of the state as a provider of posts. A preschool child care program in Colombia
services should be to supply basic services in those is managed by parents' associations (Box 5.4). In
regions that the private sector is unlikely to serve. Thailand villagers have helped to build wells and
Third, if a private delivery system is in place, the health and nutrition centers. Women in a Bangla-
state should make sure that the poor have access to desh village maintain new latrines and sewage sys-
basic care. In many countries, in other words, the tems (Box 5.5). With UNICEF's assistance, wom-
government should continue to expand basic ser- en's NGOs in Indonesia have produced training
vices but with a greater emphasis on access for the booklets about child survival for twelve different
poor. religious groups. By January 1988 these booklets
Chile again offers an example. Expansion of pri- had reached about 7 million women at a cost of 10
mary health care services, with an emphasis on cents a woman.
vulnerable groups such as mothers and young Chile has transferred responsibility for many
children, was part of the reforms implemented af- public health services from the central government
ter 1974-75. New financing mechanisms were used to municipalities. The reforms are incomplete, but
to promote increased coverage. The government coverage has already been broadened substan-
encouraged the private sector to offer prepaid tially. The capacity of municipal health posts, for
health insurance plans that were to be financed instance, has gone up 45 percent since the reforms,
through payroll deductions. User fees were intro- and the number of service hours offered has in-
duced on the basis of ability to pay. creased by more than 80 percent in rural areas.
Perhaps the most important aspect of the re- Chile's long-term commitment to improving its ba-
forms was the decision to decentralize responsibil- sic health services is having results. Infant mortal-
ity for delivering primary health care and public ity declined from 103 per thousand in 1965 to 21
health care services. In many countries primary per thousand in 1985, and maternal mortality was
health care programs are designed and imple- reduced from 131 to 47 maternal deaths per
mented by highly centralized administrations. 100,000 live births during the same period.
Building health centers and training health work- China also illustrates the importance of local in-
ers can be managed from the center, but supplying volvement in the provision of basic health care.
services calls for decentralized management. In After the introduction of the economic responsibil-
Box 5.4 A community child care and nutrition program in Colombia
An estimated 25 percent of Colombia's population lives ment, to meet 80 percent of the daily requirements of
below the poverty line. In the towns the worst poverty each child. The service is much in demand. It started
is borne disproportionately by children, who are at risk on a broad scale in 1987 and by late 1989 had served
from malnutrition, illness, neglect, isolation, and vio- half a million children. Its goal is to reach all children at
lence. To address these problems, the Colombian gov- riskmore than a million of themby 1994.
ernment and local NCOs developed a system of pre- Participating children receive improved nutrition and
school child care that includes a feeding program and care as well as exposure to preschool learning activi-
health monitoring. ties. Community mothers are benefiting from addi-
The target groupchildren age 2 to 6 and their tional income, and parentsoften single female heads
parentsis drawn mainly from the poorest 20 percent of householdgain an opportunity to seek remunera-
of the population. A group of parents selects a "com- tive employment outside the home. The cost of the
munity mother" to provide day care and other services program, at about $11 a month per child, compares
for fifteen children in her home. With help from the well with $33 a month for day care provided through
National Family Welfare Institute, the community the institute's centers. In addition, the program's sub-
mother receives training, a small monthly stipend, and sidies are better targeted to the poor; day care centers
a credit to upgrade the home to minimum standards of largely serve a middle- and lower-middle-income clien-
hygiene and safety. The institute also provides food, tele.
including a domestically produced nutritional supple-
85
Box 5.5 Mirzapur, Bangladesh: water supply, sanitation, and hygiene education
Bangladesh, heavily populated and one of the poorest trines. An important element of the project was an
countries in the world, suffers from endemic health education program designed to persuade villages to
problems. The mortality rate for children under 5 was use handpumps for all domestic purposes, to consume
204 per thousand in 1980-85the highest in Asia and more water, to store it hygienically, and to use and
one of the highest in the world. Diarrheal diseases ac- maintain the water-sealed latrines.
count for 29 percent of all deaths of children under age One hundred and forty-eight Tara handpumps and
5 and for roughly one in five deaths in all older age 754 latrinesone for every householdwere provided.
groups. Sixty percent of children under 5 experience Villages made a financial contribution for the equip-
moderate to severe malnutrition. These problems ment. Classes in hygiene were complemented by fur-
stem, at least in part, from the poor quality and man- ther training and household visits over several months.
agement of Bangladesh's water, which in turn are Twelve local women were trained to empty the latrine
linked to frequent floods. pits. They were paid about $1.30 each per pitmuch
More than 82 percent of rural villagers say they use less than the price quoted by professional sweepers.
tubewell water for drinking, but only 12 percent use it The project achieved high rates of handpump water
for all their domestic needs. This means that the vast consumption. The villagers used and stored water
majority still uses fecally contaminated surface water more hygienically than before, and 90 percent of the
for other purposes, such as bathing and washing uten- latrines were in regular use.
sils and clothes. The effect of the project on health was evaluated
In 1984 the Interregional Handpump Project, spon- through a series of surveys. In the baseline year the
sored by the United Nations Development Programme incidence of diarrheal diseases in the project area was
and the World Bank, initiated a study on the health similar to that in a control area. After the intervention
impact of handpump water supplies, pit latrines, and the incidence of persistent diarrhea was 40 percent
hygiene education in the Mirzapur region of Bangla- lower in the project villages than in the control area,
desh. The project sought to encourage people to use and child dysentery was 30 percent lower. In the last
well water instead of surface water for domestic pur- two years of the project, children in the project area
poses by supplying new and more efficient Tara hand- suffered from diarrhea for, on average, thirty-five days,
pumps. It also introduced hygienic water-sealed la- compared with sixty-three days in the control area.
ity system, village-based rural health insurance tion will serve both efficiency and equity objec-
programs collapsed. Local communities re- tives. Recommending such a shift does not mean
sponded with a variety of schemes to secure ade- denying the importance of higher-level services.
quate financing for basic carefrom health mainte- Any country that wants to compete in the world
nance organizations with prepaid insurance in economy needs a comprehensive education policy
some of the larger cities to arrangements that fi- that includes spending on higher education, sci-
nance village health posts out of charges for sales ence and technology, and professional training.
of medicine and for curative treatment. Hospitals are an essential part of a balanced health
care system, and they are the educational and re-
Financing health and education search institutions that train new staff and gener-
ate new knowledge. These considerations, how-
Public funds provide most of the resources for the ever, do not justify using higher-level services as a
social sectors. The demand is increasing: basic ser- means of transferring government money to privi-
vices need to be expanded so as to reach the poor, leged students and urban patients from middle-
and rapid population growth in some regions and and upper-class backgrounds. Financial reform,
aging populations in others add to the burden on including greater reliance on alternative financing
all the social services. But the supply of funds is mechanisms such as student loans and health in-
being squeezed. Slow economic growth, tight bud- surance plans, can free up resources that can be
get constraints, and competing calls on the govern- used to expand and improve basic services and
ment purse from outside the social sectors mean provide better access to the poor.
that priorities must be examined carefully. In many cases more government spending on
A shift in the allocation of funds from higher- the social services (especially at the cost of military
level services to basic health and primary educa- expenditures) would be fully justified. In some
86
Table 5.1 Expenditures on primary education There is also ample evidence that people are wifi-
as a percentage of GDP, 1985 and 2000 ing to pay for higher education. Higher fees for
2000 secondary and higher education would garner re-
(at current 2000 sources for improving the quality of education in
enrollment (universal
Region 1985 rates) enrollment) general and, where necessary, for expanding pri-
2.53
mary education. The effect of fees on the demand
Sub-Saharan Africa 1.41 1.35
East Asia 1.42 0.60 0.60 for primary education is less clear. If there are few
South Asia 0.95 0.63 1.08 public schools, private schools can charge substan-
Eastern Europe 1.17 0.71 0.71 tial fees by catering to the rich. But if the aim is to
Middle East and North Africa 2.05 1.47 1.76 increase the enrollment rates of the poor, fees are
Latin America and the counterproductive.
Caribbean 1.06 0.83 0.92
Both approachesshifting resources from higher
levels of care to basic services and introducing fees
cases it is likely to be necessary. This is illustrated for those who can afford themtransfer resources
by Table 5.1, which shows the estimated cost of from the rich to the poor. If reducing poverty is a
achieving universal primary education by 2000. primary objective, this policy is fully justified.
Projections for population and economic growth
suggest that many regions will be able to achieve Creating demand
that goal by allocating to primary schooling a simi-
lar percentage of GDP as in 1985, or even a smaller Even when cheap or free social services are avail-
share. In Sub-Saharan Africa, however, although able, the poor may not take advantage of them.
enrollment levels can be raised significantly, reach- There are two reasons: they may not understand
ing universal primary enrollment will take an ex- the benefits being offered, or they may be deterred
traordinary effort, including a generous increase in by the private costssuch as working time forgone
foreign aid. Growth in GDP of 3.7 percent a year, in visiting a health center or loss of household in-
combined with an expected increase of 3.4 percent come when a child goes to school.
a year in the population age 6 to 11, wifi require an Family planning is a good example of why poor
85 percent increase in the share of GDP allocated people may not use social services. Health educa-
to primary schooling, from 1.4 to 2.5 percent. In tion campaigns are needed to explain the benefits
some countries outside Africa, too, more spending of lower fertility and to describe the various birth
on education and other social services wifi be nec- control techniques. But even when family plan-
essary. ning services are available and understood, it may
But much can also be gained in two other ways: not be in the household's private interest to have
(1) by shifting current outlays to more efficient fewer children. For instance, having fewer chil-
uses (for example, by emphasizing primary educa- dren may make it possible to give each child a
tion over higher education and vifiage health posts better education, but the benefits may not be felt
over city hospitals) and (2) by trying harder to re- for years. Where national policies have produced a
cover the costs of certain services. population growth rate that is faster than socially
How much scope is there for charging for ser- desirable, governments need to encourage family
vices? Recent evidence shows that the demand for planning by strengthening the private incentives.
medical care is sensitive to its pricemore so for Singapore, for instance, uses tax incentives and
the poor than for the rich. This implies that if fees housing policies to make having smaller families
are collected (or increased), the poor wifi be the more attractive. China has used bonuses, access to
first to drop out and that health services for poor education, and housing privileges in the same
areas wifi have to be heavily subsidized. This need way. Thailand combines community involvement,
not mean that service is entirely free; nominal fees extensive public education, mass sterilization cam-
could be charged in poor areas. Charging higher paigns, rural development programs, and direct
fees in better-off regions or in the richer parts of benefits for households with fewer children (Box
cities makes good sense and can generate substan- 5.6).
tial revenues. In addition, fees should be consid- Girls' education is another case in point. The
ered for those types of care (especially hospital private and social returns to women's education at
care) that yield mostly private benefits; this would the primary and secondary levels are at least equal
allow more resources to be devoted to public to the returns to education for men. Still, parents
health measures and preventive care. often prefer to invest only in their sons' education.
87
Box 5.6 Family planning buffaloes and vasectomy festivals in Thailand
Thailand is the home of one of the most successful and Vasectomy marathons are held on Labor Day and on
effective family planning programs in the world. The the king's birthday. In 1983 a team of forty doctors and
program relies on the use of media, economic incen- eighty nurses performed a record-breaking 1,190 vasec-
tives, and community involvement to increase the use tomies during the one-day festival. The PDA also deliv-
of contraception. ers free, convenient, and efficient vasectomy services
Thailand's striking approach stresses the immediate in clinics and mobile units. Between July 1980 and June
practical benefits from lowered fertility and uses meth- 1984 it performed 25,412 vasectomies.
ods that mix humor and audacity. The program is run Registered family planners may also rent cattle for
by an NGO, the Population and Community Develop- plowing their fieldsfamily planning buffaloesat half
ment Association (FDA), with the support of the Min- the regular price. They are encouraged to market some
istry of Public Health. of their products through the CBFPS at prices about 30
The Community Based Family Planning Services percent higher than regular middlemen offer and to
(CBFPS), a PDA program, was initiated in 1974 and buy fertilizer and seeds at 30 percent below local mar-
now reaches more than 16,000 villages and 17 million ket prices. Under a similar program villagers who prac-
peopleone-third of the nation's population. The key tice contraception may have their goods transported to
message links population growth to low standards of the marketplace at a discount or may receive free
living and family planning to economic gains. To con- piglets.
vey this message, taboos surrounding birth control had In 1985 about 60 percent of all Thai couples practiced
to be broken down. Birth control carnivals, games, some form of birth control. The total fertility rate fell
condom-blowing balloon contests, raffles, village fairs, from 6.1 in 1965-10 to 2.8 in 1985, as against an average
and weddings have served as occasions for promoting 4.1 for lower-middle-income countries.
family planning joyfully.
Measures such as increasing the number of female and indeed went into reverse, in the 1980s. If re-
teachers or paying parents to send their girls to cent trends prevailed, Sub-Saharan Africa's enroll-
school (as in Bangladesh) may help to reduce the ment rate would be a paltry 46 percent in 2000. The
present inequalities. But longer-term policies to in- pace of progress has increased recently in South
crease women's participation in the labor market Asia, and universal enrollment might be reached
will be needed if the bias against girls' education in there by 2000 (Table 5.2).
some parts of the world is to be eliminated. A similar picture emerges for mortality among
children. If long-term trends continued, all regions
Lessons for the next decade would make progress, but under 5 mortality in
Sub-Saharan Africa will remain high if the slow-
The past three decades have seen tremendous down in the 1980s is not reversed. The experience
progress in improving health, nutrition, and edu- of the 1980s indicates that accelerated progress can
cation in the developing world. In many cases be expected in South Asia, the Middle East and
progress has been considerably faster than that of North Africa, and Latin America. The total fertility
the industrial world when it was at a comparable rate is likely to come down in every region but will
stage of development. But it has been uneven. The still be a high 5.4 in Sub-Saharan Africa by 2000.
poor of the developing countries continue to suffer Thus, the picture is mixed. Although the outlook
from illiteracy, sickness, malnutrition, and early is for steady progress in the long run, more recent
death. experience shows that in some cases progress has
If the trends prevailing since the 1960s were to accelerated and in others has slowed or reversed.
continue, universal primary enrollment would be a Even if the long-term trends prevail, in thirty-six
fact by the end of the century in East Asia and countries of Sub-Saharan Africa, South Asia, and
Latin America. The countries of the Middle East Latin America, under 5 mortality rates will still be
and North Africa would approach that goal, and in at least ten times higher than in the industrial
South Asia and Sub-Saharan Africa enrollment world. That is, in 2000 almost ten million children
rates would rise from the current 74 and 56 percent under 5 will die in the developing world, for
to 88 and 86 percent, respectively. Unfortunately, broadly the same reasons that children were dying
in Sub-Saharan Africa this long-term trend stalled, there in 1900.
88
Table 5.2 Projections of social indicators to 2000
Net enrollment rates Under 5 mortality Total
(percent) (per thousand) fertility
ratee
Long-term Short-term Long-term Short-term
Region 1985 t,inda trend" 1985 trnd trendd 1985 2000
Sub-Saharan
Africa 56 86 46 185 136 153 6.1 5.4
East Asia 96 100 100 54 31 33 2.7 2.2
South Asia 74 88 100 150 98 83 4.7 3.4
Middle East and
North Africa 75 94 95 119 71 65 5.1 3.9
Latin America and
the Caribbean 92 100 100 75 55 32 3.6 2.5
Note: All data are weighted except that for under 5 mortality, short-term trend. All mortality data are projected to 2000-05.
Based on the 1965-85 trend.
Based on the 1980-85 trend.
Based on the trend used by the United Nations.
Based on 1975-80 to 1980-85 except for South Asia, 1970-75 to 1975-80.
The total fertility rate is the average number of children that would be born alive to a woman during her lifetime if she were to bear children in
accordance with the prevailing age-specific fertility rates. It is based on 1985-90 and 2000-05.
Source: United Nations and World Bank data, except for under 5 mortality, short-term trend, from Hill and Pebley 1988, and fertility, from Bulatao
1989.
To avert that shameful prospect, governments, Second, make the existing services more effective.
in cooperation with the international community, With these principles as the guide, equity and effi-
must make a long-term commitment to improving ciency can go hand in hand. Improving the quality
the social infrastructure that serves the poor. The of life of the poor will then become an achievable
key elements of such a commitment are clear. First, goaland the progress made in the past will stand
provide the basic services that the poor need most. as a sign of what can be done in the future.
89
Transfers and safety nets
Not all the poor will benefit from the policies dis- South Asia, Sub-Saharan Africa, and Latin Amer-
cussed in Chapters 4 and 5. In the first place, it ica often provide for the subsistence needs of old
may take a long time for some of the poor fishermen or of poor fishing families whose male
including the working poor and those in remote income earners have died. Old fishermen are com-
regionsto fully participate, and the old or dis- monly assigned comparatively easy tasks on
abled may never be able to do so. Second, even shore, and children from needy families are al-
among those who do benefit from the policies, lowed to take some fish from each boat. In some
there will be some who remain acutely vulnerable Senegalese villages retired fishermen are granted a
to adverse events. The first group can best be fixed share of the catch. This "tax" is paid not only
helped through a system of income transfers. The by local fishermen but also by fishermen from
second group requires a safety netsome form of other areas. Fishing communities are often well
income insurance to help people through short- suited to such arrangements because they have a
term stress and calamities. This is particularly im- high degree of social homogeneity.
portant when food security, and hence survival, is But many of the poor, particularly those of con-
threatened. cern in this chapter, are inadequately protected by
As a rule, government interventions will blend these strategies. There are limits to what those un-
elements of both redistribution and insurance. able to work can do to protect themselves, and
This chapter examines a range of policies under limits to what the households looking after them
the following headings: can do to raise household income to tolerable lev-
Food pricing and distribution els. Growing evidence points to the dissolution of
Public employment schemes family and ethnic ties and the weakening of com-
Social security munity support systems in many developing
countries. In addition, coping arrangements that
These complement the policies described in pre- work well in normal times may fail when they are
vious chapters and form a crucial ingredient in the needed most. Poor, vulnerable households, on the
overall strategy. Some of them have a surprisingly margin of subsistence at the best of times, will be
long history (Box 6.1). hard pressed to protect themselves in the face of a
Individuals, families, and communities have series of shocks. And the effectiveness of commu-
ways of coping with poverty. Individuals and fam- nity insurance depends crucially on the extent to
ilies redistribute and stabilize consumption, diver- which local incomes are affected simultaneously.
sify income sources, migrate, and give and receive Unanticipated shocks such as severe agricultural
transfers. In many parts of Sub-Saharan Africa the fluctuations, deterioration in the terms of trade,
term for being poor is synonymous with lacking and famines can create hardship for entire commu-
kin or friends. At the community level these "so- nities.
cial security" arrangements are sometimes quite The state therefore has a role in aiding house-
sophisticated. For example, fishing villages in holds or communities in times of insecurity and in
90
Box 6.1 The historical antecedents of public action
Public food distribution has a history of several thou- erable to charitable transfers, which they thought in-
sand years. In Egypt storage of food by the state, ac- duced laziness. Perhaps the most significant use of
companied by subsidized sales and public distribution public works as relief was the Works Progress Admin-
to the needy, has existed since the time of the phar- istration, which provided employment for one-fifth of
aohs. It was considered to be the moral responsibility all U.S. workers during the Great Depression of the
of rulers and was an important element in maintaining 1930s.
social stability when crises arose because of shortfalls At the turn of the century assistance to the poor con-
in domestic production, military demands, and disrup- sisted mainly of charitable transfers and poor relief
tions of trade. The distribution of a grain or bread ra- based on stringent needs testing that stigmatized the
tion in ancient Rome and Greece was limited to crisis recipients. These forms of assistance were increasingly
periods precipitated by war or climatic instability. As challenged on moral grounds as demeaning to the re-
rural transport and irrigation systems have been devel- cipients, and the world wars and the Depression cre-
oped and private and public food distribution systems ated demand that they could not meet. The rise of the
have improved, the ability of governments to avert modern industrial state allowed the development of
famine has gradually increased. formal social security systems based on contributions
Public provision of employment also has a long his- from workers and employers and on wide coverage.
tory as a means of reducing poverty. As early as the Publicly provided pensions for the aged were an im-
I fourth century B.C. Indian rulers were advised that portant feature of the new systems. Unemployment
when natural calamities struck, subjects should be em- provisions were instituted after the Depression. Fol-
ployed in building forts or waterworks in return for lowing World War II coverage became universal, with
food. Employment on public works later became the family allowances and a minimum level of benefits for
main element of strategies for famine prevention in the sick, unemployed, and aged. The existence of guar-
India, and it has proved effective. In eighteenth- and anteed public health, income support, and insurance
nineteenth-century Europe the poor qualified for relief systems has provided to the poor in most developed
if they agreed to provide labor or to live in squalid countries a degree of security that developing countries
poorhouses. Victorians considered this approach pref- still lack.
ensuring minimum levels of provision to those un- earn in public employment. This has been ob-
able to gain from the growth process. Effective pol- served in various public employment schemes in
icies will take into account what individuals and South Asia. By contrast, child feeding programs
households can do to protect themselves and what may have less than their expected impact if the
their limitations are. Government interventions allocation of food within the household shifts to-
should also try to involve people in the solutions ward other family members.
rather than treat them as passive recipients of In measuring the benefits to the poor, govern-
relief. ments must also take account of the method of
financing. If transfers are financed by printing
Public provision of transfers and safety nets money, the loss to the poor from the ensuing infla-
tion may well exceed the nominal income transfer.
In assessing policy alternatives, the tests are effec- But if progressive taxation is used to finance a pol-
tiveness in reaching the poor and the extent to icy, the system may be highly redistributive even if
which the policies compromise other objectives. the benefits go to the rich as well as to the poor.
For many practical purposes, these can be mea- Often, schemes are financed by cutting spending
sured by considering cost-effectivenessthe bud- on other social programs. The net effect on the
getary cost of delivering a given amount of benefit. poor will then depend on the balance between the
Benefits usually change the way the poor and benefits lost and gained.
others behave, and this affects cost-effectiveness. Targeting benefits to the poor can be an impor-
For instance, a public employment program tar- tant way of increasing cost-effectiveness. There are
geted at landless laborers may have important exceptions, but for a given budgetary outlay,
second-round effects on local agricultural wages; schemes that target benefits to the poor are likely
the program's overall effect on the incomes of the to have a greater impact on poverty than universal
poor may be greater than the wages participants schemes. The exceptions arise when targeting is
91
Box 6.2 Reaching the poor through public action
Ideally, government policy would identify the poor household-specific characteristics. The trouble is that
without cost and would channel benefits exclusively to the easily monitored correlates of poverty are often im-
them. In practice, policymakers lack information on in- perfect: some poor people will be missed, and some of
dividual living standards, and the costs of obtaining the nonpoor will receive benefits. Indicator targeting
this information can be high. Means testing and other can also lead to incentive effectsfor instance, non-
forms of targeting have social costs or cause distortions poor people might migrate to target areas. The chal-
because they lead to changes in behavior. In the United lenge is to find good indicators of poverty that can be
Kingdom means testing for supplementary benefits re- monitored easily by administrators but cannot be al-
quires an elaborate administrative machinery for verifi- tered easily by the nonpoor. Caste status (in India),
cation and recordkeeping. In developing countries the widowhood, and old age may be good examples.
identification of income and assets is even more diffi- A good way to design targeted programs is to make
cult, especially in rural areas, and the administrative benefits contingent on work, as in rural public employ-
problems are likely to be daunting. As a rule, means ment schemes. Another is to subsidize goods that are
testing is simply not feasible in these countries. mainly consumed by the poor. Such policies are called
An alternative is indicator targeting. The basic idea is self-targeting because the "cost" they impose will (ide-
simple. Correlates of poverty are identified from sam- ally) make participation desirable only for the truly
ple surveys or other information, and direct benefits poor. For example, work in return for wages that are
are allocated accordingly. In general, it is less costly to generally lower than other unskilled wages will dis-
identify the correlates, which may include region of suade those with better jobs or other means of liveli-
residence, landholding, nutritional status, sex, age, hood. This screening method is a key component of
and race, than to measure incomes. Food subsidy employment schemes in both Bangladesh and India.
schemes in Colombia have successfully combined geo- Often, self-targeting is more cost-effective than uni-
graphic targeting with further targeting based on form provision.
administratively costly or imposes substantial policy has to be alive to these considerations of
losses on the poor (such as income forgone in political economy.
queuing for food rations). One promising avenue
is to use incentives as a screening device so that Food pricing and distribution policies
the poor are encouraged to participate and the
nonpoor are discouraged (Box 6.2). The poor spend a large part of their incomes on
In practice, the success of public intervention in- food and can be highly vulnerable to any setback
volves more than cost-effectiveness. The demands in their ability to obtain it. An important way to
made by different sections of the population, and protect the living standards of the poor is by guar-
their ability to exert pressure on the authorities, anteeing their food securitythat is, by making
are often more influential than the government's sure that they can acquire adequate food at all
economic calculations. Fine targeting based on a times. The developing countries have adopted var-
single-minded concern for cost-effectiveness can ious forms of food subsidy: general food price sub-
reduce public interest in the vigorous implementa- sidies, food rations, food stamps, food distribution
tion of government programs to help the poor. For policies, and food supplementation schemes.
example, in the late 1970s Sri Lanka replaced a These policies can be used to raise the real incomes
universal food subsidy with a less costly targeted of the direct beneficiaries, and they can provide a
food stamp program. In time, the benefits deliv- safety net to protect a wider group of the poor
ered by the new program declined. The middle against collapses in their real incomes.
classes no longer gained from the scheme, and al-
though the new program was more cost-effective, General food price subsidies
it lost crucial political support. Similarly, a food
subsidy directed to poor consumers in Colombia General food price subsidy schemes supply unlim-
was so tightly targeted that it lacked an effective ited amounts of subsidized food to anyone who
political constituency, and it was dropped at a wishes to buy it. Brazil, China, Colombia, Egypt,
change of administration. The analysis of public Mexico, Morocco, Pakistan, Sudan, Thailand, Tu-
92
Table 6.1 Distribution of the annual income dized goods are often concentrated in urban areas,
transfer from the general food subsidy, Egypt, either by design or because market channels by-
1981-82 pass rural consumers. Rural consumers must have
Transfer as access to markets if price subsidies are to help to
Type of Amount of percentage of reduce rural poverty.
household and transfer household
income level (Egyptian pounds) expenditure A better way to reach the poor is to limit subsi-
Urban dies to commodities that are consumed mainly by
Poorest quartile 15.4 8.7 the poor and that form a significant part of their
Richest quartile 18.1 3.4 food expenditures. Such goods are not always
Rural available and are not always easy to subsidize. But
Poorest quartile 11.9 10.8 there are examples. In Egypt benefits from subsi-
Richest quartile 15.2 2.7 dies on coarse flour accrue mainly to low-income
Source: Alderman and von Braun 1984 and Alderman 1989. groups. Shifting Brazil's general subsidy to cas-
sava and away from wheat, milk, beef, and vegeta-
nisia, and, before 1979, Sri Lanka have all operated ble oils, which are heavily consumed by the
schemes of this kind, some national and some re- higher-income sections of the population, would
gional. Costs ranged from less than 1 percent of improve targeting to the poor. The choice of com-
total public expenditures in Colombia in 1978-80 to modity may also influence distribution between
10-17 percent in Egypt between the mid-1970s and sectors. For example, Mexico has a consumer sub-
1984. The experience of Egypt well illustrates some sidy on maize tortillas. Since outlets are concen-
of the advantages and drawbacks of these trated in urban areas and urban consumers buy
schemes. tortillas whereas rural consumers make them at
Egypt's marketwide food subsidy program has home, the subsidy has an urban bias. In Egypt the
been costly, but it has succeeded in reaching the urban poor gain more from the bread subsidy than
poor. In 1981-82 the transfer value represented a do the rural poor, but the reverse is true for the
sizable portion of the total expenditure of the poor- wheat subsidy.
est urban and rural groups-8.7 and 10.8 percent, Better targeting of price subsidies to the poor is
respectively (Table 6.1). possible when the government knows where the
A common feature of these schemes is that the poor live. The Pilot Food Price Subsidy Scheme in
rich receive a greater allocation per capita than the the Philippines is an example of how to improve
poor, although this amount often accounts for a cost-effectiveness by combining geographic target-
smaller share of income. In Egypt urban house- ing and consumer price subsidies. The National
holds in the top quartile received 18.1 Egyptian Nutrition Council's anthropometric reports helped
pounds (equivalent to 3.4 percent of household ex- to identify poor villages with high rates of child
penditures), compared with 15.4 Egyptian pounds malnutrition. Seven villages were selected to re-
and 8.7 percent of household expenditures for ur- ceive price discounts on rice and cooking oil
ban households in the lowest quartile. All benefits goods that account for a large part of food expendi-
going to the nonpoor add to the budgetary cost of tures. These were sold through local retailers and
reducing poverty. Indeed, if general subsidies are were available to all villagers. The scheme im-
to provide reasonable transfers to the poor, they proved the nutritional status of both preschoolers
become very expensive. In Egypt only about and adults. It was highly cost-effective; eighty-four
twenty cents of each dollar spent reached those in cents out of each dollar spent were transferred.
the lowest quartile. This, however, did not include setup costs for ex-
The bulk of Egypt's price subsidy transfer is tensive growth monitoring, which have to be
made by subsidizing bread and wheat flour sold at taken into account in determining overall cost-
bakeries and flour shops. Administrative ease is effectiveness. Few countries have in place the na-
one advantage of general food subsidies. Since tional nutrition surveillance systems needed to run
most subsidized commodities can be sold through such schemes.
private outlets, public involvement in distribution
is not necessary. Central administration of the Rationed food subsidies
price mechanism and low investments in infra-
structure help to contain transfer costs. In many An alternative to a general subsidy is to provide a
countries, however, general price subsidies fail to quota, or "ration," of subsidized food to house-
reach many of the poor. Outlets for sale of subsi- holds while permitting unlimited sales on the open
93
Table 6.2 Distribution of the monthly transfer As is true of general subsidies, ration schemes
from food subsidies, Sri Lanka, 1978-79 are often limited in coverage because the irifra-
Transfer as structure and retail networks needed to implement
Subsidy and Amount of percentage of the systems in rural regions are lacking and be-
household transfer per capita
income level (rupees) expenditure cause of a desire not to upset production incen-
General wheat and bread subsidy
tives. These are often compounded by a lack of
Poorest quintile 4.3 7.6 political will to assist the rural poor. Guaranteeing
Richest quintile 7.8 3.8 large sections of the population a basic ration on a
Targeted rice ration long-term basis may place a great burden on the
Poorest quintile 11.1 19.6 government in terms of both administrative capa-
Richest quintile 3.3 1.6 bility and cost.
Source: Edirisinghe 1987 and Alderman 1989. In the past ration programs have often been set
up for political purposes and have rarely been tar-
geted to the neediest. But it is possible to introduce
market. Schemes of this type operate or have oper- a degree of self-selection and better targeting into
ated in Egypt and in South Asia (Bangladesh, In- rationing schemes without costly means testing.
dia, Pakistan, and Sri Lanka). In addition to trans- Ration outlets can be strategically located in poor
ferring income, ration schemes ensure access to a neighborhoods. The quantities distributed should
regular supply of basic staples at reasonable prices. be small enough to allow poor households without
The absolute transfer under a general ration is sim- much cash to participate. The opportunity cost of
ilar for all income groups. Thus, rations tend to be buying a restricted quantity at a ration shop lo-
more progressive than general food subsidies. For cated in a slum is greater for the rich than for the
example, the absolute transfer to the richest quar- poor. In Northeast Brazil small amounts of subsi-
tile in urban areas under Egypt's general food sub- dized basic foods were sold through shops in poor
sidy was 20 percent greater than that received by neighborhoods that the well-to-do regarded as un-
the poorest quartile, whereas with the general ra- safe; the inconvenience of the locations and the
tion the rich received 5 percent less than the poor. limit on the size of purchase further discouraged
The Sri Lankan general rice ration scheme, affluent nonresidents from participating. The qual-
which operated between 1942 and 1978, had a ity of the subsidized commodity can also be lower
large impact on poverty both because the benefits than that available on the open market. This
that reached the poor were high and because there method was used to skew the benefits of Paki-
was extensive coverage of different types of poor stan's pre-1987 ration system toward lower-
people. To reduce costs, the government in 1978 income groups. In Bangladesh better targeting has
restricted distribution to the poorest half of the been achieved by offering a choice of wheat or a
population. The targeted scheme can be compared larger quantity of sorghum. Poorer households
with the general subsidy on wheat and bread of tend to opt for the latter, thus improving the nutri-
the same period. As Table 6.2 shows, benefits to tional impact of the program.
the poor from the curtailed rice ration are greater
both in absolute terms and in relation to income. Food stamps
Given adequate coverage of the poor, targeted ra-
tions thus represent an improvement over both Food stamps are similar to ration schemes except
general subsidies and unrestricted rations. that the quota is measured in terms of nominal
A similar pattern of transfers is seen in the distri- currency units rather than in commodity weights
bution of foodgrains through fair-price shops in or volumes. In practice, this means that the value
certain states of India. In Kerala in 1977 the poorest of the food stamps is often eroded by price infla-
60 percent of the population received 87 percent of tion. Food stamps are usually more finely targeted
the foodgrains distributed. Kerala is unique in that to needy groups than are rations.
rations are spread fairly evenly between the rural In the face of fiscal crisis, Sri Lanka in 1979-80
and urban populations. In other parts of India as replaced its four-decade-old general food subsidy
well as in other countries, including Bangladesh and ration schemes with a food stamp program.
and Pakistan, the benefits of ration systems have The total transfer was lower, and the reduction in
gone disproportionately to urban consumers administrative machinery led to substantial sav-
although poverty is mainly a rural phenomenon in ings. Stamps for food and kerosene were targeted
South Asia. to families with self-reported incomes of less than
94
300 rupees a monthroughly 50 percent of the Table 6.3 General and targeted subsidies,
population. The government had removed all Jamaica, 1988
(percent)
other subsidies on food by 1980. The share of food
subsidies in GNP dropped from 5.0 to 1.3 percent Targeted
between the mid-1970s and 1984, and their share in General subsidy
Item subsidy (food stamps)
total government expenditure fell from 15 to 3 per-
Cost as share of government
cent during the same period. But the value of food expenditure' 3.0 1.6
stamps, which was fixed in nominal terms, quickly
Proportion of transfer going to
eroded in the 1980s, falling from 83 percent of the Poorest quintile 14.0 31.0
benefits of the general subsidy at the time of intro- Richest quintile 26.0 8.0
duction to 43 percent in 1981-82. Transfer as share of expenditure
Targeted food stamps yield a more progressive per recipient
pattern of transfers than general food subsidies. Poorest quintile 2.3 9.5
Jamaica's food stamp scheme was introduced in Richest quintile 0.1 1.0
1984 to protect vulnerable groups from the full im- Proportion of households covered
pact of exchange rate movements and reduced Poorest quintile 100.0 51.0
public spending. Stamps are targeted to pregnant Richest quintile 100.0 6.0
and lactating women and to children under 5 a. Does not include administrative costs.
Source: Jamaica Statistical Institute and World Bank 1988 and 1989.
through registration at primary health care clinics.
This system encourages preventive health care and
successfully screens out wealthier households,
which tend to use private facilities. Coverage is tribute subsidized or free food through noncom-
high among the intended beneficiaries, and this mercial channels such as nutrition and health cen-
part of the program is markedly progressive. Poor, ters for direct or home consumption or through
aged, and handicapped people who are already on schools to those deemed specifically vulnerable to
welfare programs, as well as households that re- nutritional and health risks. Children under 5,
port total income of less than 2,600 Jamaican dol- schoolchildren, mothers, and mothers-to-be are
lars a year, are also eligible for food stamps. This the main beneficiaries. Additional targeting on the
component is less well targeted; there is both in- basis of growth monitoring, health status, loca-
complete coverage of eligible households and leak- tion, or income can be used to identify the neediest
age to nontarget ones. members within these groups. The scope of these
Jamaica reinstated a general food subsidy in operations can be expanded to cover wider sec-
1986. Table 6.3 shows that although coverage is tions of the population in time of crisis.
wider under the general subsidy scheme, targeting The case for food supplementation as an imme-
through food stamps has a larger impact on the diate response to severe malnutrition is undeni-
incomes of the poor, and at about half the cost. able. Measuring the benefits of these schemes,
Administrative costs (which are not reflected in the however, is far from easy. Anthropometric criteria
table) are kept low by relying on existing govern- are often used, but they require careful measure-
ment networks. The value of the stamps has ment and may not capture other important bene-
eroded over time, although adjustments were fits (such as increased activity levels) realized by
made in 1988 and 1989 and more are expected in people who are not severely malnourished. Evalu-
1990. The Jamaican food stamp program is cost- ations on this basis have tended to show that the
effective and, with some modifications, can be fur- nutritional impact on target groups is limited and
ther improved. Its success, however, owes much is achieved at high cost. Distribution is often indis-
to the availability of a highly developed health and criminate, and leakage through food sharing or
administrative apparatus and the presence of polit- substitution of home consumption has been esti-
ical will. Both are lacking in many countries. mated to be as high as 30 to 80 percent. Studies of
large-scale untargeted school feeding programs
Supplementary feeding programs show little evidence of significant nutritional
improvementalthough the effects on enrollment,
Supplementary feeding programs are a form of attentiveness, and school performance may be sig-
highly targeted ration or in-kind transfer scheme. nificant and important in their own right.
Their main objective is to reduce undernutrition. Feeding programs are sometimes ineffective be-
Government and nongovernmental agencies dis- cause they fail to recognize that malnutrition is not
95
always the result of lack of food. Parasitic diseases maternal and child health, water and sanitation,
and other health problems are often to blame. Un- household food security, child care, and nutrition
dernourished people with acute diarrhea, for ex- education. By providing child care the scheme en-
ample, are able to absorb far fewer nutrients. To be couraged the participation of very poor women
effective, therefore, nutrition programs will fre- with heavy demands on their time. Between 1984
quently need to be combined with health and 1988 it reduced severe malnutrition by 72 per-
programsincluding maternal care, immuniza- cent and moderate malnutrition by 32 percent. The
tion, sanitation, deworming, oral rehydration ther- program had fairly low costs and benefited from
apy, and micronutrient supplementation. Educa- the existence of vifiage organizational structures
tion in primary health and nutrition is also crucial and Tanzania's well-developed rural health care
where local practices are inappropriate. Education infrastructure. Bangladesh's Vulnerable Group
programs in Indonesia and the Dominican Repub- Development Program broadened its emphasis on
lic have significantly improved nutrition at low food supplementation for poor, nutritionally vul-
cost and without any food transfers. Households nerable women and children to include women's
may be unable to respond owing to lack of re- training centers and shelters. It is currently explor-
sources. In this case it is appropriate to combine ing savings and credit schemes.
nutrition education with cash or food transfers. Thus, delivering basic services through existing
It is often cost-effective to deliver nutrition ser- health care systems can be highly cost-effective.
vices through the health care systemprovided, of This approach has been successful in removing ba-
course, that an extensive health system is already sic deprivations in countries as diverse as Chile,
in place. In much of Sub-Saharan Africa and South China, Costa Rica, Cuba, Jamaica, Korea, and Sri
Asia this is not the case, and alternative delivery Lanka.
mechanisms must be sought in the short term. But
in Latin America and much of East Asia, provision Food policies in famines
through the existing social infrastructure can be
achieved at relatively low cost. International Public policies in food trade and distribution can
NGOs such as Freedom from Hunger, OXFAM, play an important role in a system of famine pre-
and Save the Children (as well as a host of local vention. Areas facing a bad harvest often see food
NGOs) have promoted health and nutrition along- prices spiral as traders anticipate scarcity. Food-
side community development. grain markets have often overreacted to impend-
Tight budgets and the proven ineffectiveness of ing shortages. For example, in the 1974 famine in
indiscriminate provision make it essential to iden- Bangladesh, rice prices increased sharply even be-
tify those in need. In schemes in Brazil and Bangla- fore floods caused supplies to decline. Public dis-
desh targeting both by geography and by self- tribution from stocks or imports can stabilize prices
reported income supplements the criterion of a and avert potentially disastrous drops in the pur-
household's containing a "vulnerable" member. chasing power of the poor. In 1984 the Kenyan
Self-selection can be introduced through the government responded swiftly to early signs of a
choice of food available or through a requirement shortfall by organizing commercial food imports.
to attend public clinics that only provide basic ser- These arrived three months before food aid did,
vices. In Chile changing milk distribution from the just as domestic stocks of maize were exhausted.
workplace to public clinics greatly improved the Zimbabwe relied heavily on its own stocks. Bot-
effectiveness of the program. Individual nutri- swana and Cape Verde made use of food aid; Cape
tional monitoring also has a role. In several pro- Verde sold the food on the open market to finance
grams, including the World Bank-assisted Tamil public employment programs.
Nadu Integrated Nutrition Program in India and Food security need not imply distribution of
the Iringa Nutrition Program in Tanzania, exten- food through public channels. Private trade can
sive monitoring of children, carried out within the often help in stabilizing prices and distributing
community, identified the needy and led to height- supplies. Holding large food stocks can be more
ened awareness of nutritional problems. costly than relying on external tradealthough
In Tamil Nadu targeted food supplements are this depends on a country's foreign exchange posi-
combined with a range of nutrition and health tion and on how quickly trade can respond in an
measures. A 53 percent reduction in severe malnu- emergency. Buffer stocks in key locations to bridge
trition was achieved within the first two years. The the gap before imports arrive may be the best op-
program in Tanzania's Iringa Province emphasizes tion. In the longer term better internal food mar-
96
Box 6.3 Drought relief in Africa: food interventions in Botswana
Like many other African countries, Botswana had to than predrought levels (see Box figure 6.3). There were
cope with episodes of severe drought in the 1980s. Un- no deaths from starvation.
like some other countries (for example, Ethiopia, Mo-
zambique, and Sudan), it has succeeded, through ap-
propriate policies, in avoiding the worst effects of
Box figure 6.3 Child malnutrition and crop production
famine. in Botswana, 1980 to 1986
During the 1979-80 drought, relief in Botswana relied
almost entirely on transporting food aid into famine-
affected areas and distributing it to the destitute. Be- Production of maize
cause of logistic difficulties, this approach was unsuc- Percentage of children and sorghum
cessful. Beginning in 1980 Botswana implemented a underweight (thousands of tons)
broad, integrated program for drought relief and recov- 35 60
ery that combines food supply management, employ-
ment generation, and agricultural assistance. 50
Effective food supply management lies at the core of
the program. Botswana received large amounts of food 30
aid in 1982-87, but its relief efforts did not depend on 40
the timely arrival of these supplies. Large-scale imports
of food from South Africa were combined with support
for rural incomes through public works and grants. 25 30
Private traders and retailers were used to distribute
food. Food prices remained broadly uniform across the
country during the drought period. 20
"Take-home" rations were distributed to all house- 20
holds that contained children under 10, pregnant and
lactating women, and destitutes. The government set 10
up a special trucking operation to supply maize to re-
mote areas. Between one-third and two-thirds of the 0
15
population received free rations during the drought.
1980 1981 1982 1983 1984 1985 1986
Intensive feeding was provided for malnourished chil-
dren at health facilities, and a feeding program for pri-
mary school children was maintained throughout. Note: The percentage of children underweight, defined here as the
percentage of children age 0- 59 months below 50 percent of expected
Health measures, including efforts to provide a clean weight-for-age, is used as an indicator of malnutrition.
The reporting system for underweight children changed at the end of
and dependable supply of water, were also important. 1984. As a result, the subsequent drop in malnutrition may be slightly
Despite large decreases in domestic food production, overaccentuated.
Source: Morgan forthcoming.
the program was effective: the percentage of children
who were undernourished had fallen by 1986 to less
kets and complementary rural infrastructure Public employment schemes
should reduce the need for government interven-
tion. Rural public employment schemes have two func-
Direct feeding and food transfer programs are tions. The first is to maintain and create rural
necessary during crises. Research in Africa and In- infrastructureexamples are road-building, irriga-
dia shows that such measures are especially effec- tion, soil conservation, and afforestation pro-
tive when used in conjunction with efforts to re- grams. The second, and the more important in this
store the purchasing power of the poor. Cash context, is to reduce poverty by providing employ-
transfers are often more effective than food ra- ment to those in need. In this vital task, public
tions: cash is faster to move and easier to adminis- employment schemes are often cost-effective.
ter, and it does little or no harm to producers and Since poor people are willing to work for low
hence to future food security. But for this approach wages, public employment programs can offer
to work food must be available elsewhere in the wages that screen out the nonpoor so that re-
affected country, and trade and markets have to sources can be used more effectively. In two large
work reasonably well (Box 6.3). rural schemes in South Asiathe Maharashtra
97
employment were provided, representing 3 per-
Figure 6.1 Seasonality and workers employed cent of total rural employment. This has undoubt-
under the Employment Guarantee Scheme, edly displaced some alternative employment, but
Maharashtra, India the rural unemployment rate in Maharashtra has
declined markedly in relation to that of other
Thousands of workers
states. There is evidence that the proportion of
poor has also declined faster than would have
1,000
been expected on the basis of initial conditions and
the growth of output.
900 Are these schemes more cost-effective than food-
pricing and food-distribution policies? They ap-
800 pear to target the poor well, but it is important to
examine the costs incurred by those who partici-
pate. If the participants join a public employment
700
scheme, they are likely to lose some of their pre-
vious income. Although such losses are difficult to
600 measure, a study for Bangladesh concluded that
the forgone earnings of participants in the Food for
500 Work Programme were equivalent to about one-
third of their gross earnings from the program. Net
400 income gains to the poorand especially the
poorestappear to be a sizable proportion of the
governments' wage disbursements.
300
Since these schemes usually produce economic
S OND J FM AM J JA assets such as roads, they may have second-round
effects on the incomes of both the poor and the
Month
nonpoor. It is claimed that in Maharashtra a dis-
E Busy season in agriculture proportionate share of these extra benefits has
Source: Subbarao forthcoming. gone to the rural rich and that this has been an
important factor in achieving political support for
the EGS. Other second-round effects are possible.
The EGS may bid up wages in general (although it
has tried to avoid recruiting during peak periods of
Employment Guarantee Scheme (EGS) in India agricultural demand). Rural public employment
and the Food for Work Programme in Bang- schemes may help labor markets to work better by
ladeshthe proportion of participants with in- reducing wage differentials between the sexes and
comes below the poverty line was at least 90 per- among ethnic groups.
cent in the early 1980s. The schemes have attracted Since the EGS is partly intended to absorb eco-
people who are often excluded from other nomic shocks, its employment has been highly
programsfor example, women and members of seasonal and has varied from one year to the
scheduled castes in the case of the Maharashtra nextalthough this variability has declined as the
scheme. Many countries in Latin America and scheme has grown and the wage rate has risen
Sub-Saharan Africa run similar programs. (Figure 6.1). By providing work, and hence pur-
The EGS began in the early 1970s. Its objective is chasing power, when it is most needed, the EGS
to provide on request employment at a stipulated has not merely raised average earnings over the
wage, within fifteen days, and no more than five course of the year but has also reduced fluctua-
kilometers away from the participants' home vil- tions from week to week and month to month (Box
lages. An unemployment allowance is paid when 6.4). A study of landless households in a drought-
this is not possible. About three-quarters of the prone region showed that families in vifiages with
EGS budget is spent on wages. The current access to the EGS had incomes that were much
scheme provides guaranteed employment year- less variable than in other villages. The same study
round. Political commitment to the program is examined other income-stabilization policies such
firm, and the effect on employment appears to be as crop insurance and found them less effective.
significant. In 1984-85, 180 mfflion person-days of Reducing fluctuations in income can matter as
98
Box 6.4 At a rural public employment site in India...
It is May 1986 in a rural area of Maharashtra State. Lata Lata seeks EGS work at this time in most years. Her
is one of a group of fifteen women working on a soil husband does so less often, as he can usually find some
conservation project organized by local officials under work at a better wage rate. For Lata the EGS wage on a
the state government's Employment Guarantee typical day is better than she could hope to get in the
Scheme. On this project she is shifting soil and doing village now, even if she could find the work. On this
light digging. Lata is in her mid-twenties, married, job she has been averaging about 6 rupees a day. Lata
with two children, three and seven years old. The likes having the wage paid partly in the form of food,
youngest child is nearby, playing with other children at since her husband has more say over how cash is
the crèche that has been set up at the project site. This spent. She is also pleased that EGS piece rates are the
allows the eldest child to stay at the school in their same for men and women; the men in her village usu-
village rather than look after her younger brother, as ally get more for the same agricultural work.
she often does when both parents are working. Lata's Like all women on EGS projects, Lata is rarely al-
husband is doing agricultural work on a nearby farm. lowed to do the more strenuous and better-paid jobs,
Their village is about three miles away. The family is such as breaking rocks for digging irrigation tanks
landless, and Lata and her husband earn about 80 per- although she knows she can do much of that kind of
cent of their income from casual agricultural labor. work. Because of the complex EGS wage rate schedule,
Their total income over the past year was about 3,000 Lata's wages vary from day to day and site to site, even
rupees, which, for a family of four, is below the rural though she is working just as hard. Sometimes there
poverty line in Maharashtra. are long delays before projects start and breaks of
The region lacks irrigation, and the land is not very many days between projects. In the past year an illness
productive. As a result, there is little multiple cropping kept her out of work for ten days. If she can stay fit, by
and the family's work is highly seasonal. Employment next month she will have worked about sixty days on
is particularly scarce at this time of the year. The family EGS projects over the course of the year. That accounts
cannot afford to save much and would have a hard for about a quarter of the paid work she does. With her
time getting by without the EGS. Things will improve EGS wages the family buys shoes, clothing, and fuel as
in a month or so, when the monsoon crop is cultivated. well as extra food.
much to the poor as raising average incomes. It can crease in response to changes in incomes from
prevent acute distress and avoid the need for other sources; this can serve as an early warning of
costly forms of adjustment such as selling produc- impending crises, such as famine. Some say that
tive assets. the EGS (which offers a guarantee) has helped to
Public employment schemes can have a vital role mobilize the rural poor as a political force. In Ma-
to play in preventing famine. This has long been so harashtra the guarantee has helped to develop
in India, and such programs were used extensively backward and resource-poor areas, where the de-
in Sub-Saharan Africa in the 1980s. In Botswana mand for jobs is often highest. And it has also
(Box 6.5), Cape Verde, and Kenya, strategies based acted as a check on the power of local officials,
in part on provision of employment have saved who might otherwise extract side payments from
many thousands of lives over the past decade. workers.
Public employment projects can also soften the im- The main drawback of a guarantee is its implica-
pact of recession, as in Latin America during the tions for the budget. The government can still keep
1980s. Chapter 7 looks in more detail at how Bo- its long-term financial commitment to the program
livia, Chile, and Peru implemented their pro- within limits, provided that it can set the wage rate
grams. at an appropriate level. But this is not always pos-
Perhaps the most important question in design- sible. The Maharashtra scheme is now required to
ing public employment programs is whether to pay wages at the statutory minimum rates, which
provide an employment guarantee. A guarantee are higher than market rates. This requirement im-
improves the schemes' effectiveness as a safety net plies substantial budgetary outlays and potentially
(and may therefore attract more political support high social costs in lost output from alternative em-
from groups that do not currently gain from the ployment.
program). With a guarantee in place, the number In Maharashtra the EGS is largely financed by
of participants in the program wifi increase or de- taxes on the more prosperous groups in the urban
99
Box 6.5 Drought relief in Africa: public employment in Botswana
The food interventions under the Drought Relief Pro- than 17 percent, and the total number of households
gramme in Botswana that were discussed in Box 6.3 engaged in farming declined by less than 5 percent.
were complemented by the Labor-Based Relief Pro- The productive value of the works created was consid-
gramme (LBRP), which provided the rural poor in erable. Other measures, including distribution of free
drought-stricken areas with opportunities for earning seed and grants for destumping, clearing, and plowing
cash income on village-improvement projects. Al- arable land, facilitated the transition from relief to re-
though the program fell short of offering guaranteed covery.
employment, it did provide employment for between Taken as a whole, the Drought Relief Programme has
60,000 and 90,000 persons each year during the been successful in preventing human suffering and
drought period. Wages were set high enough to pro- preserving the productive potential of the rural econ-
vide a meaningful return yet low enough so as not to omy in the midst of a severe and protracted crisis. Al-
attract the relatively well-off. Village committees iden- though Botswana is rich in comparison with other Sub-
tified projects and selected participants, more than 80 Saharan African countries, the critical element in the
percent of whom were women. It has been estimated program's success was the government's commitment
that LBRP projects replaced almost one-third of the to protecting the poor. The Drought Relief Programme
losses in rural incomes caused by crop failures between has become part of the election platform for the ruling
1983 and 1985. party, whose constituency is found mainly in rural ar-
In addition to stabilizing incomes and generating eas. Components of the drought programfood distri-
purchasing power, the program protected rural assets: bution, for instancehave become permanent features
the poor did not have to sell cattle or other assets to of the benefits entitlement system. This means that the
survive. Furthermore, people who might have been institutions and infrastructure will already be in place
forced to migrate were able to stay. The percentage of when drought threatens again.
traditional household farms with arable land fell by less
sector. Because the EGS is seen as one way of re- costs. This has sometimes made it difficult to find
ducing the migration of rural laborers to Bombay, suitable projects. For example, the materials for a
urban groups have not opposed this method of sealed road will cost more than the rule allows, yet
raising funds. Other schemes, as in Bangladesh, unsealed roads are soon washed away. If longer-
have used foreign food aid and have paid wages in lasting assetsmany of which would help the
kind. In Cape Verde, as already noted, proceeds poorare to be created under the programs, such
from food aid sold on the open market have been restrictions may have to be relaxed. It must be kept
used to run the programs. Donor restrictions on in mind, however, that creating the asset is not the
the monetization of food aid may prevent govern- main objective: public employment schemes that
ments from taking this route and may reduce cost- seek to reduce poverty directly need to be more
effectiveness. For example, rural public employ- labor-intensive than a conventional calculation of
ment schemes financed with food aid may not economic benefits would dictate.
have access to other funds for necessary nonlabor Projects will be all the more successful if they
inputs and materials. (This was an issue in Bangla- create or maintain assets that are especially benefi-
desh in the 1980s.) Still, the role of foreign aid in cial to the poor. Rehabilitation of agricultural land,
financing these programs, particularly in periods reforestation, and erosion control all benefit the
of stress, should be developed. poor in environmentally degraded regions, both in
If public employment programs are to act as a the short and in the long term. Public employment
form of insurance, the governmentlike any schemes are commonly used in various African
insurermust be willing to pay up when things go countriesincluding Ghana, Kenya, Lesotho, Ma-
wrong. Governments are in a better position to lawi, Mozambique, Tanzania, and Zimbabweto
cover this risk than are poor households. Any build and maintain infrastructure that will improve
scheme, however, should clearly be consistent the quality of life of the poor.
with long-term budgetary requirements.
A further policy issue in the design of the Caring for those left out
schemes is their use of nonlabor inputs. The rule
followed by the EGS, for example, is that wages Some of the poor will not be adequately reached
must account for at least 60 percent of variable by the policies discussed so far. For example, con-
100
cern for the aged and infirm in developing coun- their financing does not become a burden on the
tries is increasing. This is partly because of a shift truly poor.
in demographic profilesthe population is aging A more promising approach for some countries
in large parts of the world, including Latin Amer- involves transfers based on indicators of need that
ica, China, and South Asiaand partly because of can be monitored through local health care and
a weakening of traditional family- and community- community-based systems. Preventive and basic
based social security arrangements. The aging of health care could be provided to the elderly and
the population in many developing countries will infirm poor and could be supplemented by cash or
continue. Not all old people are poor, but provi- in-kind transfers when necessary. Self-selection in
sion for the elderly poor is an increasingly impor- attending local clinics will achieve a degree of tar-
tant issue. geting. On the whole, such arrangements are rare,
Beginning around the turn of the century, the but there are some examples. In rural China, for
rise of a large industrial labor force led to the evo- instance, the community has a moral responsibility
lution of social security systems in developed for providing basic necessities to elderly people
countries. Benefits are provided to the elderly, the without relatives and to widows and orphans, and
disabled, surviving spouses, and children through the central government helps poorer areas carry
a combination of social insurance (such as pen- out this duty. China also has "houses of respect"
sions) and social assistance (for example, health for those without relatives who are not capable of
services). The revenue base is wide, with contribu- living alone. And in some Islamic countries reli-
tions from workers and employers; coverage is al- gious land and wealth taxes have been used to
most universal; and the scale of benefits has been finance assistance to those identified as needy at
significant, accounting for 15 to 25 percent of GDP the local level. In India some states administer
in most OECD countries in the late 1970s. Reduc- pension schemes for the elderly and for destitute
tions in the incidence of poverty in developed widows. In Kerala State agricultural workers over
countries, particularly among the elderly, have 60 with incomes below a specified level receive
been strongly linked to the growth of these sys- pensions. Verification is done by local community
tems. committees that include representatives of minor-
Attempts to transplant such systems to develop- ity groups. In countries in which the aged and dis-
ing countries have, however, met with little suc- abled poor are concentrated in the agricultural and
cess. Latin American countries have had the long- informal sectors, a well-developed public health
est experience with formal social security, and care system, coupled with state support for devel-
even there the programs have failed to reach many oping and fostering local and community provi-
needy elderly people and other poor individuals. sion, may be the best answer.
By and large, coverage has been limited to people
employed in the formal sector in urban areas. The role of transfers and safety nets
Chile and Costa Rica are exceptions: both have
well-organized urban and rural labor markets and This chapter began by identifying two broad
a strong political and social commitment to social groups in need of special attention: those unable to
welfare systems. participate in the growth process and those who
The best policies for protecting the elderly poor may be temporarily in danger when events take an
and others who are left out will vary from country unfavorable turn. The first group needs a system
to country. In Chile and Costa Rica universal cov- of transfers that will ensure them an adequate
erage is feasible, and other countries with similar standard of living. The second group is best served
characteristics would be well advised to adopt by a variety of safety nets.
such programs. In Eastern Europe the large size of The appropriate mix of policies will vary enor-
the wage employment sector suggests that formal mously from country to country. It will depend on
social security will have a big role in reducing pov- a host of factorseconomic, political, social, and
erty among the aged and those unable to work. cultural. There are, however, some broad guide-
But these cases are not typical of the developing lines.
world. In most poor countries formal social secu- Most food-based policies, including general sub-
rity would fail to reduce the worst poverty, since sidies, ration schemes, and food stamp schemes,
many among the very young and aged poor, the make severe administrative demands on the gov-
sick, and the disabled would not qualify for bene- ernment. Their potential is greatest in urban areas
fits. Where formal schemes are already in place or or in more developed areas that have adequate in-
are being introduced, it is important to ensure that frastructure. In all cases, targeting is essential for
101
cost-effectiveness. Targeting can be achieved tion of environmentally degraded lands are crucial
through the choice of commodity to be subsidized, to the reduction of poverty in Sub-Saharan Africa
by locating distribution outlets in poor neighbor- and in the poor rural areas of most other regions.
hoods, or by means of indicators. In South Asia it Public employment schemes, appropriately tai-
may be best to reform the present ration systems, lored to local conditions, are a promising way to
building on experience and practical knowledge, achieve these goals while providing employment
rather than to try an entirely new approach. opportunities and stabilizing the incomes of
If the proportion of wage and salary earners is groups at risk.
high and acute poverty is usually related to old Complementary food-based interventions will
age, disability, and unemployment, formal social be needed for those who are unable to work. Di-
security systems should be developed or their cov- rect food distribution and feeding programs face
erage widened. Such systems, which provide old logistic difficulties but are often the best way to
age pensions and unemployment, disability, attack severe malnutrition. Special operations to
health, and surviving-spouse benefits, are feasible supply staples to the poor in remote areas and
in many urban areas. other high-risk regions in times of insecurity will
But most of the poor in the developing world live continue to be essential in many countries of Sub-
in rural areas, and it is there that the need for effec- Saharan Africa, Latin America, East Asia, and
tive transfers and safety nets is greatest. For those South Asia in the near future. Feeding programs
who are able to work or who have family members and transfer schemes administered through exist-
who can work, well-designed rural public employ- ing public health systems are also the best way of
ment schemes can provide insurance against a helping the elderly poor, pregnant women, and
risky agricultural environment. The key to such children in rural areas. These programs should be
schemes' success in reaching the poor is that they integrated with other measures, including im-
offer unskilled work to anyone who wants it at a provements in health, sanitation, and education.
wage rate that only the poor find attractive. Thus, The development of infrastructure to provide
targeting is achieved through self-selection by the widespread access to basic nutrition, health, and
poor. The infrastructure built and maintained un- sanitation services should be a top priority for any
der these programs can yield high economic re- country.
turns. Links to remote regions and the rehabiita-
102
The 1980s: shocks, responses,
and the poor
Many countries experienced macroeconomic diffi- measures, can have both demand-reducing and re-
culties in the 1980s as the debt crisis and interna- structuring effects. For example, changes in public
tional recession brought structural weaknesses spending can involve both generalized cuts and
into the open. But when structural adjustment is- deliberate shifts in composition, either to squeeze
sues came to the fore, little attention was paid to demand in a particular way or to support economic
the effects on the poor. Macroeconomic issues restructuring.
seemed more pressing, and many expected that This chapter is concerned with the effects on the
there would be a rapid transition to new growth poor of demand-reducing and restructuring poli-
paths. As the decade continued, it became clear cies. Since restructuring calls for a more efficient
that macroeconomic recovery and structural use of labor, it is fully consistent with the policies
change were slow in coming. Evidence of declines advocated in this Report for the long-run reduction
in incomes and cutbacks in social services began to of poverty. In Indonesia restructuring policies
mount. Many observers called attention to the sit- mainly focused on the industrial sector, whereas in
uation, but it was UNICEF that first brought the Tanzania the emphasis was on agriculture. But in
issue into the center of the debate on the design both cases the changes will help the poor over the
and effects of adjustment. By the end of the decade medium to long term by promoting the demand
the issue had become important for all agencies, for labor.
and it is now reviewed in all adjustment programs The short-run effects of adjustment, however,
financed by the World Bank. As UNICEF advo- can create difficulties, for two reasons. First, the
cated, attention is focused both on how adjust- process of economic restructuring is often sluggish
ment policies affect the poor and on the specific and uneven. Firms and labor markets take time to
measures that can be taken to cushion the short- adjust, and in the meantime economies can suffer
term costs. higher unemployment or underemployment and
labor incomes may decline. Second, demand-
Short-run policy and household welfare reducing measures may be unavoidable, and these
are likely to hurt the consumption of the poor and
The starting point for adjustment is macroeco- the nonpoor alike. The need for cuts in public
nomic disequilibrium. Whether the causes are ex- spending can lead to a particularly sharp short-run
ternal or internal, the usual symptoms are an un- conflict with two essential parts of the strategy ad-
sustainable current account deficit, internal vocated in this Reportdelivering social services
financial problemsoften linked to high inflation and providing transfers and safety nets.
and slow growth. Adjustment has two objectives: Changes in the economy affect poor households
reducing the demand for imports and domestic through two main channels: markets, which deter-
goods to stabilize economic conditions and restruc- mine private incomes and the cost of consump-
turing the economy to reach a higher growth path. tion, and public services. Developments in agricul-
Many policies, notably exchange rate and fiscal tural and labor markets are the main determinants
103
Box 7.1 Macroeconomic adjustment, household responses, and the role of women:
the experience of an urban community in Ecuador
Studies of the impact of economywide events on center in 1988 found that 80 percent were suffering
households are surprisingly rare. One such study was from malnutrition.
based on fieldwork in a low-income section of Guaya- In this community, as in many others, women have
quil, Ecuador's largest city, between 1978 and 1988. to balance several roles. In addition to earning income,
That was a period of boom and bust in the economy, many women spent more time on community activi-
driven partly by changes in the international oil ties, notably to arrange NGO services to replace declin-
market. ing public services. This put greater pressure on their
In 1978 the prospects for Indio Guayas, a rapidly role in the familyonly rarely have men taken on more
expanding community largely made up of households household tasks. For women with young children, go-
in low-wage employment, looked good. A decade later ing out to work created even more difficulties. There is
real wages had fallen by up to one-half, the prices of evidence that more women are taking control of their
some foods had soared, and local public health and fertility; in 1988 more than 40 percent had undergone
education services, which had expanded rapidly in the surgical sterilization. Teenage daughters were often a
boom of 1978-82, had been cut. help to mothers, but at the cost of schooling and ade-
Recession and adjustment had widespread effects. quate parental attention. There was concern in the
Most men were still working full-time, but at lower community about teenage boys who were roaming the
wages. In response to falling incomes, the proportion streets or using drugs. Increased domestic violence,
of women working rose from 40 to 52 percent between often stemming from disputes over the allocation of
1978 and 1988. Many women had reentered the labor men's wages to household needs, was also reported.
force despite having young children at home, and oth- Many in this community are worse off than before.
ers devoted more hours each day to paid work. House- Most men earn less in real terms, and many have re-
hold composition was changing, with married sons or sorted to temporary migration. Women have been
daughters staying in the homes of their parents. Eating forced to supplement household incomes, adding to
habits changed: households consumed much less milk, the pressures they already faced. Children have often
fish, eggs, and fresh fruit, shifted from potatoes to plá- paid the price in poorer nutrition, impaired education,
tanos, and in some cases cut out mealsfirst supper, and loss of parental care.
then breakfast. A survey of children at a local health
of incomes; pricing and subsidy policies influence be placed in one of three broad groups. First are
the prices that households face. Public spending the countries that have made progress thanks to
on services, especially in the social sectors, can af- steady growth (as in China and India) or to the
fect both the current welfare and the human capi- effective handling of macroeconomic disturbances
tal of the poor. (as in Indonesia and Malaysia). Second are those
The poor are not passive in the face of these in which poverty has worsened because of reces-
developments: they adapt. Households draw sion; this group includes most of the severely in-
down their savings and alter their expenditures debted middle-income countries and Eastern Eu-
notably, in the direction of cheaper sources of calo- rope. And third are the low-income countries
ries. Laid-off workers move into the urban infor- (most of Sub-Saharan Africa as well as such coun-
mal sector or return to rural areas. Secondary tries as Bolivia) in which slow long-term growth
workers (often women) enter the labor force, usu- and macroeconomic crises have led already acute
ally in relatively low-return activities. And house- poverty to become worse.
holds adjust their relationships with each other. A review of these three groups suggests that the
For example, the level of gifts or remittances may mix and timing of adjustment measures are cru-
change, and families may regroup. Box 7.1 looks at cial. The best approach seems to combine two ele-
the consequences of recession and adjustment for ments: (1) swift action on certain fundamental pol-
a low-income urban community in Guayaquil, icies that are designed to provide the context for
Ecuador. future growth and (2) macroeconomic policies that
can moderate reductions in private consumption
Macroeconomic policy and the poor in the transition period. Such a mix can support
effective adjustment and a relatively favorable out-
The evidence on poverty in the 1980s outlined in come for the poor.
Chapter 3 suggests that developing countries can Swift action is essential on policies that have to
104
Box 7.2 Macroeconomic policies and the poor in Ecuador: results from a modeling analysis
Box 7.1 described the fortunes of a poor urban commu- GDP, private investment, and rural and urban poverty.
nity in Ecuador during the 1980s. That account, how- It compares a "no-shock" case with two different pol-
ever, sheds little light on macroeconomic causes of wel- icy responses to an adverse external shock. The shock
fare changes. A model is helpful for understanding is similar to the one that Ecuador actually experienced:
these causes. The model used here is designed to take a combination of a fall in the terms of trade (owing to a
account of both short-run macroeconomic develop- decline in the price of oil) and reduced access to foreign
ments and sectoral developments. It is based on a de- financing.
scription of Ecuador's socioeconomic characteristics With no external shock, there is steady progress in
and contains three urban and four rural social groups. reducing poverty. Most of this decline occurs in the
Box figure 7.2 shows alternative paths for Ecuador's rural sector, especially through rural-urban migration;
Box figure 7.2 Results of three scenarios for GDP investment, and rural and urban poverty Ecuador
Real GDP Private investment
Index Index
120 120
110
100
90
80 40
2 3 4 5 6 7 1 2 3 4 5 6 7
Rural poverty Urban poverty
Percentage of total population Percentage of total population
30 30
20 20
10 10
0 0
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Year Year
Adjustment - No adjustment No shock
Source: de Janvry. Fargeix, and Sadouet (background paper).
105
Box 7.2 (continued)
urban poverty improves much more slowly. and inflation is lower. These conditions support a rapid
The "no adjustment" case assumes an essentially recovery after the third year as the economy shifts onto
passive stance in fiscal and monetary policy but a flexi- a new growth path with a stronger orientation toward
ble exchange rate policy. Maintaining public spending tradable activities, notably in the agricultural sector. In
moderates the fall in national income in the short run the short run GDP is lower than in the "no adjust-
but leads to a real exchange rate depreciation of almost ment" case, largely because of reduced demand. But
20 percent, higher inflation, higher real interest rates GDP is higher and grows more rapidly in the medium
(because of increased domestic financing of the fiscal term.
deficit), and a 40 percent decline in private investment. The clear beneficiaries are the rural poor. Thy are
This is a pattern seen in many Latin American coun- protected in the short run, in relation to their urban
tries. Welfare losses are moderate in the short run, for counterparts, by the depreciation of the real exchange
both the urban poor (maintaining spending leads to rate (increased farm incomes partially offset the effects
only small declines in labor demand and public ser- of declining wages), and they benefit in the long run
vices) and the rural poor, who benefit from higher agri- from a growth strategy with a stronger agricultural ori-
cultural prices thanks to the depreciation. But the gains entation. After seven years rural poverty is down to its
are short-lived. The shock reduces GDP in the second initial level and is falling steadily. Urban poverty rises
and third years, and low investment delays the transi- because of the initial contraction of demand and the
tion to the new growth path. Poverty worsens steadily. associated fall in real wages. Recovery along the new
The "adjustment" case involves fiscal and monetary growth path is only a minor moderating force by the
contraction and substantial depreciation of the real ex- end of the period; in Ecuador's case this reflects the
change rate, leading to a larger short-run reduction in high initial proportion of the urban labor force in ser-
GDP than in the "no adjustment" case. A decline in vices and other nontradablessectors that grow rela-
investmentprivate investment falls by 30 percent in tively slowly. In the longer term the outcome for the
the first yearagain moderates short-run consumption urban poor would be more favorable because of urban-
losses. But the domestic fiscal and monetary situation rural migration and the expansion of urban production
is more favorable than in the "no adjustment" case, of tradable goods.
do with economic restructuring. Especially impor- that the deficit is consistent with both short- and
tant are policies that concern relative prices and long-run objectives for inflation, investment, and
the management of public sector resources, includ- savings.
ing the budget and public enterprises. Poverty Analyzing alternative packages is difficult. One
cannot be reduced in the long term without approach is to use a model that consistently incor-
broadly based growth. But some policy changes porates macroeconomic factors, markets, and the
can also help the poor in the shorter term. Ex- determinants of household welfare (Box 7.2). A
change rate depreciation and other measures de- modeling approach helps to clarify issues but can-
signed to raise agricultural prices will increase the not do full justice to the complexities of the real
incomes of some of the rural poor, especially small world. In this review we therefore rely more on the
farmers. differing experiences of the countries in the three
Macroeconomic management can soften the im- groups outlined above.
pact of adjustment on consumption in several Adjustment is conditioned by initial macroeco-
ways. The government might plan for a temporary nomic imbalances, the depth of structural prob-
"pause" in investment, for increased foreign capi- lems, and the severity of external shocks. The first
tal flows, and for temporary increases in transfers group of countries, typified by many in East Asia,
from the budget. Some of these measurespublic entered the decade with relatively stable mac-
spending on investment, for instanceare directly roeconomies, flexible production and trade struc-
subject to government control. Others, notably tures, and a history of investment in human capi-
private investment, will respond to other aspects tal. These countries suffered relatively mild
of the economic environment. When transfers are shocks, taking the decade as a whole (Table 7.1).
used, the budget deficit will be higher than other- As a result, adjustment there did not require a sus-
wise; this will be justified only if effective action is tained reduction in demand. Countries in the sec-
taken on other aspects of public finance to ensure ond group, including many in Latin America, were
106
living beyond their means at the beginning of the
decade, and they suffered larger shocks. Their ad-
justment was inextricably tied up with the reduc-
tion of demand. These countries also had a greater
-
Figure 7.1 Poverty in Indonesia and Malaysia,
1984 and 1987
need for economic restructuring than the first
Incidence of poverty (percent)
group. The third group, dominated by Sub-
30
Saharan Africa, started the 1980s with substantial
macroeconomic imbalances and severe structural
weaknesses, and it experienced the worst shocks. 25
In discussing each of the groups in more detail, it
is helpful to focus on one or two representative
countries. 20
The benefits to the poor of effective action
15
Indonesia and Malaysia represent the first group.
In Indonesia progress in reducing poverty was the
result of gains across the board; in Malaysia reduc- 10
tions in rural poverty outweighed a small rise in
urban poverty (Figure 7.1). These successes are the
result of effective adjustment. Both countries acted
swiftly on policy fundamentals while relying on
changes in investment, debt, and public spending
to prevent or moderate declines in consumption. 0
These countries needed to adjust because of the 1984 1987 1984 1987
fall in the prices of oil and other commodities. Indonesia Malaysia
Once it became clear that the shocks were not tem-
porary, both countries acted decisively to stabilize 0 Rural share 0 Urban share
their economies and establish a framework for eco- Note: The total incidence of poverty for Malaysia is slightly less
nomic restructuring. They adjusted their fiscal pol- than that reported in Chapter 3 because specific poverty lines
for rural and urban areas have been used here.
icies, depreciated their currencies, liberalized their
Table 7.1 External shocks in the 1980s, by region trade regimes where necessary, and deregulated
(percentage of GDP) their industries. These policies raised the relative
Terms of Interest price of agricultural goods (Table 7.2), which pro-
trade rate tected the incomes of farmers in the short run and
Region and country shock shock' Total
encouraged continued growth in farm output. The
Sub-Saharan Africa 10.1 4.4 14.4 gains were fully passed through in the case of ex-
East Asia (excluding China) 3.9 4.3 8.1 port crop farmers, who are a significant part of the
China 0.6 0.6 1.2 farming population in both countries. Rice farmers
South Asia (excluding India) 7.9 2.3 10.2 in Indonesia also enjoyed significant gains. Well-
India 4.6 1.0 5.6 developed rural infrastructure and marketsthe
Latin America and the fruits of the policies and investments of previous
Caribbean 6.3 4.0 10.3 decadesgreatly reinforced the benefits to
Note: Numbers may not sum to totals because of rounding. The table farmers. The governments gave a high priority to
shows changes in external conditions for 1985-88 compared with the maintaining public infrastructure, despite the need
average for 1970-80; it thus gives a measure of those shocks that
persisted into the second half of the decade. Aggregations by region for fiscal stringency.
are not weighted. Although restructuring was effective, demand-
The difference between changes in export prices and changes in
import prices between the two periods. Prices are in dollars and are reducing policies were also necessary. But these
weighted by the share of exports or imports in GDP. policies had only a moderate impact on private
Calculated from changes in the real interest rate weighted by the
debt-to-GDP ratio. The real interest rate is derived from the implicit consumption. Investment was high at the begin-
nominal interest rate (the sum of public and private interest pay- ning of the adjustment, following the increase in
ments less interest receipts on reserves divided by total debt) and
from U.S. inflation. revenues from oil exports in the early 1980s, and
107
Table 7.2 Changes in key variables, Indonesia
and Malaysia, 1984 and 1987
(1984 = 100, unless otherwise indicated)
Indonesia Malaysia
Indicator 1984 1987 1984 1987 Box 7.3 The poor in Eastern Europe
GDP per capita 100 107 100 98
in the years of crisis
Private consumption per
capita 100 107 100 85
Fixed investment as a The evolution of poverty in Eastern Europe in the 1980s
percentage of GDP 22 19 32 23 was dominated by macroeconomic developments. As
in many highly indebted countries, structural defects
Real effective exchange were masked by overborrowing in the 1970s. Box fig-
rate' 100 55 100 76 ure 7.3 shows the pattern of changes in poverty in
Agricultural terms of trade 100 116 100 129 Poland and Yugoslavia. Whereas Poland was, until re-
Urban wage 100 115 100 97 cently, a centrally planned economy, Yugoslavia has
Rural wage 100 115 100 99 followed decentralized socialist policies since the
a. In terms of foreign currency per unit of domestic currency. 1950s.
Demand-reducing measures, coupled with halting
attempts to reduce subsidies, led to declines in real
urban wages in both countries and to increased unem-
there was room for a "pause" in investment with-
ployment in Yugoslavia. This was in spite of substantial
out a decline to unusually low levels (Figure 7.2). declines in investment, which in turn helped protect
Both countries cut public investment substantially, overall private consumption. Farm incomes fell less
especially in capital- and import-intensive activi- sharply because of exchange rate changes (notably in
ties. Private investment declined in response to the Yugoslavia) and because farm households have greater
contraction in economic activity, tighter monetary access to the "second" economy that operates outside
conditions, and greater uncertainty. The convinc- official markets. Urban poverty increased substantially.
Although reform was already under way in some
ing character of the adjustment policies provided Eastern European countries in the 1980s, much more
the basis for a recovery in private investment, radical measures are being implemented or are under
consideration in the 1990s. These steps are likely to put
added pressure on urban labor. A substantial shakeout
of employment from the state sector will be necessary;
the private sectors of these economies, although grow-
Figure 7.2 Fixed investment in Indonesia and ing fast, are still very small. Subsidies are a major prob-
Malaysia, 1979 to 1988 lem; in 1988 they were 14 percent of GDP in Poland, 12
percent in Hungary, and 9 percent in Yugoslavia. The
task is clearly immense. Even so, the principle of effec-
Ratio of fixed investment to GDP tive and early action on policy fundamentals, together
40 with measures to smooth consumption, applies here
too.
30
which was already beginning in 1988. Despite
lower receipts from the oil sector, the governments
were able to contain the budget deficits through
reduced public investment, greater stringency in
current spending, and increased revenues (nota-
bly, in Indonesia, from higher domestic fuel
prices). In Malaysia overall fiscal control proved
consistent with holding the consumer price of rice
steady in nominal terms, thus increasing the fiscal
10 transfers to rice consumers. Indonesia was able to
combine adjustment and an increase in foreign
borrowing without losing its creditworthiness.
1979 1982 1985 1988
Concessional assistance also made a valuable con-
tribution.
108
This mix of measures achieved macroeconomic
stability with only moderate losses in welfare and
(especially in Indonesia) fostered a political envi-
ronment that supported strong adjustment. Ini-
tially favorable structural conditions helped short-
run macroeconomic management as well as
restructuring. The three most important factors at
the beginning of the decade were a well-managed
Box figure 7.3 Poverty and wages in Poland
fiscal policy, high investment, and a relatively low
and Yugoslavia debt burden.
The outcome for the poor was favorable. In both
countries most of the poor are farmers, and land
Poland
Incidence of poverty
distribution was relatively even. Smallholders
(percent) Real wages (1978 = 100) were able to benefit from higher farm incomes. Ur-
30 120 ban and rural households that depended on wages
were potentially more vulnerable to the price in-
creases associated with devaluation in both coun-
tries and with the declines in national labor de-
mand as a result of recession in Malaysia. Wages
did fall in Malaysia, causing the small rise in urban
poverty seen in Figure 7.1, but in rural areas
higher incomes for farm and nonfarm enterprises
offset that effect. The decline in real wages was
also moderated by a fall in the share of profits,
especially in nontradables, and by the policy of
stabilizing rice prices. Although public wages de-
clined in Indonesia, the overall buoyancy of the
1978 79 80 81 82 83 84 85 86 87 labor market (thanks to deregulation and the rapid
growth of nontraditional exports) appears to have
Yugoslavia raised real wages in the private sector. Survey evi-
Incidence of poverty dence for Indonesia indicates that the incomes of
(percent) Real wages (1978 = 100) poor households went up for all sectors and
30 120 occupations.
The problem of sharply reduced demand
The second group consists of middle-income coun-
tries that were living beyond their means at the
beginning of the decade as a result of easy borrow-
ing in the past. There, reductions in demand were
unavoidable, and there was little scope for using
debt to smooth consumption. Because of these un-
favorable initial conditions, it was hard to avoid a
squeeze on the incomes of the poor in the wake of
the debt crisis. In many of these countries the poor
1978 1983 1984 1985 1986 1987 depend heavily on the labor market and are there-
fore vulnerable to a widespread drop in the de-
Urban share 0 Rural share Urban wage
mand for labor. The group includes much of Latin
Note: For Poland wages are adjusted for shortages of consumer goods
in 1980 and 1981.
America and parts of Eastern Europe (Box 7.3).
Source: Milanovic (background paper) and Posarac (background paper). Many are severely indebted. Brazil and Costa Rica
can serve as examples. Both experienced a sharp
increase in poverty when wages fell in 1983 (Figure
7.3). Costa Rica was able to reverse this trendby
1986 poverty was below the level of the late
109
employment stagnated. Policy had much to do
Figure 7.3 Poverty and wages in Brazil and with the subsequent outcome. Failure to reduce
Costa Rica the deficit undermined the effort to tackle infla-
tion. High domestic borrowing by the public sector
raised real interest rates and expanded the share of
Brazil financial profits in the economy (at the cost of the
Incidence of poverty
(percent) Real wages (1980 = 100)
shares of labor and of nonfinancial profits). In 1986
40 200
the Cruzado Plan sparked an economic boom that
reduced poverty substantially. But the rise in con-
sumption and wages was not sustainable, owing
to rising imports and resurging inflationary pres-
sures. When the inevitable contraction came, pov-
erty rose again. In 1987 the incidence of poverty
was higher than at the beginning of the decade.
The country faced continuing fiscal difficulties and
a larger debt burden. Brazil had undergone the
recession of 1983 for nothing.
Brazil's policies led to high inflation and
(through widespread import controls) an appreci-
ated real exchange rate, in relation to what it
1980 983 1985 would otherwise have been. Each of these effects
had complex implications for poverty. In general,
Costa Rica however, they made the poor worse off. The "in-
Incidence of poverty flation tax" is probably much more regressive than
(percent) Real wages (1978 = 100) traditional tax instruments. And greater deprecia-
60 120 tion would have helped the rural pooralthough
to a smaller extent than in Malaysia, because more
of Brazil's rural workers depend on wages. Evi-
dence from the Philippines, another middle-
40 80 income country that failed to adjust adequately in
Table 7.3 Changes in key variables, Brazil,
20 40 1980 to 1987
(1980 = 100, unless otherwise indicated)
Indicator 1980 1981 1983 1986 1987
GDP per capita 100 92 87 99 100
0 0 Private consumption
1978 1983 1986 per capita' 100 92 89 99 96
Urban share 0 Rural share Fixed investment as a
Formal sector earnings - Informal sector earnings
percentage of GDP
Public sector deficit as a
23 23 18 19 22
Manufacturing wage Agricultural wage percentage of GDPt 3.6 6.2 4.2 3.6 5.5
Annual inflation
(percent) 90 108 141 144 209
1970sbut Brazil's recovery was incomplete and
Real effective
faltering. The difference can be linked to action on exchange rater 100 92 118 108 100
policy fundamentals. Formal sector earnings 100 115 98 135 95
Brazil failed to act decisively on the fundamen- Informal sector
tals: its public sector deficit (adjusted for inflation) earningsd 32 37 30 53 35
remained high, and its real exchange rate did not Adjusted with the use of the GDP deflator.
The public sector deficit is the operational deficit after deducting
change significantly (Table 7.3). The recession of the component of nominal interest payments that can be attributed
1983 was probably unavoidable. Poverty increased to inflation.
In terms of foreign currency per unit of domestic currency.
because of falling wages and incomes in urban and The index of informal sector earnings is in relation to the 1980
rural areas; informal employment grew as formal formal sector wage.
110
Box 7.4 Macroeconomic policy choices and income distribution in the Philippines
The recession of 1983-85 in the Philippines illustrates Why is this? Recession and contraction in the labor
the costs to the poor of declining labor demand, lack of market cause declining labor productivity and rising
real exchange rate movement, and inflation. As in underemployment, which hit the poor hard. Accelerat-
many highly indebted countries, the crisis had its roots ing inflation also hurts the poor because nominal
in past overborrowing and structural defects. Stabiliza- wages fail to keep pace and because the poor lack ac-
tion began with a nominal devaluation and strict im- cess to assets that are protected from inflation. The
port controls, followed by drastic cuts in public spend- poor gain from real exchange rate depreciation because
ing and a tight monetary policy. The effects were exports are intensive in unskilled labor (and are often
initially stagflationary: a sharp decline in output, faster produced by peasants) and because tradables account
inflation, and an appreciating real exchange rate. The for only a small part of the consumption basket of most
currency did not depreciate in real terms until 1986. poor people in the Philippines. Government spending
How did all this affect the poor? An econometric mainly benefits the middle classes, and cuts here hurt
study that used quarterly data for 1980-86 found that the poor less.
contraction in the labor market and increases in infla- The country's approach to stabilization was unfavor-
tion reduce still further the share of the poor in national able to the poor in this period. If the government had
income; depreciation of the real exchange rate works in chosen not to ration imports, relative prices would
the opposite direction. The results indicate that when have changed more sharply and without such a large
the inflation rate increases by 10 percent, the share of rise in inflation. Demand reduction through changes in
the poorest fifth of the population falls by 10 percent fiscal and monetary policy was necessary, but greater
(other things being equal). By contrast, when the real reliance on changes in relative prices would have mod-
exchange rate depreciates by 10 percent, the share of erated the recession and protected the poor. Changes
the poorest fifth rises by 20 percent. The study also in the composition of public expenditures, although
found that lower public spending and higher real inter- politically difficult, could also have made adjustment
est rates lead to smaller decreases in the incomes of the less painful for the most vulnerable.
poor than of the rich.
the early 1980s, supports the view that real depre-
ciation and lower inflation help the poor (Box 7.4). Figure 7.4 Investment pause versus
In Brazil and the Philippines alike, high inflation investment crash
and overvalued exchange rates, combined with
general uncertainty, also encouraged capital flight, Ratio of fixed
which benefited the rich and hurt the poor. investment to GOP
Costa Rica stands in sharp contrast to Brazil. It 30
suffered a deeper recession at the beginning of the
period but undertook a substantial macroeconomic
adjustment. Between 1980 and 1986-87 the gov-
ernment cut its budget deficit from 8 percent of
GDP to 2-3 percent, and the real effective ex-
change rate depreciated by 30 percent. By 1986
output and wages had recovered and inflation had
declined. These changes account for the significant
improvement in poverty seen in Figure 7.3.
The success or failure of efforts to protect the
poor does not depend on whether the government
attempts to cushion the decline in consumption
during adjustment. The successful and the unsuc-
cessful alike have taken this route. Indeed, in both
Latin America and East Asia the counterpart of a
cushioned fall in consumption was lower invest-
ment (Figure 7.4). But Latin America's investment
1981 1983 1985 1987
rate is now 27 percent lower than in the early
1980sand far below East Asia's. This illustrates
111
decrease in the incomes of the poor was probably
inevitable, even with significant declines in invest-
Figure 7.5 Consumption and agricultural
purchasing power, Sub-Saharan Africa ment. As the model described in Box 7.2 illus-
and Ghana trated, governments can maintain spending above
sustainable levels only temporarily and at great
cost. But with appropriate policies, governments
Private consumption per capita can minimize the decline. Demand-reducing mea-
Index (1981 = 100)
sures that moderate inflation and lead to a more
120
competitive real exchange rate hurt the poor less
than the alternatives. This is even more the case if
110 the measures are resolute enough to maintain do-
mestic economic stability and thus prevent capital
100 flight. In the 1990s avoiding a further squeeze on
living standards will require an increased inflow of
90 capital; where policy reform is on course, this will
be warranted. Beyond the short term, however,
80 the only way to help the poor is to bring about
sustainable recovery based on a growth path that
70
involves efficient use of labor and widespread in-
vestment in human capital.
1981 1982 1983 1984 1985 1986 1987 1988
The problem of severe distortions
Agricultural purchasing power per capita Sub-Saharan Africa dominates the group of low-
Index (1981 = 100) income countries with severe distortions. These
120 distortions were exacerbated by the macroeco-
nomic shocks of the 1980s, and the region suffered
110 badly. By the end of the decade many countries
had initiated significant reform programs. As in
100
the other groups, balancing the need to adjust
with the need to protect the poor called for a mix-
ture of (in some cases radical) changes in funda-
90
mental policy and measures to cushion consump-
tion. Ghana provides an illustration. Like many
80
Table 7.4 Changes in key variables, Ghana,
70 1980 to 1988
1981 1982 1983 1984 1985 1986 1987 1988 (percent, unless otherwise indicated)
- Ghana Indicator 1980-84 1985-88
Sub-Saharan Africa Average growth rate
GDP per capita 4.1 1.7
Note: Private consumption per capita is calculated on the basis of Private consumption per capita 3.4 1.6
the total population; agricultural purchasing power is calculated Agricultural purchasing power
on the basis of the rural population.
per capitaa 10.3 106
S ha re of GDP
Fixed investment 5.0 99b
Fiscal deficit 4.2 0.5
the limits of a pause in investment as a way of Foreign aid receipts 0.8 3.8
cushioning consumption: a pause makes sense
Average value
only if it is followed by an investment recovery Real effective exchange rate' 172 32
that prepares the way for future growth in Black market premium (percent)' 597 62
outputas in Costa Rica and Malaysia. Otherwise This is the growth in nominal value added in agriculture per
a country may start with a smaller reduction in capita rural population, deflated by the consumer price index as a
consumption but end with a permanently lower rough measure of terms of trade gains for farmers.
1985-87.
level. Based on the official exchange with 1980 = 100 and expressed in
terms of foreign currency per unit of domestic currency.
A substantial reduction in demand was unavoid- The percentage of the black market exchange rate above the offi-
able in Brazil and Costa Rica. As a result, some cial ratethis had declined to 4 percent by 1988.
112
countries in this group, its economy has suffered The changes in official prices in Ghana were sim-
from serious distortions, a long-term decline in its ilar to those made by many countries in the late
infrastructure and institutions, and adverse exter- 1980s. The alternative, which was commonly fol-
nal shocks. Its experience since the mid-1980s, lowed in the early 1980s, was to avoid changes in
however, illustrates the benefits of effective policy nominal exchange rates by rationing imports. This
(Figure 7.5). led to appreciation of real exchange rates and ex-
In Ghana movement on the policy fundamentals pansion of parallel markets for foreign exchange.
meant large rises in many official prices. These in- Attempts to control domestic prices usually led to
creases were needed to deal with the distortions the dominance of parallel product markets, in
between official and parallel markets and between which prices were more strongly influenced by the
agriculture and the rest of the economy (Table 7.4). black market exchange rate than by the official
They were supported by strong fiscal adjustment, rate. This happened, for example, in prereform
with an effort to redirect public spending toward Ghana and in Tanzania. And in countries in which
the poor. This change in priorities allowed for the general rationing and price controls were com-
rehabilitation and expansion of some social ser- pounded by other sources of acute instability (such
vices. Ghana's approach, in other words, was en- as war, as in Angola and Mozambique) rural mar-
tirely consistent with the development strategy ad- kets disintegrated. Farmers were forced to with-
vocated in this Report. Moderating consumption draw into subsistence production.
losses by reducing fixed investment was not an In assessing the impact of adjustment on pov-
option. Investment was already depressed in the erty, the main concern is with rural incomes. Rural
mid-1980s, and it has yet to recover to a level that areas accounted for 80 percent or more of those in
would be sufficient merely to replace worn-out poverty in countries such as Ghana in the 1980s. A
capital. Foreign aid was therefore an essential ele- strategy of rationing imports and reducing official
ment in supporting Ghana's recovery in consump- producer prices clearly hurts rural areas. But how
tion per capita, which grew at 1.6 percent a year in have the radical changes adopted in Ghana and
1985-88. Continued aid will be necessary if invest- other countries influenced the lives of the rural
ment is to recover and consumption is to continue poor? The answer is that many have been helped,
to grow in the 1990s. but some have been hurt (Box 7.5).
Box 7.5 The effect of macroeconomic adjustment on farmers in Sub-Saharan Africa
Regina Ofo is a farmer in Bendel State in Nigeria. For respond to higher prices by increasing food production
her, the structural adjustment program of 1986, which because he lives in Imo State, where land is scarce.
included measures to increase farm prices, has meant Malawi's reforms, which date back to 1981, have in-
something concrete: she is better off. Because she earns cluded substantial price increases for maize. These in-
more from farming, she can afford to buy new clothes creases have both helped and hurt small-scale farmers.
for herself and her two daughters, and she was even Maize farmersmany of whom are poortripled their
able to prepare a Christmas feast for the less fortunate output in the early 1980s and have helped Malawi re-
in her village. Babatunde Akinola, a former municipal vive its export trade. But some lost out in the short run.
employee, has also benefited. As a result of the scrap- Dester Mlondo is unable to produce enough maize to
ping of the government-run Cocoa Marketing Board in feed her family year-round. She has to sell her maize at
1986 and the subsequent currency devaluations, inde- harvest time, when prices are low, to obtain cash for
pendent merchants are offering prices for cocoa that other household needs. The removal of maize subsi-
are many times higher than prices in the early 1980s. dies in 1987 has meant that maize becomes expensive
Mr. Akinola spent his pension rehabilitating his fa- in the months before harvest, when Mrs. Mlondo must
ther's cocoa farm in Ondo State. His income has risen buy maize to feed her family.
substantially. Since 1986 he has built a large house, and Adjustment measures that favor agriculture will
he is able to send his children to some of the best gradually pull up the whole rural economy. In the
schools in Nigeria. shorter term they have helped many farmers but hurt
Others have been less fortunate. Rising food prices others. Even the losers might soon have been worse off
are not always welcome to farmers who have to buy without the reformsbut that is little comfort to people
food part of the time. Before the change in policy, Nse such as Mr. Nnachukwu and Mrs. Mlondo. Measures
Nnachukwu was able to feed his family of six with to protect the most vulnerable during adjustment are
produce from his plot and his proceeds from petty discussed further in the section on "Public expenditure
trading. But higher prices for food and other basic restructuring and the poor."
goods have outrun his income. Furthermore, he cannot
113
that are little affected by international price and
Figure 7.6 Agricultural terms of trade, Nigeria exchange rate changes, and they have suffered
and Tanzania from the contraction in economic activity; this was
important in Nigeria. Finally, some of the poor in
rural areasin Malawi, for exampleare net pur-
Nigeria chasers of food. These mixed short-run effects do
Index (1978 = 100)
not diminish the case for strong price incentives in
350
agriculture. But they underline one of the central
300 messages of Chapter 4: a coherent overall strategy
that includes improved marketing and rural infra-
250
structure is needed to support growth in rural in-
200 comes.
Rural areas tend to gain from adjustment, espe-
150
cially in the medium term. Indirect evidence sug-
100 gests that it is the urban poor who are more at risk.
Real urban incomes may fallbut not always be-
50 cause of higher prices, as is often supposed. In
0 many cases poor urban consumers are paying
1978 1980 1982
parallel-market prices rather than official prices
1984 1986 1988
when adjustment begins, and higher official prices
Real effective exchange rate a often have neutral, and perhaps favorable, effects
Agricultural terms of trade b on the prices actually paid by the urban poor. The
Tanzania (export crops)
main reason for decreases in urban income is the
Index contraction in the demand for urban labor. In
200 many countries this was unavoidable. Urban
spending, especially in the public sector, reached
levels in the 1970s that were clearly unsustainable.
150 Even before the adoption of reform measures, both
Ghana and Nigeria experienced a radical decline in
the position of urban labor. Urban-rural migration
100
has become significant in both countriesthree-
fifths of all internal migrants in Ghana during
50 1982-87 came from Accra, the capital.
In this third group of countries the problems of
adjustment and poverty are at their most severe.
0 But, as before, the appropriate path is to combine
1977/78 1979/80 1981/82 1983/84 1985/86 1987/88 effective action on policy fundamentals with as
a much support for private consumption as macro-
Real effechve exchange rate (1978 = 100)
economic resources allow. Two special factors de-
Prices received by marketing boards
Prices received by producers
serve to be stressed regarding this group. First,
price distortions are in many cases still so great
In terms of domestic currency per unit of foreign currency. that dramatic price reforms may be either neutral
The value added deflator divided by the consumer price index.
or beneficial for the poor, who in many cases have
gained little from attempts to control official prices.
Second, usually the only short-run option for
Price reform helps the rural poor by raising their moderating consumption declines is foreign aid.
incomes as producers. Sometimes, however, even Given the depth of the structural problems in
large changes in the official exchange rate and in many countries, a lengthy transition during which
prices have failed to have much effect on poverty. aid finances consumption as well as a recovery in
Figure 7.6 illustrates the experience of Nigeria and investment will probably be necessary.
Tanzania. Potential gains from depreciation of the
Macroeconomic policy choices and political feasibility
exchange rate have not always been passed on;
sometimes, notably in Tanzania, they have been Protecting the poor through macroeconomic policy
absorbed by inefficient marketing chains. More- has proved easier in some countries than in others.
over, many of the poorest farmers produce crops The East Asian countries were in a better position
114
Box 7.6 The political economy of adjustment
Effective adjustment can ease the burden on the poor 1986, Turkey in 1983, Costa Rica in the mid-1980s, and
in the short run and reduce poverty in the long run. Yet so on. In these cases, leaders built on discontent with
adjustment poses dilemmas for political leaders. The previous forms of economic management and de-
policies often impose costs on constituentssometimes fended market-oriented policies as "progressive." Un-
permanent costs, as when monopoly profits or trans- der these circumstances the fit between the politics and
fers are cut. They also involve a tradeoff between economics of effective adjustment can be close. Where
present and futurea tradeoff that politicians, with new economic teams have come to power and the poli-
elections to win, interest groups to satisfy, and coups cies of the previous team have been discredited, there
to deflect, find awkward. The success of adjustment is scope for ambitious reform. In such cases demand-
measures may depend on the feasibility of building reducing and restructuring measures should be intro-
coalitions of those who benefit and on careful sequenc- duced as quickly as is technically and economically fea-
ing with respect to political as well as economic sible. Swift action establishes the credibility of the
objectives. program, limits the opportunity for resistance to coa-
Favorable initial conditions, which many East Asian lesce, and increases the likelihood that new leaders will
countries enjoy, reduce the political as well as the eco- reap the political fruits of reform. These lessons are
nomic costs of adjustment. The openness of the Malay- particularly relevant for the new Eastern European de-
sian economy, Korea's strong export orientation, and mocracies.
Indonesia's long-standing stress on rural development A mix of swift action on the fundamentals and efforts
created important constituencies that quickly benefited to cushion consumption is often justified on welfare
from adjustment. This reduced the risks for political grounds. The political case for this approach can also
leaders and contributed to a virtuous circle of sound be strong. But there is an important difference between
policy and satisfactory economic performance. the two rationales. Concern with welfare puts the em-
Severe imbalances and large external shocks increase phasis on the poor, whereas a preoccupation with poli-
the temptation to delay adjustment. If people think, as tics points the other way, dictating that losses among
they did in Zambia, that economic difficulties are exter- politically powerful groups (such as formal sector
nally caused and adjustment policies externally im- workers) be moderated. In designing programs to com-
posed, there is a high risk of losing political support. pensate losers among the poor, some leakage to such
Weak or divided governments and those facing elec- groups may be both economically unavoidable (if the
toral challenges are likely to procrastinate. The govern- truly disadvantaged are to be reached) and politically
ments of Costa Rica and Ghana in the late 1970s and advisable.
early 1980s, Brazil in 1986, and the Philippines in 1984 As painful as they may be, crises can strengthen sup-
all delayed because of political uncertainties. This in- port for policy change, weaken antireform interest
creased the costs of adjustment. groups, and increase politicians' willingness to rely on
Some have argued that authoritarian regimes are bet- technocrats. For those governments that are willing to
ter at effecting adjustment, since they have no (parlia- act, the political risks of undertaking difficult measures
mentary) opposition demanding compromise and they can be relieved by outside support. When govern-
have the luxury of longer time horizons. The govern- ments lack the resources to cushion consumption, ex-
ments of Chile after 1973, Korea in the early 1980s, and ternal capital inflows can play a vital role in the political
Ghana from the mid-1980s are examples. But there are sustainability of reform. Adjustment in both Ghana
many cases of strong reform under democratic and Turkey was facilitated by timely outside support
accountabilityJamaica in the 1980s, the Philippines in for committed governments.
I
at the beginning of the decade, they reacted swiftly choice of policies for dealing with macroeconomic
to the shocks by taking macroeconomic and re- disequilibrium (Box 7.6). Conflicts between politi-
structuring measures, and they used a pause in cal imperatives and the goal of reducing poverty
investment appropriately to moderate declines in are common. This is especially true when demand-
consumption during the transition. Their econo- reducing measures are required or when growth-
mies responded powerfully, and the poor were rel- promoting policies help the rural poor but hurt the
atively well protected, notably through changes in towns. If experience is a guide, strong executives
relative prices. A weaker initial position undoubt- with a mandate for change seem best placed to
edly contributed to the plight of the poor in other pursue reformand, given the opportunity, it
areas. But policy was of critical importance, as the makes sense to act quickly. In any case, external
progress achieved in Costa Rica and Ghana capital flows and measures to compensate losers
testifies. play a crucial political role as well as a purely eco-
Political factors have a decisive influence on the nomic one.
115
Public expenditure restructuring and the poor Preparing the way for a recovery in investment.
Although a decrease in public investment is an ap-
Macroeconomic and structural adjustment policies propriate initial response to a shock, a subsequent
can evidently have a significant impact on the recovery is essential.
poor. In many countries reducing poverty has
There are clearly tradeoffs here, and some of
been facilitated by good macroeconomic policy
them become sharper in a macroeconomic crisis.
choices. But many of the poor have suffered and Public spending to cushion declines in consump-
continue to suffer in the wake of the shocks of the
tion becomes all the more important, but so does
1980s. Even if policy promotes the expansion of
maintaining and expanding the capital stock. Ac-
labor demand by reducing biases against labor-
curate targeting is especially valuable in the face of
intensive activities, economies take time to re-
these difficulties, but by definition, it entails cuts
spond, and some groups of laborers or small-scale
that hurt the rich or middle classes, making it polit-
producers will lose during the transition. Also,
ically as well as administratively awkward. Never-
dealing with shocks and excessive public debt of-
ten involves cuts in public spending, which may
theless, countries such as Chile and Indonesia
have managed to alter their public spending priori-
hurt the poor directly. Table 7.5 shows how reduc-
ties as part of an overall fiscal adjustment.
tions in aggregate spending affected spending on
This section focuses on short-run changes in
social services in Sub-Saharan Africa and Latin
budgetary expenditures. But changes in govern-
America in the early 1980s. Social spending fell in
both regionsbut in Sub-Saharan Africa changes
ment receipts can also be important. In many
countries there is a case for reforming the structure
in allocation in favor of the social sectors provided
of revenues in a manner that both serves long-run
a little protection, whereas in Latin America there
were actually shifts in composition away from so-
efficiency and is broadly progressive; well-
designed direct and indirect tax structures and
cial spending.
user charges for utilities, energy products, and so-
With overall public spending under pressure, cial services have these characteristics. Where such
how far is it possible to protect the poor? There are
reforms can be introduced swiftly, increased reve-
three main elements in that task. nues can reduce the need for spending cuts. A
Cushioning consumption. The poor lack savings good example is reduced subsidies on petroleum
and access to credit, and they are ill-equipped to products in Indonesia. Often, however, reform of
protect their consumption on their own. the tax system takes longer.
Maintaining physical and human capital. Adjust-
ment may mean a long-term setback for the poor if Restructuring public expenditure during adjustment
it undermines the resources that enable them to Many countries have cut spending across the
work as small-scale producers and laborers. Health
board. The Philippines in the mid-1980s is an ex-
and education, irrigation works, and rural roads ample. The share of spending on the social sectors,
are some of the investments that should not be
agriculture, and poverty-oriented programs re-
allowed to deteriorate.
mained roughly constant at about 30 percent of
noninterest spending, but because of the overall
Table 7.5 Fiscal contraction and social spending, cutbacks real per capita spending in these areas in
Sub-Saharan Africa and Latin America 1986 was about two-thirds of the 1981 level. Since
overall public spending in the Philippines benefits
Change
Region and indicator 1980 1985 (percent) the nonpoor disproportionately, they also suffered
Real expenditure per capita (1978 = 100)
substantial losses from the cuts. But it is the poor
Sub-Saharan Africa who are most vulnerable, and the pattern of
Total noninterest spending 96 64 33 spending did not change to protect them.
Social spending 85 63 26 Indonesia also had to cut its public spending af-
Latin America ter oil prices declined in the early 1980s. Between
Total noninterest spending 110 92 16 1982-83 and 1987-88 its real public spending fell by
Social spending 107 87 18
17 percent. But despite an 80 percent increase in
Social spending as a percentage of total noninterest spending
Sub-Saharan Africa 23 26 13
interest payments, spending to maintain economic
Latin America 23 20 16 and social infrastructure was protected, and trans-
Note: The figures include all countries in which total real noninterest fers to the provinces increased by 29 percent. The
spending declined during the period. reallocation was feasible because of a substantial
116
reduction in development spending and a real de-
cline in the wage bill. Within development spend- Figure 7.7 Who gained from public spending on
ing, resources shifted from import-intensive in- education? Chile, 1974 to 1986
dustry and mining to human resources. The net
effect was to increase the employment content of
public spending and preserve adequate resources Percentage of educational expenditure
for maintaining and selectively expanding social 50
and economic infrastructure. The Indonesian gov-
ernment has traditionally avoided explicit trans-
fers, but shifts of public spending into activities
with a high employment content helped to main-
tain the incomes and consumption of the poor.
The pattern of spending within sectors is as im-
portant as the pattern among sectors. It is often 30
argued that when social spending is cut, services
for the poor are cut most. The evidence is inconclu-
sive. In some African countries primary education
and health services suffered disproportionately, 20
and the quality of service declined. Cuts tend to
bear hardest on supplies and equipment; person-
nel costs are harder to adjust. (In Cameroon sala-
10
ries rose to 99 percent of total recurrent health
spending in the fiscal squeeze of 1985-87.)
Chile managed to protect services to the poor
during its fiscal adjustment. Despite lower public 0
spending on goods and services overall, basic
health and child nutrition programs targeted to the 1974 1981) 1983
poor expanded. This helped to sustain a continued
improvement in social conditions in the 1980s, in- Income group
cluding further declines in mortality rates for chil- O 30 percent poorest 0 30 percent middle
dren under 5. In education, too, the government o 40 percent richest
maintained quality by ensuring an adequate sup- Source: Castaneda forthcoming.
ply of materials; expenditure on these items rose in
real terms despite cuts in total education spending.
The government also tilted its spending in favor of
primary education. Survey data confirm that this These transfers often form part of the social action
reallocation was progressive. A longer-term reori- programs that many countries have recently intro-
entation of public spending on education toward duced (Box 7.7).
the poor continued in the 1980s despite increasing
fiscal austerity (Figure 7.7). SUBSIDIES. General subsidies often benefit the
urban poor (and occasionally the rural poor, as in
Protecting the poor through transfers Egypt and in Kerala, India), but leakages to the
in a macroeconomic crisis nonpoor are substantial. As a result, fiscal outlays
must be large to have much of an impact on the
Should transfers rise or fall during adjustment? poor, and the costs of maintaining them during a
Reducing transfers may seem an obvious part of macroeconomic crisis are high. Better targeting is
growth-oriented fiscal adjustment, but it can add highly desirablebut difficult.
to the burden of the poor. The alternative is better Jamaica and Sri Lanka, for instance, both shifted
targeting, but this too is difficult at the best of from general to targeted food subsidies in the late
times and rarely gets any easier in a macroeco- 1970s and early 1980s. The switch succeeded in
nomic crisis. There is some scope, however, for reducing aggregate subsidies: in Sri Lanka these
using three sorts of transfersubsidies, public em- declined from 5 percent of GDP in 1975-79 to less
ployment schemes, and compensation for laid-off than 1 percent in 1984, and in Jamaica they went
workersto help the poor during adjustment. from a peak of 6 percent of GDP in 1977 to less
117
Box 7.7 Packaging interventions: the role of social action programs
The late 1980s saw an increasing number of special health, drugs, nutrition, and shelter, all with a strong
multisectoral schemes that sought to mitigate the social orientation toward community involvement and the
Costs of adjustment; those in Bolivia, Ghana, and Mad- participation of indigenous NGOs. Problems of foreign
agascar are the most advanced and best known. These and domestic coordination have badly delayed imple-
programs included short-run measures to relieve dis- mentation. Both programs have served three important
tress (such as public employment, severance pay- purposes: mobilizing foreign assistance, raising do-
ments, and credit schemes for displaced workers) and mestic awareness of poverty inside and outside the
other measures (ranging from urban sanitation to pro- government, and easing the introduction of difficult
vision of textbooks) that form part of a longer-term adjustment measures.
strategy for reducing poverty. They often involve many Neither scheme is an ideal model for other countries,
donors and local and international NGOs. especially in AfricaBolivia's because of its special, au-
Bolivia's Emergency Social Fund (ESF) and Ghana's tonomous character and Ghana's because of its com-
Programme of Action to Mitigate the Social Costs of plexity. Newer schemes have been less complex than
Adjustment (PAMSCAD) have entirely different de- PAMSCAD, but most still have many components and
signs. The ESF involved a new institutionessentially mix short- and long-run measures. This raises prob-
a domestic financing agency for projects chosen by lo- lems of coordination and may divert attention and re-
cal communities and executed by private contractors. sources from where they are needed most. Sometimes
Thanks to strong leadership, highly motivated person- immediate measures to start dealing with long-term
nel (who received salaries above civil service rates), problems will be desirable, but they should be compati-
and minimal government involvement, the scheme ble with longer-term strategies. The Economic Manage-
was implemented rapidly. By contrast, PAMSCAD ment and Social Action Project in Madagascar, for ex-
works with established government agencies. It in- ample, includes measures to provide drugs and
cludes twenty-three antipoverty interventions that support family planning, but these will feed into
cover public works, credit, training, low-cost water, longer-term health interventions.
than 1 percent in the mid-1980s. But some of the opportunity cost of subsidies in a period of adjust-
poor lost out because coverage was incomplete. In ment is high.
Sri Lanka the real value of food stamps declined.
Furthermore, there is evidence that income de- PUBLIC EMPLOYMENT SCHEMES. Most countries
creased for the poorest fifth of the population, that are undergoing adjustment have to cope with
whereas less poor households were protected by a temporary drop in labor demand. Public employ-
the general rise in economic activity and growth in ment schemes are usually better than subsidies at
real urban wages. Jamaica did somewhat better, reaching the poor under these circumstances. Bo-
especially in providing food in clinics to pregnant livia, Chile, Peru, and, more recently, Ghana and
women and children under 5. Self-selection im- Madagascar have used employment schemes in re-
proved targeting for this group. sponse to macroeconomic shocks. Similar pro-
These and other cases suggest three conclusions. grams in South Asia provide a safety net for the
First, when subsidies are already well targeted rural poor. These programs often take the form of
say through feeding programs for vulnerable traditional public works schemes, but in some
groups that are held in clinics attended by the countries, including Bolivia, Ghana, and Madagas-
poorthere is a powerful case for maintaining car, the government acts as a financing agency for
them. (Chile is again a good example.) In such labor-intensive projects carried out by private
well-designed systems temporary increases in the contractors.
ration may be considered. Second, where better The poor may face a temporary decline in real
targeting is feasiblethrough food stamps, a shift wages owing to reduced demand for labor or shifts
to goods consumed mainly by the poor, or geo- in relative priceshigher food prices, for instance.
graphic targetingit should be pursued. Third, if (A study in Bangladesh found that real agricultural
improved targeting is practically or politically diffi- wages first fell and then recovered after an increase
cult, it may be better to cut subsidies across the in rice prices. Full adjustment took three to four
board, releasing resources for other programs. The years.) Public employment schemes can help the
118
poor in both cases. If targeting is to be through
sell-selection, however, the wage must be set low. Figure 7.8 How public employment schemes
The public employment programs in Chile and alleviate unemployment: Chile, 1970 to 1989
Peru are good examples. At their peak in 1983
Chile's programs provided employment to a re-
Percentage of total labor force
markable 13 percent of the labor force (Figure 7.8).
35
Afterward, the programs were run down as the
labor market recovered. At its peak in 1986 Peru's
Programa de Apoyo de Ingreso Temporal (PAIl) 30
employed 3.5 percent of the labor force. Both
schemes paid low wages, thus reinforcing target- 25
ing through self-selection. In 1986-87 two-thirds of
the employees in the Chilean programs were from
the poorest 20 percent of the population, and both 20
schemes attracted many women (hall of the pro-
gram's workers in Chile and three-quarters in 15
Peru). Low wages also meant that relatively broad
coverage was affordable, especially in Chile, where
the total cost of the program was 1.4 percent of 10
GNP at the peak in 1983. Chile's program was fi-
nanced from domestic sources, partly through 5
lower spending on civil service wages. Peru's pro-
gram, at its height in 1986, cost only 0.2 percent of
0
GNP and 4 percent of total public investment.
Both schemes were controversial. Chile's 1970 1974 1978 1982 1986 1989
aroused much hostility and was viewed by some Unemployment 0 Public employment schemes
as a make-work scheme to reemploy sacked public Total
sector workers. Peru's program was popular
among workers but became discredited when it Source: Riveros 1989 and World Bank data.
was increasingly used for political purposes. Nev-
ertheless, both programs successfully targeted the
poor and provided social security for the unem-
ployed during the recession. ing into a famine. The temporary shocks to the
Bolivia's Emergency Social Fund (ESF) included rural labor market under consideration here are
employment measures with similar objectives but very similar to the reductions in purchasing power
with a different design: a special agency, substan- that occur in a drought. In one respect adjustment
tial donor finance, and the use of private contrac- is easier: since the initial source of the shock is not
tors to hire construction workers at market wages. an interrupted food supply, the government can
It succeeded in channeling donor finance to local focus on employment without having to worry
infrastructure projects, and it softened the decline about the food distribution system. But since there
in activity in an economy that was contracting is not a visible crisis, there is less information
sharply. It recruited male workers almost exclu- about destitution and less chance of special foreign
sively and was much less finely targeted than the assistance, if a government is to use public em-
Chilean and Peruvian schemesno doubt because ployment schemes to best effect, it must develop
it paid market wages. Fewer than hall of the ESF mechanisms for monitoring vulnerable groups,
workers came from the poorest 40 percent of Boliv- and it must be willing to cut spending that benefits
ian households. the nonpoor. The demand for work provided
These cases confirm the potential for public em- through employment schemes can itself supply
ployment schemes, especially those targeted to ur- the government with information on the extent of
ban labor markets. But can such schemes reach the need.
rural poor, too? Chapter 6 concluded that a mixed In Africa donor finance is often available, but the
strategy of cash-for-work schemes and measures complementary supervisory and material re-
to increase the availability of food through private sources are much scarcer than in Latin America
and public channels can stop a drought from turn- and South Asia. It is too early to judge the schemes
119
currently being implemented, although some of in the public sector are likely to be heavy, and they
them have already proved effective in dealing with will lead to deepening poverty. Unemployment
drought (for example, in Botswana and Cape compensation, which may be politically necessary
Verde in the 1980s). The use of employment in the 1990s in any case, will also make sense as a
schemes to transfer income to the poor should be means of reducing poverty.
paralleled by efforts to introduce efficient labor- If compensation is necessary, direct payments
intensive techniques in infrastructure programs. are better than special credit and retraining
Ghana provides an example of this parallel ap- schemes. Special credit programs have been tried
proach: a food-for-work scheme run by a local in Ghana, Guinea, Mauritania, and Senegal. In
NGO has recently been substantially expanded general, such programs have had a sorry record
with the aim of providing employment for the ru- and have brought few benefits to the poor (see
ral poor. Although the organizers expect some Chapter 4). Measures to promote investment by
benefits in the form of improvements to rural infra- small-scale enterprises are an important part of
structure, transfers are the main objective. In addi- any growth strategy, but they should not be mixed
tion, the Ministry of Transport has adopted effi- up with compensation schemes. As for training,
cient labor-intensive methods of building feeder the emphasis should be on developing a training
roads. The Ministry works through local contrac- system that increases the flexibility of skilled labor
tors, who employ roughly four times as much la- rather than on special initiatives to equip displaced
bor as with other common methods and use 40 workers for specific new jobs.
percent less foreign exchange than is usual.
Lessons for adjustment in the 1990s
COMPENSATING THE LOSERS. The most visible
losers in adjustment are often formal sector urban Even if the 1990s prove less turbulent than the
workers who lose their jobs. Governments often 1980s, many countries will confront the need to
target benefits to laid-off workersfor example, adjust. Some will face new adverse shocks, and
through severance or other redeployment pay- many will be continuing the process of adjustment
ments, special credits, and retraining schemes. that started in the 1980s. The experience of the
Such measures are frequently prominent in social 1980s suggests that efforts to restructure econo-
action programs. The newly unemployed often mies in the wake of a macroeconomic crisis are, by
suffer losses, but they are generally not the poorest and large, consistent with a medium-term shift to-
members of society. Civil servants, for example, ward a pattern of growth and human capital for-
are usually well equipped to withstand a spell of mation that effectively reduces poverty. In the
unemployment, and in many countries with over- short run, however, some of the poor may lose
extended civil services, public sector workers have out. A combination of effective action on the policy
other sources of income. One study of Nigeria fundamentals (notably changes in relative prices to
found that many laid-off civil servants had main- favor agriculture) and efforts to moderate declines
tained their land rights and were able to return to in consumption (through a pause in investment,
farming (although this was less true of the young). for instance) can help many of the poor in most
Reducing public employment is often an essen- cases. But shifts in the pattern of public spending
tial part of adjustment, especially in Sub-Saharan toward goods and services consumed by poor peo-
Africa. But it is politically difficult, and some form ple and transfers targeted to them will often be
of compensation may be necessary. Payment of necessary as well. Increased capital inflows can
compensation, however, implies that fewer re- also help to soften the impact of adjustment on the
sources will be available to soften the consumption poor, and they could be particularly important in
declines of the poor, protect the capital stock, and Latin America, Eastern Europe, and Sub-Saharan
support a recovery in investment. By contrast, in Africa. But without sound economic policies, in-
Eastern Europe those who lose their jobs are also creased capital flows can provide only temporary
likely to be among the poor. There, notably in Po- relief. They are no substitute for domestic action to
land, state employees made up an increasing share protect the poor.
of the poor in urban areas in the 1980s. Job losses
120
International factors
in reducing poverty
Although domestic policies are the essential ingre- oping countries themselvesfor example, regard-
dients of a strategy for reducing poverty, interna- ing exchange rates, levels of protection, and other
tional factors play an important part. This chapter interventionsare largely responsible for the
discusses world trade, international debt, and the countries' success or failure in world trade. They
supply of foreign aid to developing countries. In also have direct and indirect effects on poverty.
each case it asks what the implications are for the These and other policies are discussed elsewhere
poor. in this Report. This chapter addresses some of the
ways in which industrial country trade policies af-
Trade and poverty fect the poor in developing countries.
Trade performance and prospects differ greatly Industrial country protectionism
among developing countries. Some middle-
income countries have a diversified production The developing countries face many obstacles in
base and substantial manufacturing and exports. selling their products abroad. High on the list are
The newly industrializing economies of East Asia the farm trade policies of the industrial countries. In
are the most obvious examples, and a number of OECD countries assistance to farmers through
Latin American countries also fit into this category. price supports, direct payments, and supply man-
Most of these economies, especially those in East agement programs seriously distorts world trade
Asia, have made substantial headway in reducing in agricultural commodities. On the whole, these
poverty. Other middle- and low-income countries policies raise domestic farm production, leading to
have an increasingly diversified economic base but more exports, fewer imports, and lower world
have not yet attained such high levels of develop- prices. Price decreases help those developing
ment. They include the emerging "second tier" of countries that import the commodities concerned,
newly industrializing economiesnotably Indone- but they hurt those that are themselves net pro-
sia, Malaysia, and Thailandand some low- ducers (and are often more efficient producers
income countries such as China and India. Many than their industrial country counterparts).
of these, too, have reduced poverty dramatically Non tariff barriers also thwart some agricultural ex-
over the past two decades. Still other countries, ports from developing countries to OECD mar-
including many in Sub-Saharan Africa, produce kets. Quantitative import restrictions most fre-
and export mainly primary commodities and lack a quently affect sugar and animal and dairy
diversified economic base. Very few of them have products, but processed fruits, groundnuts, to-
had much success in reducing poverty. bacco, and rice are also common targets. About
Given this pattern of trade and poverty, the ef- one-third of the agricultural exports of developing
fects on the poor of changes in international trade countries appear to be affected by such barriers.
relations are likely to differ according to country The effects can be serious (Box 8.1).
characteristics. The domestic policies of the devel- Tariff escalation has a particularly damaging effect
121
Box 8.1 Industrial country trade poiicy and the poor in the Dominican Republic
The Dominican Republic demonstrates vividly the im- state sugar company kept their jobs, the field workers
pact that industrial countries' trade policies can have had to fall back on subsistence farming, compete for
on the poor in developing countries. One U.S. trade meager wages in the depressed day labor market, or
policyits sugar import quotahas greatly harmed the migrate to the towns.
country's rural poor, whereas its "807 program," U.S. sugar policy is not, of course, the sole determi-
which encourages assembly operations using materials nant of rural poverty in the Dominican Republic, but it
imported from the United States, has apparently has certainly made development much more difficult.
helped to generate urban employment. The country's real per capita income has stagnated
The U.S. sugar import quota varies with domestic since the early 1980s, and rural per capita income has
production, but it is largely determined by political in- fallen by as much as 40 percent. Infant mortality and
terests. The total import quota for all exporting coun- malnutrition rates, particularly in rural areas, are closer
tries was reduced by about 70 percent between 1982 to those prevailing in neighboring Haitithe poorest
and 1987. The Dominican Republic, which traditionally country in the Western Hemispherethan to the aver-
has had the largest national allocation, saw its quota age for Latin America. This is in spite of the fact that
reduced proportionately. During the same period the the per capita GNP of the Dominican Republic is al-
European Community (EC), the world's largest sugar most equal to the Latin American average.
exporter, continued to export more than six million In sharp contrast, the U.S. 807 program has helped
tons of sugar a year. The combination of a shrinking the economy. If an import under the program is assem-
preferential U.S. quota and booming, subsidized EC bled from U.S. material (an example would be shirts
exports wreaked havoc on the world sugar market. sewn from U.S. cloth), tariffs are levied only on the
Throughout the 1980s the world price was only about value added by the exporter. Since 1980 the number of
one-third of the U.S. protected price. The Dominican firms in export processing zones (EPZ5), which mainly
Republic is one of the world's more efficient producers assemble U.S.-produced parts, has risen from 70 to
of cane sugar, yet it found itself unable to compete at 330. The firms now account for about half of all manu-
the unprecedentedly depressed world price. At the facturing employment and have since 1980 generated
same time, its access to the U.S. preferential market the bulk of the nation's incremental merchandise ex-
was shrinking. Its only recourse was to cut sugar pro- ports. Employment in EPZ firmswhich mainly use
duction and exports drastically. Four sugar mills were semiskilled workershas increased from 16,000 to
closed by 1990, and production plunged 40 percent fol- 130,000. Since 1980 such firmsstimulated by the U.S.
lowing the reduction in the quota. 807 programhave provided one-fifth of the urban
Since sugar is the main rural economic activity in the jobs created in the Dominican Republic.
Dominican Republic, the repercussions were severe. U.S. trade policy has substantially altered the shape
The average real incomes of cane workerswho were of the Dominican economy. EPZ exports have replaced
already among the lowest-paid people in the rural sugar as the country's largest foreign exchange genera-
economycollapsed. About two-fifths of the estimated tor. Although some new higher-paying jobs have been
50,000 expatriate Haitian canecutters lost their jobs, to- created for the urban poor, the net effect of the policies
gether with perhaps half of their Dominican counter- has been increased impoverishment in the Dominican
parts. Although most mill employees of the Dominican Republic.
j
on low-income countries. Industrial countries' tar- are in the primary stage; more than 90 percent of
iff duties generally increaseor "escalate" their cocoa and sugar exports are unprocessed.
according to the degree of processing. The pur- The amount of labor used in processing varies
pose of this protection is to retain high value from product to product, but many such activities
added processing in the developed countries. would be labor-intensive, and low-income coun-
Products for which escalation is most pronounced tries could have a comparative advantage in them.
include tobacco, rubber, leather, paper, cotton, Industrial country trade restrictions concerning
jute, and iron. Average industrial country tariffs textiles and clothing are another example of how
on fruit preparations are double the tariffs on fresh protectionism can hurt low-income countries. Ex-
fruit; tariffs on vegetable preparations are more ports of these products by developing countries
than double those on fresh vegetables. Escalation amount to $30 billion. Much of the world's trade in
discourages the processing of commodities in de- textiles and clothing is governed by the Multifibre
veloping countries. More than 70 percent of devel- Arrangement, which sets bilaterally negotiated
oping countries' meat, fish, and vegetable exports quotas designed to slow the growth of exports
122
from low-cost (mainly developing country) pro- Table 8.1 The effect of complete trade
ducers (Box 8.2). liberalization on selected countries
The industrial countries have put in place a num- Change GNP per capita
ber of special arrangements with the avowed pur- in exports 1987
Economy (percent) (dollars)
pose of stimulating exports from developing coun-
tries. The main such arrangement is the High and middle income
Hong Kong 25.9 8,180
generalized system of preferences. With some ex-
Korea, Rep. of 21.6 2,900
ceptions, these schemes have done little to offset Yugoslavia 14.0 2,480
the damage done by the much broader range of Dominican Rep. 13.0 740
overtly protectionist devices. The benefits from Tunisia 11.4 1,190
such preferences have been mainly concentrated Mauritius 10.5 1,500
in a handful of middle-income countries with rela- Thailand 10.3 850
Morocco 8.9 620
tively few poor people (Box 8.3). Singapore 7.2 8,010
Brazil 6.8 2,030
Assessing the effects of trade liberalization Low income
on developing countries Sri Lanka 20.9 410
China 13.0 300
Developing countries will reap substantial gains Pakistan 10.7 340
Haiti 9.3 360
from trade liberalization by the industrial coun- India 8.6 310
tries. It is difficult, however, to make a precise cal- Bangladesh 1.0 160
culation of these benefits. Results would vary from Tanzania 3.3 210
country to country, and many countries would Burundi 5.5 250
gain only over the long-term. Table 8.1 presents Nepal 9.6 160
Somalia 24.3 170
the results for selected developing countries of a
a. Percentage change in value of exports in 1983 attributable to full
simulation exercise showing the short-run changes trade liberalization.
in exports if trade were completely liberalized. Source: Karsenty and Laird 1987 and World Bank data.
Box 8.2 Bangladesh is not spared from MFA quotas
The emergence of an export-oriented garment industry quotas covered only two product groups: cotton coats
in Bangladeshone of the world's poorest countries, and men's and boys' shirts. Seven additional groups
with a GNP per capita of only $170 and with 57 percent were brought under quotas in 1986. By 1987 there were
of its population in povertywas a great success in the thirteen quotas.
early 1980s. Quota restrictions on Korea's textile ex- Initially, with stocks of garments building up in
ports under the Multifibre Arrangement led Korean ex- warehouses, on docks, and aboard ships in U.S. ports,
porters to look elsewhere for trade opportunities, and a the quotas had a big effect in Bangladesh. Many factor-
Korean joint venture was a catalyst in building up an ies closed. The United States later increased its quotas,
efficient industry in Bangladesh. By the end of 1985 and exporters in Bangladesh managed to diversify their
there were about 700 garment factories in the country. markets as well as their products. In 1987 Bangladesh
From a negligible start around 1980 Bangladesh's gar- exported over $300 million in thirty-five garment cate-
ment exports rose to more than $100 million in a couple gories, including products made of cotton, wool, and
of years. The garment industry provided employment man-made fabrics. These products accounted for
for many poor people, mostly women. The production nearly 30 percent of the country's exports.
and export of garments is a clear example of the kind of The story does not appear to be over. In 1988 most of
efficient labor-intensive activity advocated in this Re- Bangladesh's clothing and textile products subject to
port. quotas in the United States were again reaching their
In 1985 Canada, the United States, and some Euro- quota limits: the overall quota utilization rate was more
pean countriesnotably the United Kingdom and than 90 percent. The dismantling of the Multifibre Ar-
Franceimposed export quotas on Bangladeshi gar- rangement would reduce the instability in Bangla-
ments. About 80 percent of Bangladesh's garment ex- desh's most important industry, permit the country to
ports were going to the United States at that time, yet fulfill its great potential as a garment exporter, and lead
their share in total U.S. imports of clothing and textile to substantial employment gains for poor workers in
products was less than 0.5 percent. The initial U.S. the industry.
123
Box 8.3 The generalized system of preferences: who benefits and how much?
The generalized system of preferences (GSP) is com- mies have received an estimated 44 percent of the total
plex. Each industrial country in the schemeincluding benefit; each gains three times more than the next larg-
the United States, Japan, and the members of the Euro- est beneficiary, Brazil.
pean Communitygrants its own preferences, and An assessment of the GSP found that of the thirty
there is considerable variation among them. Most principal beneficiaries, only fourBangladesh, China,
schemes are limited to manufactured exports, but a few Haiti, and Indiawere low-income countries. Except
cover relatively small quantities of agricultural prod- for Haiti, incremental exports attributable to the CSP
ucts. Some donors charge no duties at all on specified were tiny.
products from qualifying countries, and others have An effort to work out what the effects would be if
duties but at a preferential rate; the least-developed GSP schemes were extended without any limitation on
countries are usually given a zero rate. product or country coverage found that imports by do-
Most studies agree that the GSP does little to in- nors from preference-receivers would increase by
crease trade. It has been estimated that total imports in about $20.7 billion. But the poorest countries would
most preference-granting countries are less than 0.5 actually face some loss of trade because they currently
percent higher as a result of the system. In the benefi- enjoy better treatment than other GSP beneficiaries in
ciary countries the effects are generally about I or 2 most markets. The same would also be true of some
percent of total exports, with important exceptions. beneficiaries under the Lomé Convention, a system of
The most exhaustive effort to assess the direct trade trade preferences granted by the EC.
effects of all the GSP schemes concluded that in 1983 The various GSP schemes may have helped some
about $6.5 billion of the imports by GSP donors from poor people in the middle-income countries to the ex-
the preference-receiving countries could be attributed tent that the expanded exports were labor-intensive in
to the various schemes. This represented just 2.3 per- production, but the poor in low-income countries can-
cent of total imports, which amounted to $280.6 billion. not have gained much at all. The clear tendency to
The gains, with a few exceptions, have been concen- exclude agricultural products from the schemes
trated in a few middle-income economies, particularly strengthens this conclusion.
Hong Kong, Korea, and Taiwan, China. These econo-
/
In general, the successful middle-income ex- preferences would be lost. Many countries pro-
porters, which have relatively little poverty, would duce few of the goods that might experience a rise
benefit. The outcome would also be positive for in demand as a result of tariff cutting in the Uru-
some low-income countries, including those with guay Round or other forums. And the low-income
the largest numbers of poor people, China and In- countries suffer from structural rigidities that make
dia. But for many low-income countriesmainly them slow to adopt new patterns of production
but not exclusively in Sub-Saharan Africa and and exports.
mainly exporters of primary productsany bene- Despite this, industrial country liberalization
fits from industrial country liberalization would be should continue to be the goal. It would have the
slow to materialize. Several countries might actu- following favorable effects.
ally lose in the short run from complete trade liber- Poor people in middle-income countrieswho
alization because their existing trade preferences number about 175 millionwould probably gain,
would disappear. Moreover, net food importers since many of the exports, especially manufac-
would suffer some losses because liberalized trade tured exports, that would be stimulated by liberali-
in agricultural goods would probably raise world zation are labor-intensive.
food prices. In some sectors, notably textiles and clothing,
Much depends on the commodity composition low-income as well as middle-income countries
of exports (Figure 8.1). Countries that produce and would be likely to benefit immediately from liberal-
export mainly primary commodities, such as the ization. If all bilateral quotas and tariffs under the
low-income African countries, could not be ex- Multifibre Arrangement were removed, develop-
pected to profit much from liberalization in the ing countries would realize an estimated benefit of
short run. Tariffs are already low or nonexistent on $11.3 billion, or more than one-third of their total
most such commodities, demand in the importing exports of clothing and textiles. The expansion of
countries is often insensitive to price, and existing trade brought about by abolition of the Multifibre
124
Figure 8.1 Commodity composition of exports, 1988
(percentage of total value in dollars)
100
75
50
25
0
5$.
C, Cf39
U Food 0 Agricultural raw materials 0 Ores and metals (excluding fuels) 0 Manufactures
Note: The data for Sub-Saharan Africa do not include Botswana and Lesotho.
Source: United Nations Statistical Office.
Arrangement might raise employment in these almost 80 percent in the case of coffee, 76 percent
countries' textile and clothing industries by 20 to for wool, and 52 percent for cocoa.
45 percent. China might gain about $2.3 billion and In the end the gains for the poor could be large.
Bangladesh about $340 million (or 44 percent of its But the diversification required in many countries
total exports in 1986). cannot take place overnight. It will take time for
Trade liberalization would have longer-term the poorest countries to reorient their production
effects that would eventually benefit even the low- and export bases to take advantage of the more
income countries that are largely dependent on liberal trade environment that might emerge from
primary commodities. But this would depend on the current round of multilateral trade negotia-
whether these countries adopted policies that tions. Policy and institutional reforms will be re-
would encourage a supply response. If tariff esca- quired. These efforts will call for substantial invest-
lation could be substantially reduced, for example, ment, and in the years ahead this is bound to be
it would be easier to diversify away from primary difficult. External assistance, including debt relief
commodities. One study found that the removal of and the provision of concessional finance, will thus
all industrial country tariffs on processing would have a significant role in these countries for at least
increase such activities in developing countries by the next decade.
125
Debt and poverty As with the conversions of concessional loans to
grants, however, the amount of actual debt relief
delivered by the Toronto accords is limited so far.
Limited export earnings are likely to be an impor-
The cash flow savings for twelve Sub-Saharan Af-
tant constraint on the development prospects of rican countries participating in the Special Pro-
many countries. Debt burdens are another prob-
gram of Assistance in 1989 were about $50 million
lem. There are twenty-six severely indebted low-
(in relation to rescheduling under standard terms),
income countries, of which twenty-four are in Sub-
or 2 percent of their debt service.
Saharan Africa, and nineteen severely indebted
middle-income countries. Chapter 1 discussed the Projections indicate that Toronto-terms resched-
debt problems of the second group of countries ulings are likely to have only a small effect on the
and the international initiatives for dealing with future stock of debt. if there is no change in the
them. This chapter focuses on the debt burdens of debt-relief options offered by creditors and if To-
low-income countries. ronto terms are applied repeatedly (that is, if re-
In absolute terms the debt of the severely in- scheduled debts coming due are again rescheduled
debted low-income countries$103 billion in on the same terms), the total reduction of debt to
all bilateral creditors by the end of 2000 would be
1988is not high compared with the $516 bfflion
about $2.0 billion, or only 11 percent of long-term
owed by the middle-income group. But some indi-
nonconcessional debt in 1988.
cators of debt burden are actually much higher for
the first group. The ratio of debt to GNP for the In 1988 interest payments on the external debt of
low-income countries, for example, was 111 per- the low-income countries of Sub-Saharan Africa
cent in 1988; for the middle-income countries it (including Nigeria) totaled about $2.9 billion. This
was 54 percent. represented about 27 percent of net disbursements
A number of international initiatives for dealing
of all official development assistance (ODA) to
with the debt burdens of low-income countries these countries in that year. The ratio of interest
payments to pure grants was more substantialit
have been undertaken. Since 1978 many bilateral
donors have converted part of the official debt was 47 percent in Kenya, 52 percent in Ghana, 50
owed by these countries into grants. It is estimated percent in Madagascar, 73 percent in Togo, and 57
that $3 billion of official debt has been canceled in percent in Zaire. Such high levels of interest pay-
this way; about $2 billion represented claims on ments severely limit the contributions that aid can
low-income Sub-Saharan African countries. Al- make to increasing consumption and investment
and reducing poverty.
though this is an important achievement, the can-
celed debt amounts to only about 3 percent of the A number of other low-income countries have
total outstanding debt of the low-income African had increasing recourse to nonconcessional bor-
countries at the end of 1988. Moreover, many of rowing as a result of stagnation in aid flows. For
these countries wifi not benefit greatly from fur- example, over the past decade India's ratio of debt
ther conversions of loans to grants, since most of service to exports has increased from less than 10
their concessional loans have already been for- percent to about 30 percent. In 1988 its interest
given. payments on external debt actually exceeded net
The agreement reached at the Toronto economic disbursements of ODA (interest payments were
summit in June 1988 provides debt relief on official $3.1 bfflion and net disbursements were $2.1 bil-
bilateral nonconcessional debt under the auspices lion). Pakistan's debt service is about 25 percent of
of the Paris Club for severely indebted low-income exports. In 1988 interest payments on external debt
countries that are undertaking adjustment pro- were about 45 percent of the net disbursements of
grams. As of March 1, 1990, the "menu" of op- ODA that it received, if India, Pakistan, and a
tions agreed on at Toronto (partial write-offs, number of other non-African low-income coun-
longer repayment periods, and more concessional tries are to avoid the fate of the severely indebted
interest rates) had been applied to sixteen Sub- low-income countries, they wifi clearly need to
Saharan African countries. The Toronto accords boost exports substantially. But they wifi also re-
were a breakthrough in dealing with the debt of quire continued substantial inflows of external as-
low-income countries because they officially sanc- sistance on concessional terms.
tioned the principle of debt relief. It was acknowl- Many low-income countriesespecially but not
edged that these countries could not simply "grow exclusively in Sub-Saharan Africafind them-
out of debt." selves with daunting debt and debt service bur-
126
dens at a time when they need to invest more (in need aid just to maintain their present low quality
order to improve their long-term prospects) and, of life, when it should go toward improving their
simultaneously, to increase the consumption of long-term prospects. Viewed from the perspective
large numbers of people in poverty. Further efforts of this Report, another fact about aid is particularly
by the international community wifi be needed to worrying: in many countries aid has done much
reduce their debts and to increase concessional as- less than might have been hoped to reduce pov-
sistance to them. These efforts should be condi- erty. In its comprehensive review Twenty-Five Years
tional on appropriate policy reform in the coun- of Development Cooperation, which covers the period
tries concerned. Aid and debt relief wifi be of no 1960-85, the OECD's Development Assistance
avail if appropriate policies are not in place. Committee (DAC) concluded that "the most trou-
bling shortcoming of development aid has been its
Aid and poverty limited measurable contribution to the reduction-
as distinguished from the relief-of extreme pov-
Official development assistance in 1988 was $51 erty, especially in the rural areas of both middle-
billion-half the net receipts of external capital by income and poor countries."
developing countries. For the low-income coun- This part of the Report asks why aid has appar-
tries, where most of the world's poor live, aid rep- ently been less effective than it might have been in
resents close to 70 percent of net external finance. reducing poverty and examines how future pro-
In many of these countries aid is a much more grams can be strengthened and improved. It notes
important source of foreign exchange than are ex- that much aid-especially bilateral aid-has simply
ports (Table 8.2). Given the scale of this effort, it is not been concerned with economic development
important to assess the effectiveness of external or poverty reduction and that although aid specifi-
assistance and to learn from past successes and cally devoted to these objectives has had many
failures. Aid donors have increasingly recognized clear successes, there have also been some prob-
this and have been modifying their policies- lems.
through improved coordination, for example-to
take account of the lessons of experience. The motives for supplying aid
Although many countries have benefited signifi- Aid is provided by three broad groups of donors:
cantly from aid, some of the poorest countries bilateral donors, multilateral agencies, and non-
have become trapped in "aid dependency." They governmental organizations. Their respective
shares in the total aid receipts of developing coun-
tries are shown in Figure 8.2. Box 8.4 discusses
Table 8.2 The relative importance of aid some characteristics of the aid they provide.
in selected developing countries, 1987 Bilateral donors, in particulai, provide aid for
Total aid Aid receipts many reasons-political, strategic, commercial,
receipts as from DAC countries and humanitarian. Reducing poverty is only one
a percentage as a percentage of
Country of GNP exports to OECD markets motive, and it is usually far from the most impor-
Guinea-Bissau 89.3 956.0 tant. This has several results.
Gambia, The 55.6 89.1 Not all aid goes to low-income countries. In
Mozambique 51.7 509.2
Chad 28.3 332.2
1988 about 41 percent of it was directed to middle-
Malawi 23.6 61.0 and high-income countries.
Zambia 22.5 53.0 There are great disparities in the per capita
Tanzania 17.0 270.4 allocation of aid, and the most generous per capita
Lesotho 15.6 1,041.7 allocations do not necessarily go to the poorest
Bangladesh 9.4 105.9
Benin 7.9 108.5
countries (see Table 8.3). Despite their poverty,
Sierra Leone 7.6 26.4 China and India have received almost the least aid
Kenya 7.5 60.3 per capita.
Myanmar 3.6 261.6 Many "aid" programs in donor countries
Pakistan 2.7 18.2 cover an assortment of activities (including com-
Indonesia 1.9 7.7
India 0.7 11.8 mercial and strategic initiatives) which often have,
China 0.5 3.9 at best, a tenuous connection with development.
Source: Organisation for Economic Co-operation and Development Only about 8 percent of the U.S. aid program in
1990. 1986, for example, could be identified as "develop-
127
support to countries with balance of payments dif-
Figure 8.2 Sources of developing countries' net ficulties and by financing development projects in
aid receipts, 1987 specific sectors. How far has aid of this sort helped
(percent) to reduce poverty?
A clear link between aid and the reduction of
poverty has been hard to find. The main reason is
Nongovernmental (14) that aid is only one of many factors influencing
poverty and is often far from the most important.
Domestic policies, the institutional and managerial
capacity of the recipients, and other variables often
matter more. Although the overall effects of aid
may be unclear, detailed studies of individual
countries have highlighted its role in specific in-
Multilateral (22) stances.
Aid contributed in an important way to growth
and declining poverty in Korea, for example. A
recent study concluded that aid-financed invest-
ments in infrastructure paved the way for the
country's successful export-oriented growth in the
1960s, that assistance to education was vital to the
development of a literate labor force, and that aid
Note: The total amount of aid receipts was $40.3 billion. The amount
promoted agricultural and rural development.
received from nongovernmental organizations was $5.5 billion, of Korea has moved from an initial heavy depen-
which $2.2 billion represented funds that these organizations dence on aid toward self-reliance. In the 1950s con-
received from official aid agencies and transferred to developing
countries. cessional assistance financed about 70 percent of
Source: Organisation for Economic Co-operation and Development
1989 and 1990. total imports and accounted for more than 75 per-
cent of all savings. Later, external support from
official nonconcessional sources increased; Korea
has received a cumulative total of $7 billion in such
ment assistance devoted to low-income coun- assistance from the World Bank. The country now
tries." receives substantial volumes of private capital
A substantial amount of bilateral aid is "tied";
from commercial sources.
recipients are required to buy goods and services
from donor countries. Approximately two-thirds
of all aid supplied by DAC members is in this cate-
gory. Table 8.3 Per capita aid receipts, 1988
(dollars)
Donors prefer to finance physical capital in-
stallations that help their own firms and exporters, Aid receipts GNP
Country per capita per capita
and they are reluctant to support the operating
Israel 282.07 8,650
("recurrent") costs of aid-funded undertakings. Jordan 108.95 1,500
But many of the initiatives required for reducing Gambia, The 102.63 200
poverty in the developing world are in sectors Senegal 78.85 650
such as health and education that make intensive Zambia 63.73 290
use of recurrent resources. Egypt 29.91 660
Nepal 22.05 180
In sum, a substantial proportion of aid is provided Ethiopia 21.05 120
at least partly for purposes other than to promote Syrian Arab Republic 16.34 1,680
Bangladesh 14.62 170
development. It stands to reason, therefore, that Pakistan 13.32 350
the impact of aid on poverty has been smaller than Myanmar 11.22
it might have been. Indonesia 9.34 440
India 2.58 340
Aid for development China 1.84 330
Nigeria 1.09 290
When aid has served development, it has done so a. GNP per capita estimated at less than $500.
Source: Organisation for Economic Co-operation and Development
in two main ways: by providing general budgetary 1990 and World Bank data.
128
Box 8.4 Foreign aid for development
Official aid is disbursed to recipient countries from bi- which they have particular experience and expertise,
lateral and multilateral sources. It comprises grants and and they are often responsible for novel or innovative
loans (with at least a 25 percent grant element) from approaches. Multilateral agencies generally give more
official sources that have promotion of economic devel- weight to developmental criteria, and some play a lead-
opment and welfare as their main objectives. In addi- ing role in the policy dialogue with recipients and in
tion, a great many nongovernmental organizations in the aid coordination process. But many regard these
industrial countries also provide aid. Measures of aid agencies' approach as excessively technocratic, and the
effort differ considerably depending on the figures conditionality attached to assistance from the develop-
used. Among DAC members, the United States is the ment banks is often a bone of contention. Nongovern-
largest donor in terms of total volume; it provided mental organizations are particularly adept at local in-
$10.1 billion in aid in 1988. Japan, with $9.1 billion, was terventions, participatory approaches, and poverty
the next largest provider of total aid. But the United reduction. But they are small, and many have ques-
States is near the bottom among DAC donors when aid tioned whether they are capable of providing large-
is calculated as a proportion of GNPthe figure is only scale delivery of services on a sustained basis.
0.21 percent. At the other extreme, Norway gives 1.10
percent of GNP. Aid from Arab countries in 1988 to-
taled $2.3 billion, of which $2.1 billion was from Saudi Box figure 8.4 Aid as a percentage of GNP 1988
Arabia and $108 million from Kuwait. Saudi Arabia's
ratio of aid to GNP was 2.70, the highest for any donor,
and Kuwait's was 0.41, above the average for DAC
members (Box figure 8.4).
U.S. aid is mainly directed to countries of key strate-
gic importance such as Egypt and Israel; French and
British aid goes disproportionately to former colonies.
Some donorsthe Nordic countries, for example
have emphasized the reduction of poverty more than
others.
Multilateral aid is provided by two main categories of
suppliers: the multilateral development banks and the
functional agencies of the United Nations system. Net
disbursements of concessional assistance from the mul-
tilateral development banks totaled about $5 billion in
1988; of this, assistance from IDA accounted for about
$3.8 billion. The United Nations assistance agencies,
such as the United Nations Children's Fund, the
United Nations Development Programme, and the
World Food Programme, provided about $3.8 billion.
Nongovernmental organizations in industrial coun-
tries supplied about $5.5 billion in financial resources
to developing countries in 1987. In addition, they are
extremely important in educating the public on devel- o DAC member countries 0 Others
opment issues. DAC average
Bilateral donors are frequently criticized for provid-
Source Organisation for Economic Co-operation and Development 1989.
ing aid for political, strategic, and commercial reasons.
But they are able to focus on countries or sectors in
A number of other countries have moved toward lending to Indonesia was concessional and was ex-
sell-reliance, and most of them have also made tended through the International Development
good progress on poverty. Indonesia, where the Association (IDA). Virtually all Bank lending to In-
headcount index of poverty fell from 58 to 17 per- donesia since then has been in the form of noncon-
cent between 1970 and 1987, is a good example. cessional loans. The country advanced partly be-
During that time it received more than $12.9 billion cause of its newfound oil wealth, which the
in foreign aid, making it the seventh largest aid authorities managed very well. Nevertheless,
recipient. Between 1968 and 1974 all World Bank there is general agreement that without aidfor
129
agriculture, education, and family planning, for aid recipients have often followed policies that
examplethe country's efforts to develop and were harmful to both growth and poverty reduc-
thus to reduce poverty would have made slower tion, and donors have not always monitored pro-
progress. grams carefully or linked assistance to policy re-
The story is different in other countries form.
Tanzania, for example. Tanzania's GDP, which had
grown by more than 6 percent a year in the imme- Projects and poverty
diate postmdependence years (1961-67), increased
4.3 percent a year during 1967-73, 2.5 percent dur- A large part of the aid that flows to poor develop-
ing 1973-79, and less than 1.4 percent during 1979- ing countries finances specffic projects. Again, the
85. All the while, aid poured in. Between 1970 and effect of aid on the poor is difficult to judge. One
1988 the country received about $8.6 billion in con- reason is that sometimes external assistance for
cessional assistance. specific antipoverty projects indirectly finances
Tanzania's espousal of African socialism, with its other undertakings that may not help the poor and
stress on equity and the fight against poverty, ap- may even hurt them. Nevertheless, observers
pealed greatly to bilateral and multilateral donors. have generally concluded that projects in at least
The country's development strategy was based on the following areas have reduced poverty: agricul-
cooperative "villagization" in agriculture, the pro- tural research (particularly on crops grown pre-
vision of health, education, and water to the rural dominantly by the poor) and extension; the intro-
poor, and a headlong drive toward industrializa- duction of high-yielding varieties of foodgrains in
tion. But the economywide policies pursued were Asia and especially India (the "Green Revolu-
frequently ineffective; they actually discouraged tion"); irrigation that benefits many small farmers,
agriculture and supported large, inefficient state especially in Asia; basic infrastructure in rural ar-
enterprises. The public sector took on a greater eas, especially farm-to-market roads in many low-
role in every aspect of the economy, contributing income Sub-Saharan African countries; primary
to severe policy distortions and misallocation of education, basic health care, and nutrition; and re-
resources. This combination of policies meant that lief from natural and man-made disasters.
Tanzania's aid-financed initiatives largely failed. The World Bank's experience with rural develop-
Per capita real private consumption has fallen by ment and urban poverty projects is discussed in
43 percent since 1973, and poverty has increased. Boxes 8.5 and 8.6. Such projects have registered
Whereas Korea and other aid recipients greatly re- many successes, as have many efforts of other do-
duced their reliance on aid, Tanzania came to de- nors. A review of experience reveals, however,
pend on it increasingly. In recent years, however, that some aid projects intended to reduce poverty
Tanzania has undertaken significant institutional have gone wrong, although it is impossible to give
and policy reforms, and the World Bank and other a precise number. Among the reasons why some
donors have provided support for these efforts. aid projects have been ineffective are the follow-
Other aid recipients have also traveled the road ing.
toward aid dependency. Sudan received about
$9.6 billion in foreign aid during 1970-88, making it LACK OF COMMITMENT. Some poverty-oriented
the ninth largest aid recipient in absolute terms; projects are undertaken mainly to satisfy donors
Zaire received $5.8 billion over the same period. and, often, because the government wants funds
Neither has made much headway on growth or for something else. Without a commitment to help
poverty. Mozambique, Niger, Togo, and Zambia the poor, agricultural credit projects intended to
are similar cases. Outside Africa, Haiti has also benefit small farmers end up making cheap loans
received substantial aid and yet has failed to re- available to large farmers, public works projects
duce poverty. benefit contracting firms rather than the poor, and
One lesson stands out from this picture: success the provision of water supply and other basic ur-
and failure in aid programs spring from the actions ban services is diverted to wealthier neighbor-
of both donors and recipients. When countries hoods. Enormous sums have been spent for pur-
have adopted domestic policies that use aid effec- poses that have nothing whatever to do with
tively, and when donors have backed good advice reducing povertymilitary expenditures, for ex-
with assistance aimed at overcoming development ample, and lavish "cathedrals in the desert." In
bottlenecks, the result has been a virtuous circle of 1986 the developing countries spent $159 billion on
growth and diminishing dependence on aid. But their armed forcesabout five times what they re-
130
Box 8.5 The World Bank's experience with rural development projects
The World Bank's approach to poverty has varied over production directly in project areas. Eighty-three
the years. During the presidency of Robert S. McNa- projects undertaken between fiscal years 1974 and 1979
mara (1968-81), the Bank emphasized targeted inter- are estimated to have benefited about 4.7 million farm
ventions explicitly designed to benefit the poor in both families directly. For these projects the cost per benefi-
rural and urban areas. In recent years these projects ciary family at project completion was about $1,100 (as
mainly begun during the mid- and late 1970shave against an estimate of $960 when the projects were first
been subjected to a number of assessments by the Bank appraised).
itself. Although it is exceedingly difficult to generalize By now a number of lessons from these rural devel-
from a vast body of diverse project experience, several opment projects have been documented. A few are
conclusions have emerged. given here.
The basic goal of the Bank's rural development
Many integrated projects were too ambitious and
projects was to shift the focus of agricultural develop-
complex, and they often placed impossible demands
ment toward smallholder farmers and to increase these
on local leadership and institutions. As a result, they
farmers' productivity, output, and incomes. Rural de-
performed worse than projects with simpler designs.
velopment projects usually included agricultural credit
In many cases the benefits flowing from projects
to poor farmers, research on crops grown mainly by
were not sustained beyond project completion. One
such farmers, extension services, the construction of
reason was the Bank's tendency to create independent
infrastructure (primarily rural access roads), and some
project management units. Although this may have fa-
"directly productive" investments. Many projects had
cilitated implementation over the short term, it was
an "area development" focus and were undertaken in
difficult to deal with national bureaucracies once dis-
areas that suffered extreme poverty even by the rural
bursements on projects were completed. Another rea-
standards of developing countries. For example, in
son was that project beneficiaries were often not in-
Funtua Statethe site of the Bank's first rural develop-
volved in decisions, and they felt that they had little
ment project in Nigeriathe rural population had the
stake in sustaining the projects. Projects that encour-
lowest per capita income of any state in the country.
aged participation worked better.
Many projects were "integrated": they involved a
Although there was awareness of the importance
specified sequence of actions by a variety of agencies
of an appropriate policy framework, finance was often
within the recipient countries and by the Bank itself. In
granted in unpromising circumstances in the hope that
fact, the phrase "integrated rural development
governments would be encouraged to change their pol-
project" became synonymous in many people's minds
icies. As it turned out, the larger policy environment
with the Bank's whole rural development effort in the
was perhaps the single most important factor in the
1970s. There were other kinds of rural development
success or failure of the projects. Government policies
projects, toosome nationwide projects, for example,
on prices, interest rates, and input supplies were fre-
included a substantial component for small farmers or
quently at variance with project objectives. Moreover,
low-income producers but were not targeted on the
the projects themselves often proved ineffective levers
rural poor in a specific geographic area. Several nation-
for influencing overall national policies for agricultural
wide agricultural credit projects in Mexico and a na-
development. Many tended to be successful "en-
tionwide agricultural research and extension project in
claves" within national agricultural systems that were
Brazil are examples.
still largely inefficient and inattentive to the needs of
In 1987 the Bank's Operations Evaluation Depart-
poor farmers.
ment conducted a thorough review of 112 completed
rural development projects undertaken during the The balance sheet of successes and failures for rural
McNamara years. Total Bank lending for these projects development projects during the McNamara years has
was about $2.7 billion. The review used the simplest influenced the Bank's subsequent approaches to rural
and most commonly accepted Bank measure of a poverty. Today's projects pay more attention to overall
project's "success" or "failure' 'the economic rate of sectoral policies, place greater stress on institutional
return. A return of 10 percent normally qualifies as development, put less emphasis on complex multisec-
successful. The evaluation found that 63 percent of the toral approaches, and are increasingly realistic about
projects were successful and that the projects overall the feasibility of focusing project benefits precisely on
had an average rate of return of 17 percent. About 85 narrow target groups.
percent of the projects appear to have increased food
131
Box 8.6 The World Bank's experience with urban poverty projects
In response to the diverse problems of cities in devel- of the least expensive government-subsidized housing.
oping countries, the World Bank in the 1970s began to In El Salvador houses in the sites and services project
expand its lending for urban development. Lending cost less than half as much as the cheapest conven-
increased from a modest $10 million in 1972 to more tional houses in the public sector.
than $2 billion in 1988. Much of this lending has sought The shelter projects were, however, not without
to address poverty. In a sample of projects for which problems. Assessments of completed projects have
poverty data are available, nearly three-quarters had at provided a number of important lessons.
least 40 percent of their funds allocated to the poor.
Almost 60 percent of the Bank's total urban lending The larger policy and institutional environment
since 1972 has gone to shelter operations, although can hamper urban poverty projects, just as it has rural
lending for the direct provision of shelter has declined development projects. The concentration of land in the
substantially in recent years. Lending for transport and hands of large property owners often made it difficult
for water supply and sanitation are the other main to acquire land for housing projects. Zoning regula-
components of the Bank's urban operations, and there tions also generally discriminated against the poor. The
has also been considerable lending for citywide finan- inefficiencies of service-delivery organizations often
cial and institutional reforms. thwarted efforts to provide municipal services to poor
areas.
As with the rural development programs described
in Box 8.5, the Bank's poverty emphasis in urban areas It is difficult for urban poverty projects to reach the
poorest people. The most comprehensive study con-
is prominently identified with the McNamara years.
The Bank's shelter strategy of the time attracted inter- ducted by the Bank found that people in the middle
rather than the lower part of the urban income distribu-
national attention. It relied primarily on sites and ser-
tion tended to gain most.
vices housing projects and slum-upgrading projects.
Project-level interventions, such as shelter
In sites and services projects, land parcels equipped
with rudimentary urban services were provided to peo- projects, often do not have much influence on the over-
all urban policies of recipient countries. Some old plan-
pie who then either constructed their own dwellings or
contributed to the construction. The projects were de- fling and design criteria gave way to lower-cost solu-
tions, but the laws, codes, and regulations that provide
signed to reorient typical public sector housing policies
the framework for private housing development were
in developing countries, which had benefited mainly
generally left unchanged. The most recent assessment
middle-income households and not the poor. The
projects provided beneficiaries with core houses, some of the Bank's urban projects concluded that in most
infrastructure, social facilities, and employment sites.
countries sites and services projectsagain like many
They also made available financing for plots, houses,
rural development projectsbecame "enclaves."
and building materials. Typical "upgrading" projects Rarely did governments establish programs indepen-
also relied greatly on self-help schemes. They sought to dent of external donor support. As a result, the direct
improve infrastructure through a comprehensive pack- provision of shelter did not have the broad, long-term
impact on the sector that had been expected.
age that included water, sanitation, drainage, solid
waste removal, roads, and footpaths. The Bank's new approach to the urban sector
How many of these urban shelter projects were suc- stresses broad policy and institutional issues and
cessful? As with rural development projects, it is hard strives for a sectorwide impact. New priorities include
to be precise. About thirteen million households are better urban administration, improved mobilization of
estimated to have benefited from shelter projects be- municipal resources (which may involve the recovery
tween 1972 and 1989. The Bank's shelter strategy of costs for the supply of services), more efficient urban
showed that "housing the poor" is possible. The land markets, and a cleaner urban environment.
changes in design standards introduced by some of the Project-level interventions, such as the provision of ba-
projects did in fact lower shelter costsin some cases sic urban infrastructure, will continue to be important.
dramatically. In Zambia, for example, houses in sites But projects will need to pay much greater attention
and services projects cost less than one-fifth the price than in the past to these wider issues.
ceived in aid. In 1984 they spent almost as much THE SECTORAL AND MACROECONOMIC CONTEXT. In
on arms imports as on all their health programs. the 1970s the presumed shortcomings of "trickle-
The continued provision of aid needs to be care- down" approaches to poverty reduction led many
fully reexamined if it seems that developing coun- donors to emphasize direct, project-specific inter-
try governments are not serious about reducing ventions. Many of these projects were successful.
poverty. But projects explicitly intended to reduce poverty
132
were often undermined by sectoral and macroeco- tributes less than it should to institution building
nomic policies such as pricing policies in the agri- in the recipient countries. In recent years attempts
cultural sector and land use policies in urban areas. have been made to improve this aspect of aid, but
Moreover, even when projects to reduce poverty much remains to be done.
were successful on their own terms, many of them
were essentially "enclaves" with little influence on INADEQUATE PARTICIPATION. Donors and recipi-
national efforts to reduce poverty. ents have given too little attention to sociocultural
and political factors and have not been sufficiently
FADDISM. Frequent shifts in donor concerns and aware of the important role that the poor them-
policies have often overshadowed the need for sta- selves can play in initiatives designed to assist
ble long-term strategies. A study of six Sub- them. Evidence supports the view that involving
Saharan African countries concluded that changes the poor in the design, implementation, and evalu-
of direction were one of the main reasons that aid ation of projects in a range of sectors would make
had not accomplished more. The study found that aid more effective. Involvement of women has
aid in the 1970s was characterized by "a single- contributed to the attainment of objectives in many
minded emphasis on food security and poverty al- agricultural development projects in Sub-Saharan
leviation that undermined export-crop production Africa; participation of local community organiza-
and neglected the technological factors affecting tions has improved performance in many urban
improvement in food-crop productivity," whereas poverty projects; organizations of beneficiaries in
in the 1980s there was "a sharp and simplistic aid-supported irrigation schemes have made im-
swing of donor attention toward efficiency and portant contributions to the maintenance and op-
away from equity concerns." These aid-induced eration of project works; and involvement of orga-
"zigzags" were apparent not only in agriculture nized groups of low-income borrowers has
but also in education and other sectors. They made facilitated repayment of loans in small-scale credit
projects more complicated, when success in attack- programs.
ing poverty called for sustained attention to a few
simpler matters. For instance, the contributions Toward a greater role for aid in poverty reduction
that basic infrastructure can make to reducing pov-
erty in rural areas have often been neglected in the This review of successes and failures provides im-
pursuit of a grander vision of "rural develop- portant lessons for future assistance. Although aid
ment." Faddism also leads to competition among will continue to serve many diverse development
donors for "good" projects. The result has often objectives, an assessment of countries' policies
been that too many donors are doing the same and their likely effects on poverty should play a
thing in the same country oreven worseare do- more important role in allocating aid resources. An
ing the same thing differently. This wastes effort aid strategy revised to make external assistance a
and means that the total impact of development more effective weapon against poverty should
aid is less than the sum of its parts. have as its centerpiece appropriately designed
support for countries that are pursuing policies in-
NEGLECT OF INSTITUTIONAL DEVELOPMENT. Re- tended to reduce poverty. This means looking at
ducing poverty through aid calls for more than the effects on the poor of the whole range of gov-
money; building capacity is crucial. Donors have ernment policiespolicies designed to expand the
unduly neglected the institutional and managerial income-earning opportunities of the poor (Chapter
aspects of poverty-oriented projects and pro- 4), extend social services (Chapter 5), effect income
grams. The reason is not want of funds; technical transfers and provide safety nets (Chapter 6), and
assistance accounts for more than a quarter of total implement structural reforms while incorporating
aid. Rather, much of this assistance fails to reflect a concern for the effects of these measures on the
borrowers' priorities. The main form of technical poor (Chapter 7).
assistance has been the supply of expatriate talent.
The United Nations Development Programme Aid allocation
(UNDP) estimates that donors support between
75,000 and 100,000 foreign experts a year in devel- Such a revised aid strategy has implications for the
oping countries. Donors often prefer to gather allocation of aid from both bilateral and multilat-
these experts in project units outside the normal eral sources. IDA, the World Bank's concessional
bureaucratic structure. The result is that aid con- lending affiliate, already extends about 95 percent
133
of its credits to low-income countries. Povertyits Country assistance strategies
extent and the efforts that individual countries are
making to reduce itis one of the principal criteria A revised aid strategy also has implications for the
in allocating these resources. In recent years coun- design of country assistance strategies and the use
tries that are judged to be doing too little to reduce of a variety of aid instruments. These would vary
poverty have received a smaller proportion of re- according to country circumstances; to a large ex-
sources from IDA. The approach recommended tent they would depend on the domestic policies
here would reinforce this trend. being pursued by recipient countries.
Past practice in the World Bank has tended to In countries in which policies are already consis-
focus on measuring the effect on poverty of partic- tent with the reduction of poverty, assistance to
ular projects (by attempting to assess, for example, the public sector can be delivered in three ways:
what proportion of project benefits accrues to the balance of payments support linked to the restruc-
poor). The approach recommended here is econo- turing of public expenditure, "time-slices" of pub-
mywide. The central question is the impact on the lic expenditure programs, and investment
poor of government policies as a whole. The most projects. The appropriate mix would depend on
recent negotiations for the ninth replenishment of the most pressing needs of individual countries.
IDA stressed the importance of poverty criteria in Aid for balance of payments support is especially
allocating the organization's resources, and broad important in debt-distressed low-income coun-
assessments of the effects on poverty of general tries. It is already a key feature of the Special Pro-
country policies increasingly receive special atten- gram of Assistance for debt-distressed countries in
tion in the economic and sector work undertaken Sub-Saharan Africa that are undertaking adjust-
by Bank economists. These efforts should be ment programs. Support for time-slices of public
broadened and intensified so that external assis- expenditure programs is also under way in a num-
tance from the Bank and other donors can more ber of casesfor example, in the Bank's support
effectively support the kinds of policies advocated for the education sector in Morocco. If the pro-
in this Report. posed aid strategy is to be effective, disbursements
In assessing the overall impact on poverty of for both kinds of support would have to be based
country policies and in monitoring their imple- on agreed changes in the structure of public
mentation, it is necessary to avoid an excessively spending.
mechanistic approach. Decisions wifi have to be A particular use of general budgetary assistance
made on the basis of informed judgments concern- is contingency funding to support safety nets. Ex-
ing the extent to which governments' policies are ternal shocks and natural disasters often coincide
appropriate for reducing poverty. Other govern- with (or cause) fiscal difficulties. At precisely the
ment and donor objectives must also be taken into time that the need for transfers to the poor is great-
account. Both the assessment and the monitoring est, governments may be least well positioned to
of the influence of government policies on the poor supplement or even maintain existing programs.
would be greatly facilitated, however, by more sys- There is a good case for increased external assis-
tematic compilation of social and income indicators tance at such times. Emergency food aid already
that could measure progress in implementing the serves this purpose during droughts and famines
two-part approach outlined in this Report. For ex- (Box 8.7). When governments have in place effec-
ample, data on per capita expenditures on primary tive safety nets, such as public works schemes, this
education and nutrition programs could help to basic idea could be extended to cover other shocks.
measure progress in providing social services to Thus, aid programs could include agreements to
the poor. On the income side, it should be feasible the effect that external assistance would be in-
to collect on a systematic and regular basis data on creased in the event of adverse macroeconomic
producer prices for small farmers, wages of daily shocks to provide temporary relief while efforts to
agricultural laborers, urban unskilled wages, and restructure the economy are put in place.
relevant cost of living indices. These would com- Free-standing investments should also continue
plement direct assessments of the status of the to play a central role. Such investments will be
poor, such as those obtained from household ex- vital for building and maintaining basic infrastruc-
penditure surveys. The World Bank has collected ture and for undertaking projects in the social sec-
survey data in many countries through its Living tors. As discussed in Chapter 4, however, the
Standards Measurement Studies, and additional scope for specific investments will vary between,
surveys are in preparation. at one end of the range, areas with high productive
134
Box 8.7 Food aid and poverty
In 1989 food aid to developing countries amounted to also help to finance measures to protect the standard of
approximately $2.5 billion. Food aid constituted 5 to 6 living of the poor during adjustment.
percent of bilateral aid from all DAC members; it repre- Food aid is also provided for emergencies and such
sented 12 percent of Canada's aid program and 18 per- specific purposes as food-for-work or supplementary
cent of total U.S. assistance. The main multilateral pro- child feeding projects ("project" food aid). An issue
vider is the World Food Programme, which accounts with these forms of food aid is whether they are effec-
for about 21 percent of all food aid from United Nations tive in providing transfers and safety nets to the poor-
sources. Another significant donor is the European est groups. Ensuring food security for the poor is the
Community; in 1989 it contributed 10 percent of all central objective of such assistance, but this does not
food aid. necessarily entail physically supplying poor people
A strong argument on behalf of food aid is that much with food. Simply making more food available in a
of it represents U.S. and EC production surpluses that country can help protect the poor by stabilizing prices,
would not otherwise benefit developing countries. and food aid can be used to finance effective poverty-
Since food aid is likely to continue to be a significant reducing programs that do not involve food transfers.
component of foreign assistance, it is important that it Direct distribution can, of course, be important in
be an effective instrument for reducing poverty. At the emergency situations where there is a large food deficit
global level allocation of food aid (as of aid in general) and markets are not functioning well. Examples in-
could be based more strongly on poverty criteria, as clude the famines in Ethiopia, Somalia, and Sudan as
opposed to political considerations. Reform of domes- well as the severe floods in Bangladesh in 1988, when
tic policies can also greatly improve the effectiveness of the World Food Programme coordinated 400,000 tons
this resource in reaching and assisting the poorest of food aid through a quick-response capacity devel-
within a given country. oped in dealing with previous floods.
Most food aid is provided for general balance of pay- Monetization of food aid through sale in the recipient
ments support ("program" food aid). Questions about country is often appropriate and can greatly improve
the effectiveness of program food aid mainly concern the poverty-reducing potential of this form of assis-
the appropriateness and efficiency of public sector ex- tance. The cash generated can be used to finance such
penditures. Given good administration and appropri- domestic programs as rural employment schemes. Pro-
ate policies, program food aid can free resources for grams of this kind are able to reach the poorest rapidly
poverty reduction and other development efforts with- and effectively. By relying on regular market channels
out interfering with the functioning of local food mar- to bring food into deficit areas, they avoid production
kets. In India, for example, funds generated by the sale disincentives and high internal distribution costs. In
of food aid, together with the associated foreign ex- addition to providing direct assistance, monetized food
change savings, accounted for 30 to 40 percent of gov- aid can serve as a resource for investments in human
ernment revenue in the mid-1960s. Some of these re- and physical capital and so help to bring about sus-
sources were used to create the infrastructure for the tained improvements in food security and in the well-
Green Revolution, which in turn led to considerable being of poor people.
increases in food production. Program food aid can
potential and, at the other, resource-poor regions private investment. It will also continue to have a
that have little productive potential but many poor role in more directly promoting labor-intensive ac-
people. tivities in the private sector.
Since the private sector is central to growth in The provision of aid through these instruments
labor demand, an aid strategy should support this cannot, however, be recommended for those coun-
sector. But aid will be effective only if incentives tries whose policies are not conducive to meaning-
are appropriate. An evaluation of the broad frame- ful poverty reduction. Yet there are large numbers
work of incentives must form an integral part of of poor people in such countries. In these cases
the overall assessment of country strategies. Bal- difficult decisions will have to be confronted. Di-
ance of payments support can also be important recting limited quantities of aid in a highly targeted
hereprovided that it is linked to macroeconomic fashion toward the poorest groups would seem to
management that facilitates adequate private in- be the appropriate response. For example, aid
vestment. Aid can be used to reduce the demand could support health clinics that serve poor
placed on the banking system by the public sector, women and children, immunization programs for
and this in turn will ease financial pressures on children, and well-targeted feeding programs.
135
Much of the work of agencies such as UNICEF and At $51 billion, the total volume of aid in 1988
OXFAMin Ethiopia, Kampuchea, Sudan, and might seem large in absolute terms. But it is small
elsewhereis precisely of this kind. The World when viewed in a wider perspectiveespecially
Bank is also undertaking such work, in Zaire, for when it is recalled that a substantial part of this
example. The goal must be to protect the welfare of amount, mainly that from bilateral sources, is not
the poor as far as possible while efforts to reform oriented toward development. Total aid from DAC
country policies continue. But large volumes of aid members amounts to only 0.36 percent of their
should not go to such countries until policies more combined GNP. Moreover, the development assis-
consistent with the reduction of poverty are put in tance provided by industrial country donors in re-
place. cent years has been only about 5 percent of their
Many countries present intermediate cases, and military expenditures. These donors spent $31.6
the choices for deploying aid are not so well de- bfflion for aid and $666 bfflion for military purposes
fined. Again, informed judgments will have to be in 1986. In recent years aid has represented only
made on what sorts of initiatives can be effectively about 1.4 percent of central government expendi-
supported by development assistance in such tures for DAC members (and only about 0.8 per-
countries, if, for example, a country is not provid- cent for the United States). A substantial increase
ing adequate social services but is doing reason- in the resources for fighting poverty in the poorest
ably well at increasing the incomes of the poor, aid countries appears entirely affordable. It is a matter
should be directed at improving social services, if of political commitment and the reassessment of
policies are not conducive to the productive use of donors' priorities.
poor people's labor but the provision of social ser- DAC projects an increase of only 2 percent a year
vices to the poor is satisfactory, aid can help to in real terms in its members' aid over the medium
maintain the stock of physical capital while efforts term. This would yield aid of $64 billion in 2000,
to encourage growth-promoting policies continue. but it means that the growth of official develop-
So, some assistance to intermediate cases would ment assistance would be slower than the growth
be justified. But the disbursement of substantial of members' GNP, which is estimated at about 3
volumes of aid should generally be confined to percent in this Report. The ratio of aid to GNP
countries that are pursuing appropriate policies would decline even further.
designed to generate income-earning opportuni- This chapter has stressed that in the longer term
ties and are providing social services efficiently to the progress of developing countriesincluding
the poor. their progress in reducing povertywill demand
substantially more reliance on the countries' own
Implications for the volume of aid domestic resources and on their income from for-
eign trade and considerably less dependence on
This strategy implies that an undifferentiated case external official support through aid. But economic
for more aid cannot be madewhether aid should self-reliance is stifi far off for many poor countries.
be increased depends on how many countries are Real growth in aid of only 2 percent a year is an
seriously pursuing the reduction of poverty. Nor unacceptably weak response to the challenge of
can it be argued unequivocally that aid should be global poverty. The international community
reallocated from less poor to poorer countries, al- needs to do bettermuch better. At a minimum, it
though the case for disbursing large volumes of aid should ensure that aid does not fall as a proportion
to middle-income countries is weak. Such a reallo- of donors' GNE A 3 percent annual increase in aid
cation would again depend on country policies. between now and 2000 would produce a total
There are two important reasons, however, for amount of $73 billion in 2000. Preferably, donors
believing that a more substantial volume of aid will should aim much higher. Aid could reach $108 bil-
be required in the short- to medium-term. First, lion in 2000 if donors with aid-to-GNP ratios of less
the outlook for exports and debt relief is not good than 0.5 increased their aid to that proportion and
for many low-income countries. Second, and more those with ratios of more than 0.5 maintained their
encouragingly, more countries are adopting poli- current performance. Attainment by all donors of
cies consistent with the priorities urged in this Re- the widely accepted international target of ODA as
port. These include the countries eligible for the 0.7 percent of GNP (with those donors having al-
Special Program of Assistance to Sub-Saharan Af- ready attained the target remaining at their current
rica, and there is guarded optimism that other aid ratios) would increase aid to $144 billion. More-
countries may join their ranks. over, if increases in aid volume were accompanied
136
by significant reallocations toward countries that Donors must be prepared to supply substantial
have policies geared toward the reduction of pov- volumes of aid to countries that are serious about
erty, the effects on the external resources available reducing poverty. Recipients must increasingly
to those countries could be even more dramatic. demonstrate that seriousness. Together they can
The goal must be to make aid a more effective learn from past successes and failures and make
weapon in the war against global poverty. This aid a more effective instrument for reducing pov-
entails challenges for donors and recipients alike. erty in the future.
137
Prospects for the poor
In the coming decade 850 million people will be specific details on domestic policy are discussed
born in the developing world. The chance that an below, region by region. The projections are based
individual will be born poor and will grow up in on the relatively favorable assumptions about
poverty will depend mainly on the region of the global economic conditions described in Chapter
world in which he or she is born. But this does not 1growth in industrial countries of about 3 per-
mean that economic policy is unimportantfar cent a year, falling real interest rates, rising com-
from it. Policy choices by domestic governments modity prices over the decade, and a successful
and the international community can make a criti- conclusion to the trade talks at the Uruguay Round
cal difference for hundreds of millions of the poor. of the GATT and other forums. This is the Report's
This chapter examines the prospects for reducing assessment of the most likely outcome. But fear
world poverty over the next decade. remains that the problems of the 1980s will persist.
The projections should be interpreted, therefore,
Poverty at the end of the century as indicating what can reasonably be expected. It
would be possible to do somewhat betteror
This Report has emphasized a dual approach to much worse.
reducing poverty. The elements of this twofold Table 9.1 shows projected changes in two social
strategy are: indicators: under 5 mortality and primary school
Efficient labor-intensive growth based on ap- enrollment. The projections are based on long-
propriate market incentives, physical infrastruc- term trends in each region. Under 5 mortality is
ture, institutions, and technological innovation expected to decline throughout the developing
Adequate provision of social services, including world. In South Asia, for example, the rate is ex-
primary education, basic health care, and family pected to fall dramatically, to 98 per thousand. If
planning services. the experience of recent years continues, even
greater progress can be expected. Moreover, the
In addition, transfers are needed to help those developing world will have attained or nearly at-
who would not otherwise benefitthe extremely tained universal primary enrollment, although
destitute, the sick, and the aged, and safety nets some countries in South Asia and in Sub-Saharan
must be provided to protect those most vulnerable Africa show a significant lag.
to income-reducing shocks. In Sub-Saharan Africa a return to long-term
The projections in this chapter are intended to trends holds the promise of substantial progress.
show what might be achieved if the recommended Under 5 mortality is likely to fall but to remain well
strategy gained wider acceptance. They do not as- above 100. Similarly, the percentage of children at-
sume that all countries will fully adopt the strat- tending primary school will increase significantly,
egy. They do assume that where it is in place, although it will still be somewhat lower than in
countries will persevere and that where it is not, other regions. If these projections are to prove ac-
countries will at least move in that direction. More curate, increased spending on the social sectors
138
Table 9.1 Social indicators, by developing region, tinue, the projections of Table 9.1 would not be
1985 and 2000 realized.
Net primary
enrollment Under 5 Regional differences in poverty
ratio mortality
Region 1985 2000 1985 2000
Regional variations reflect differences in the provi-
Sub-Saharan Africa 56 86 185 136
sion of social services, in the rate and pattern of
East Asia 96 100 54 31
China 93 95 44 25 economic growth, and in population growth. They
South Asia 74 88 150 98 also reflect differences in the structure of poverty
India 81 % 148 94 and in administrative capacity for dealing with its
Eastern Europe 90 92 25 16 problems.
Middle East and
North Africa 75 94 119 71
Latin America East Asia
and the Caribbean 92 100 75 52
Total 84 91 102 67 East Asia has long followed the broad approach to
reducing poverty advocated in this Report. By 1985
Note: For under 5 mortality, regional figures are weighted averages,
1985 refers to 1985-90, and 2000 refers to 2000-05. the number of poor in the region had declined to
Source: For 1985, United Nations data and Table 5.1; for projections, 280 mfflion, most of whom were in China, and by
World Bank estimates.
the end of the decade it is projected to fall to about
70 million. Growth of GDP in the region should
will be required. A repetition of the experience of remain strong, at almost 7 percent a year. The
the 1980s would imply much higher mortality countries are expected to continue to provide infra-
among children and lower enrollment rates. structure and appropriate incentives for efficient,
Expanded provision of social services for the labor-intensive manufacturing, to maintain do-
poor needs to go hand in hand with a greater em- mestic terms of trade that encourage agricultural
phasis on growth that makes productive use of growth, and to allocate a substantial share of gov-
labor. Table 9.2 shows what the two together might ernment spending to investment in human capital.
achieve. Between 1985 and 2000 the incidence of (The Philippines may be an exception; as in the
poverty in the developing world would fall from 33 past, macroeconomic imbalances and an unfavor-
percent to 18 percent and the number of poor from able distribution of land may lead to slow growth
1.1 billion to 825 million. This would be remarkable and prolonged underemployment.)
progress by many standardsbut one-seventh of Several factors, however, sound a cautionary
humanity would still be living lives of acute depri- note. First, since China accounts for three-quarters
vation. Moreover, some regions will progress of the region's population, any setback there will
much faster than others. In Sub-Saharan Africa,
the number in poverty wifi rise by 85 million, to
265 million by the end of the century. The distribu- Table 9.2 Poverty in 2000, by developing region
tion of global poverty will shift dramatically. Asia's
Number of poor
share of the world's poor will decline to 53 percent Incidence of poverty (millions)
from 72 percent in 1985; Sub-Saharan Africa's will Region 1985 2000 1985 2000
double, from 16 to 32 percent. Sub-Saharan Africa 46.8 43.1 180 265
If the global economy performs less favorably East Asia 20.4 4.0 280 70
than forecast, the level of world poverty wifi be China 20.0 2.9 210 35
considerably higher. A slowdown in industrial South Asia 50.9 26.0 525 365
country growth, persistently high real interest India 55.0 25.4 420 255
Eastern Europe 7.8 7.9 5 5
rates, rising protectionism, or setbacks in the cur- Middle East,
rent debt reduction initiatives could all seriously North Africa,
undermine progress. In that case, by 2000 the and other Europe 31.0 22.6 60 60
number of poor would still be hovering around 1 Latin America and
bfflion. In most regions the expected improvement the Caribbean 19.1 11.4 75 60
in social indicators should prove less vulnerable to Total 32.7 18.0 1,125 825
external events, but, as in the 1980s, Sub-Saharan Note: The incidence of poverty is the share of the population below
Africa may be an exception. If the deterioration in the poverty line, which is Set at $370 annual income (the higher line
used in this Report).
services that occurred in the 1980s were to con- Source: For 1985, Table 2.1; for 2000, World Bank estimates.
139
have a significant effect on regional poverty. The lems than most other regions, and its social and
projected 6.8 percent growth rate of GDP in the physical infrastructures are seriously debilitated.
1990s assumes that internal reforms will continue. In addition, the region's population is projected to
The projected drop in the number of poor also de- continue growing at more than 3 percent a year for
pends on maintaining the favorable distribution of the next decade. Such rapid growth exacerbates
income that has marked China's development. An the difficulty of eliminating poverty by undermin-
erosion of the agricultural terms of trade or a fail- ing efforts to increase labor income and increasing
ure of lagging regions to join in growth could com- the cost of expanding social services.
promise China's overall progress, especially in ru- The sharp rise in poverty projected for Sub-
ral areas. At the same time, greater reliance on Saharan Africa distinguishes it starkly from other
market forces and decentralization could further regions. The 3.7 percent annual growth in GDP
undermine the community-level system of health that underlies the projections is somewhat higher
care and social security. The challenge that China than the average growth achieved over the pre-
faces is to encourage gains in efficiency through vious twenty years, but that will be barely ade-
market reform while maintaining or replacing so- quate to hold living standards steady over the next
cial safety nets threatened by reform. If external ten years. In that time span the population will
conditions are unfavorable and internal reforms grow by another 165 million, and an additional 70
are not implemented, annual growth is unlikely to million people will be living in poverty.
exceed 5 percent in the 1990s. This would leave The plight of Sub-Saharan Africa requires a com-
China with 90 million poor by 2000. A fall in the mitment to fundamental changes in domestic poli-
real incomes of the rural poor (caused, for in- cies and development priorities. During the sec-
stance, by a shift in relative prices) could easily ond half of the 1980s many countries introduced
boost this number to more than 100 million. major reform programs. These need to be contin-
A second reason to temper optimism about the ued and strengthened. In particular, there must be
region's poor is that the pattern of poverty is a continued commitment in three critical areas: re-
changing. Rising incomes and strong employment ducing regional fertility rates, expanding social in-
growth in the middle-income countries of East vestment (especially in education and health care),
Asia have meant that the poor are increasingly and rehabilitating the physical infrastructure
confined to those groups that are unable to benefit needed to encourage the expansion of agriculture,
from employment opportunities and rising real commerce, and industry. Many of the policies for
wagesnotably, the elderly and the infirm. Demo- achieving this objective were described in a recent
graphic and social forces are also increasing the World Bank report, Sub-Saha ran Africa: From Crisis
need for state provision of health care and safety to Sustainable Growth. The analysis of future pov-
nets for these groups. Economies such as Korea erty presented here strengthens that report's con-
and Malaysia have the resources and administra- clusions.
tive capacity to put social assistance schemes in To see the scale of the task confronting Sub-
place, but a higher share of domestic transfers will Saharan Africa, consider what it would take to pre-
be needed for this purpose in the future. vent the number of poor from increasing. Even
Finally, Southeast Asia, in particular, is vulnera- with a supportive policy environment that in-
ble to protectionism in its export markets. In these cludes greater provision of social services and
countries exports are an exceptionally large share some improvement in income distribution, a
of GNP. In several product areas their potential for growth rate of about 5.5 percent a yearnearly 2
further growth has run up against industrial coun- percent higher than the projected ratewould be
try barriers to imports of, for example, garments, needed to raise per capita consumption by enough
shoes, and electronic goods. For East Asia more to meet this target. To achieve this, the region
than any other region, growth prospects and do- would need much more, and better designed, for-
mestic employment opportunities will be im- eign assistance. Such assistance will be effective
proved by further trade liberalization under the (and warranted) only if domestic policy is consis-
GATT tent with the reduction of poverty and the devel-
opment of indigenous capacity.
Sub-Saha ran Africa
South Asia
Sub-Saharan Africa lies at the other extreme. It has
the furthest to go in adopting the strategy pro- In South Asia there is considerable potential for
posed here, it has faced more severe external prob- progress toward reducing poverty, particularly in
140
India. But there is also a threat of stagnation, espe- ucation. By contrast, Pakistan has a sizable manu-
cially elsewhere in the subcontinent. Close to half facturing sector, ample administrative skills, and a
the world's poor live in the region, and what hap- per capita income twice that of Bangladesh. De-
pens to them will go far toward determining the spite these advantages, it has made little progress
success or failure of efforts to reduce world pov- toward improving the social indicators of poverty.
erty. Per capita GDP is still low, but it grew at a Reducing poverty requires a much higher level of
steady 3 percent during the 1980s and is likely to investment in social services for the poor.
continue at this rate through the 1990s. If this Aid has generally been used more efficiently in
strong performance can be maintained, the num- Asia than in Sub-Saharan Africa, but despite the
ber of poor is projected to fall from 525 million to vast numbers of poor, the flow of aid to the region
365 mfflion. is much smaller in relation to total population or
Poverty in India is projected to fall dramatically. GDP. One important conclusion of this Report is
The key to the gains in India will be sustained that aid allocations should be based on a demon-
growth and further internal policy reforms. Fiscal strated commitment to the goal of reducing pov-
reforms are needed to curb growing government erty. More aid should be forthcoming where seri-
deficits, contain foreign and domestic debt, and ous efforts are being made to reduce poverty,
ensure adequate savings to support domestic in- whether in Africa or in Asia, and the increase
vestment. Servicing domestic and foreign debt ac- should not come at the expense of the other re-
counts for a growing share of government spend- gion. Additional aid from industrial countries will
ing and may threaten the government's ability to be needed.
meet its investment targets. If investment remains
squeezed by low domestic saving rates and exter- Latin America and the Caribbean
nal borrowing, India's growth is unlikely to exceed
2 percent per capita a year over the decade. Such Nowhere in the developing world are the contrasts
slippage would imply a poverty count of 370 mil- between poverty and national wealth more strik-
lion people by 2000, eroding much of the potential ing than in Latin America and the Caribbean. De-
gain. spite average per capita incomes that are five to six
Additional domestic measures will be needed in times those in South Asia and Sub-Saharan Africa,
the fight against poverty. Further trade liberaliza- nearly one-fifth of the population still lives in pov-
tion would encourage exports and allow efficient, erty. This is because of the region's exceptionally
labor-intensive import substitution. Agriculture high degree of income inequality. Raising all the
needs access to new technology, and rural infra- poor in the continent to just above the poverty line
structure must be expanded and maintained, par- would cost only 0.7 percent of regional GDPthe
ticularly in lagging subregions. Exports are pro- approximate equivalent of a 2 percent income tax
jected to become a more important source of on the wealthiest fifth of the population.
growth as public spending is trimmed back. Regional prospects for reducing poverty rest
Economywide policies must be supplemented heavily on domestic policy reforms in several ar-
with better-targeted interventions. Public works eas. Credible macroeconomic stabilization mea-
schemes, for example, have proved successful and sures are of primary importance. Stabilization is a
could be extended and developed. prerequisite for restoring investor confidence, en-
The prospects for other countries are bleaker, couraging the return of flight capital, and breaking
and strong measures are needed to prevent pov- the cycle of economic crises that has characterized
erty from deepening significantly over the decade. many countries in the region. Policies to promote
At the projected growth rates, the number of poor inequality-reducing growth are equally important.
in Bangladesh and Pakistan would increase. Al- First, reforms to remove biases that favor the use
though Bangladesh is endowed with fertile land of capital are needed to ensure that future growth
and natural resources, in other respects it resem- will generate productive employment that reaches
bles Sub-Saharan Africa. Lack of administrative ca- the poor. This includes reforms in price and finan-
pacity, weak infrastructure, rapid population cial policy to encourage private investment in effi-
growth, and highly distorted domestic policies cient labor-intensive and outward-oriented indus-
hamper the reduction of poverty. For external as- tries. Moreover, since 40 to 50 percent of the poor
sistance to have a significant impact, fundamental wifi still reside in rural areas in the 1990s, the im-
efforts are needed to improve aid administration provements in incentives for agricultural produc-
and to direct resources toward improving preven- tion that occurred in the 1980s need to be comple-
tive health care, nutrition, sanitation, and basic ed- mented by active support for rural development.
141
Second, there is a need to maintain and in some dence of poverty is expected to remain roughly
areas to expand the provision of social services to constant at about 8 percent of the population. Two
the poor. Finally, transfers, such as emergency em- important issues face the region's governments.
ployment schemes of the type undertaken in Bo- One is the speed of reform, which will determine
livia, Chile, and Peru, may be needed to protect the severity of social dislocation and the time re-
the poor, particularly during the stabilization quired before market reforms boost productivity.
process. The second is the tradeoff between the reduction
The average GDP growth of 4.2 percent a year of fiscal deficits and state interventionsan essen-
projected for Latin America in the 1990s assumes tial step if markets are to flourishand the need to
that during the second half of the decade the debt maintain state-provided safety nets. A significant
burden will no longer be a serious constraint on part of the work force is likely to be dropped from
regional investment and that programs to restore state employment rolls at the same time as subsi-
economic stability will be put in place over the next dies on food, housing, and services are cut back.
few years. If, in addition, income distribution im- Temporary unemployment insurance and worker
proves with growth (as it did in Indonesia and training programs will therefore be needed.
Colombia, for instance, during the 1970s and The potential for raising regional output, pro-
1980s), significant progress can be made in reduc- ductivity, and labor incomes over the medium and
ing regional poverty. The incidence of poverty is long run is enormous. if the institutional and legal
forecast to fall from 19 to 11 percent by 2000. transition to a market-based system can be man-
Adverse developments in the global economy, as aged quickly and technical and financial assistance
described in Chapter 1, would darken the outlook (including foreign investment) can be obtained
for growth and for reducing poverty in the region. from abroad, consumption growth could be much
if efforts to ease the debt burden failed, terms of higher than in the projections. With a broadly un-
trade grew worse, and world demand for the re- changed distribution of income, growth in GDP of
gion's exports fell, the result would be to slow 4 percent a year over the decade would eliminate
growth and probably undermine adjustment ef- absolute poverty.
forts. Under these circumstances growth in per
capita consumption would not rise above 1 per- Middle East and North Africa
cent, and an additional 25 million people, or 5 per-
cent of the region's population, would be in pov- Heavy state intervention in domestic markets has
erty by 2000. To avert this outcome, political also characterized the countries of the Middle East
commitment to adjustment programs in the region and North Africa, but the challenges facing them
is needed, along with adequate external assistance differ from those in Eastern Europe. Many North
and debt relief to prevent the debt crisis from per- African countries have relied on food subsidies to
sisting through the 1990s. help the poor, despite ample evidence that subsi-
dies are an extremely inefficient form of transfer. In
Eastern Europe Algeria, Egypt, Morocco, and Tunisia food subsi-
dies account for a substantial share of the govern-
The condition of the poor in Eastern Europe is un- ment budget. More cost-effective means of provid-
usual because of the state's large role in providing ing transfers to the truly poor are needed. Several
employment, housing, and other services. Poverty countries also have severely distorted domestic
is largely an urban problem associated with low markets, persistent current account imbalances
real wages and, increasingly, with outright unem- (caused by unsustainable fiscal and exchange rate
ployment. In several countries poverty increased policies), and high debt-to-exports ratios. The pro-
during the 1980s. Structural problems accompa- longed regional conflicts in the Middle East have
nied by low productivity growth and chronic mate- been extremely costly and have diverted resources
rial shortages have caused a drop in real wages for from investment and the needs of the poor. Until
a large part of the work force. Although poverty is these conflicts are fully resolved, prospects for the
already serious and threatens to deepen further poor will remain bleak.
during the transition to a market-based system, Assuming that peace comes to the region in the
the number of poor in Eastern Europe is much 1990s and that structural adjustment programs are
lower than in Sub-Saharan Africa and South Asia. put in place, growth in per capita GDP should av-
Per capita GDP growth over the next decade is erage 2.1 percent a year. Despite high population
projected at only 1.5 percent a year, and the mci- growth (projected to remain at more than 2 percent
142
a year), modest improvements in income distribu- and would leave many more people in poverty in
tion and more cost-effective targeting of the poor 2000. Second, a failure to undertake policy re-
could hold the number of poor to about 60 mfflion forms, to reduce income inequality, or to safeguard
in 2000, the same as in 1985. On less favorable social services and real incomes where progress
assumptions, the number of poor could rise to has already been made would sharply reduce the
about 85 million. potential gains. Together, these might stifle all
progress toward reducing poverty over the coming
The potential for action decade. Third, the projections show that between
1985 and 2000 the number of poor in Sub-Saharan
The decade started with momentous changes in Africa will rise by perhaps 85 million. Only
the U.S.S.R. and in many Eastern European coun- through exceptionally bold action by the interna-
tries. These may point the way toward a world less tional community and the governments of the re-
gripped by superpower rivalry and more devoted gion can this be avoided.
to improving the quality of life. This Report has Preventing the number of poor from rising in
discussed a quieter but equally momentous Sub-Saharan Africa while reducing the number
changethe steady movement of households out elsewhere in the developing world by nearly 400
of poverty. Public action has often been immensely million (as compared with 1985) is an ambitious,
effective in supporting this process, but more than but achievable, target for the end of the century. In
a bfflion people are still poor. In some parts of the many areas the political and economic obstacles
world the number in poverty increased in the are daunting. Effective action to help the poor in-
1980s. volves some costs for the nonpoor in both devel-
The projections presented in this chapter are in- oped and developing countries. But these costs are
evitably uncertain, but they do illustrate the bene- modest even in the short term, and they are mas-
fits that can flow from appropriate public action. sively outweighed by the advance in human wel-
Three conclusions are especially clear. First, a less fare that a sustained attack on poverty would
buoyant external environment would place a bring.
greater strain on resources in developing countries
143
Bibliographical note
This Report has drawn on a wide range of World van Wijnbergen. Box 1.2 is based on Dornbusch
Bank reports and on numerous outside sources. and Edwards 1989, Sachs 1989, and the back-
World Bank sources include ongoing research as ground paper by Roubini. Box 1.4 draws mainly on
well as country economic, sector, and project van Wijnbergen 1990 and Box 1.5 on Borrell and
work. The principal sources for each chapter are Yang 1990. Ajay Chhibber, Alan Gelb, and Paul
noted below. These and other sources are then Meo provided valuable comments.
listed alphabetically by author or organization in
two groups: background papers commissioned for Chapter 2
this Report and a selected bibliography. The back-
ground papers, some of which will be available Numerous World Bank, United Nations, and aca-
through the Policy, Research, and External Affairs demic studies on poverty were consulted for the
Working Paper series, synthesize relevant litera- chapter. The three household profiles were in-
ture and Bank work. The views they express are spired by Hartmann and Boyce 1983, Tremblay and
not necessarily those of the World Bank or of this Capon 1988, and household survey data from the
Report. World Bank's Living Standards Measurement
In addition to the principal sources listed, many Study project, with help from Paul Glewwe and
persons, both inside and outside the World Bank, Dean Joliffe. Gabriela Vega also provided useful
helped with the Report. In particular, the core suggestions. Box 2.1 is based on the background
team wishes to thank Ravi Kanbur, Michael Lip- paper by Anand and Harris and on Haddad and
ton, Martin Ravallion, and T. N. Srinivasan for Kanbur 1989 and Ravallion 1988. Box 2.2 draws on
their extensive support. Others who provided the literature on poverty measurement, particu-
notes or detailed comments include Surjit Bhalla larly Sen 1981, with policy applications from
and Herman van der Tak. Ravallion and van de Walle 1988. The discussion in
Box 2.3 is based on Collier and Lal 1986, Collier
Chapter 1 and others 1986, Drèze forthcoming, and Greer
and Thorbecke 1986. Box 2.4 was prepared by Mar-
This chapter draws principally on IMF, OECD, and tin Ravallion and Apparao Katikineni from data
World Bank sources. Projections and background files provided by ICRISAT, India. Gaurav Datt and
analysis were prepared by the International Eco- Etienne van de Walle made useful comments.
nomic Analysis and Prospects Division, Interna-
tional Economics Department, World Bank; Far- Chapter 3
doust and Dhareshwar 1990 was an important
background source. The preparation of special The discussion on changes in poverty draws on
forecast scenarios was assisted by Robert Lynn, World Bank sources and on work by Elaine K.
Christian Pedersen, and Karsten Pedersen. The Chan on poverty decomposition and income distri-
sections on debt draw mainly on World Bank 1989f bution. Box 3.1 is based on Keyfitz 1985; and Box
and on discussions with Constantijn Claessens, 3.2 on Bevan, Collier, and Gunning 1988, Collier
Charles Humphreys, Ishrat Husain, and Sweder forthcoming, Collier and others 1986, Wagao 1986
144
and material from Douglas Rimmer; Box 3.4 on the Augusta Molnar, and Robert Repetto provided
background paper by Williamson and Polak. Wil- comments. Box 4.1 is based on Binswanger 1989
ham Ascher made substantial contributions to Box and Box 4.2 on Ahmed and Hossain 1988. Box 4.3
3.5. Box 3.6 and the discussion of critical tradeoffs draws on material prepared by Elizabeth King. Box
are based on the background paper by Bourgui- 4.4 is based on background notes by Friedrich
gnon. Data for figure 3.4 were provided by Juan Kahnert and Oded Stark. Box 4.6 was prepared by
Luis Londoño. Peter Hazell and Shem Migot-Adholla and draws
on Blare! and Place 1990. Box 4.7 draws on material
Chapter 4 prepared for World Bank 1989c. Box 4.8 is based on
Singh and Bara 1988, Box 4.9 on Kahnert 1989, and
This chapter draws extensively on World Bank Box 4.10 on Piazza and Doolette 1990. Douglas
sources and on the Bank's operational experience. Barnes, David Beckmann, Clive Bell, Hans
The section on government policy and rural devel- Binswanger, Peter Hazell, Barbara Herz, Peter
opment is based on the background paper by Pa- Hoperaft, Friedrich Kahnert, Robert Liebenthal,
panek; on Adelman 1984, Booth 1989, Cavallo and Luis Riveros, Ricardo Silveira, Roger Slade, and
Mundlak 1982, de Janvry and others 1989, Norman Uphoff provided valuable comments on
Krueger, Schiff, and Valdés 1988, and Schiff and the chapter.
Valdés forthcoming; on Lele 1989 and other mate-
rials from the Bank's project on Managing Agricul- Chapter 5
tural Development in Africa; and on material pro-
vided by Jean-Jacques Dethier. The discussion of This chapter draws heavily on World Bank experi-
infrastructure and technology draws on Anderson, ence in education and health. The section on pri-
Herdt, and Scobie 1988, Binswanger 1990, mary education is based mainly on Lockheed, Ver-
Binswanger, Khandker, and Rosenzweig 1989, spoor, and others 1990 and on World Bank 1986b.
Hayami and Ruttan 1985, Hazell and Ramasamy The discussion on government intervention in
1988, and Lipton and Longhurst 1989. Sources for health owes much to World Bank 198Th. Data on
the section on encouraging farm-nonfarm linkages under 5 mortality are from United Nations, De-
include Haggblade and Hazell 1989 and Hagg- partment of International Economic and Social Af-
blade, Hazell, and Brown 1989. The section on ur- fairs 1988 and Hill and Pebley 1988. The Westing-
ban and industrial growth benefited from de Soto house series of Demographic and Health Surveys
1989, International Labour Organisation 1985, (DHS) was a valuable source of information on
Krueger 1983, Lee 1989, Lee and Anas 1989, and child and maternal mortality, fertility, and family
Little, Mazumdai and Page 1987. The discussion planning indicators. Ralph Henderson made avail-
on access to land draws on the background paper able data generated by the Expanded Programme
on land rights by Bell, on Binswanger and Elgm on Immunization of the World Health Organiza-
1988, and on material provided by Peter Hazell and tion. Nutrition information is drawn from United
Peter Hoperaft. The section on access to credit is Nations 1987 and has benefited from suggestions
based on the background paper on credit by Bell by Alan Berg and John Mason. Nancy Birdsall pro-
and on material prepared by Sharon Holt; other vided material on demographic and family plan-
sources include Braverman and Guasch 1989, Hos- fling issues. Box 5.1 was inspired by work in
sam 1988, Huppi and Feder 1989, U.S. Agency for progress in the Population, Health, and Nutrition
International Development 1989, Von Pischke Division of the World Bank, under the direction of
1989, and Von Pischke, Adams, and Donald 1983. Anthony Measham. Box 5.2 derives from the back-
Lynn Bennett, Millard Long, Elizabeth Rhyen, ground paper by Lau, Jamison, and Louat. Box 5.3
J. D. Von Pischke, and Jacob Yaron provided de- was drafted by Ann Tinker and Box 5.4 by Eleanor
tailed comments. Material on access to infrastruc- Schreiber. The information for Box 5.5 was made
ture and technology came from the background available by the UNDP-World Bank Interregional
papers by Holt and by the International Fund for Handpump Project. Barbara Herz, Dennis de Tray,
Agricultural Development and from Cernea 1985, Emmanuel Jimenez, Oded Stark, and Adriaan Ver-
Korten 1980, Korten and Siy 1988, Lewis and oth- spoor provided detailed comments on the chapter.
ers 1988, and Uphoff 1986. The discussion on
resource-poor areas is based on material prepared Chapter 6
by Sharon Holt and on Lele and Stone 1989 and
Leonard and others 1989; John Doolette contrib- This chapter draws on a range of World Bank and
uted material, and Gloria Davis, Jeffrey Leonard, other sources and benefited from discussions with
145
people in and outside the Bank. The discussion of Bank. Francois Bourguignon, and Frances Stewart
fishing communities is based on a background provided valuable comments on the chapter. Fur-
note by Jean-Philippe Platteau and on Platteau ther assistance came from Perla Aizenman, Benoit
forthcoming. The section on food policies draws Blarel, Fernando Luis Quevedo, and Luis Riveros.
on Alderman forthcoming, Alderman and von
Braun 1984, Berg 1987, Edirisinghe 1987, Jamaica Chapter 8
Statistical Institute and World Bank 1988, Pinstrup-
Andersen 1988, Ravallion 1987, Sen 1981, United The discussion of trade and poverty draws on Fin-
Nations Children's Fund 1989, and World Bank ger and Messerlin 1989, Finger and Olechowski
1986c. Margaret Grosh provided information on 1987, Karsenty and Laird 1987, Laird and Yeats
Jamaican policies. The section on public employ- 1990, Tyers and Anderson 1986, United Nations
ment schemes draws mainly on Acharya and Conference on Trade and Development 1989,
Panwalkar 1988, Drèze and Sen 1990, and Raval- Whalley 1985, and Yeats 1981 and 1989 and on ex-
lion 1990. Box 6.1 is based on material from the tensive discussions with Refik Erzan, Sam Laird,
background papers by Atkinson and by William- Paul Meo, and Alexander Yeats. Essential data
son and Polak and on Howard 1943. Box 6.2 draws were obtained from United Nations Conference on
on Besley and Kanbur 1988, Ravallion 1990, and Trade and Development 1988. The discussion of
background notes by Jean Drèze. Boxes 6.3 and 6.5 the Multifibre Arrangement profited from the
are based on material in Drèze forthcoming and work of Erzan, Goto, and Holmes 1989 and Trela
Morgan forthcoming. Box 6.4 draws on informa- and Whalley 1988. The discussion of debt relies
tion in Acharya and Panwalkar 1988. Helpful com- primarily on Greene 1989, Humphreys and Un-
ments were provided by Alan Berg, Jean Drèze, derwood 1989, and World Bank 1989f. Materials of
Judith McGuire, Per Pinstrup-Andersen, Shlomo the Development Assistance Committee, espe-
Reutlinger, Nicholas Stern, Kalanidhi Subbarao, cially Organisation for Economic Co-operation and
and Joachim von Braun. Development 1985, 1989, and 1990, were the prin-
cipal sources on aid and poverty; Josefina G. Vale-
Chapter 7 riano provided invaluable research assistance.
Other published studies on aid that were ex-
The analysis of the relationship between macroec- tremely useful included Cassen and associates
onomic policy and the poor draws on the back- 1986, Lele 1989, Mosley 1987, Riddell 1987, and
ground papers by Ahmed and Peters; Berry; Bour- World Bank 1986a. Box 8.1 was prepared by Paula
guignon, de Melo, and Suwa; and Fox and Morley. Holmes and Paul Meo and Box 8.2 by Refik Erzan
Other sources were Beaudry and Sowa 1989, Ma- with the assistance of Paula Holmes. Box 8.3 draws
zumdar 1989, World Bank country work, and pa- mainly on Karsenty and Laird 1987. Box 8.4 owes
pers prepared for the project on Labor Markets in much to the OECD sources, especially 1989. Box
an Era of Adjustment sponsored by the University 8.5 is based largely on Ayres 1983 and World Bank
of Warwick and the Economic Development Insti- 1988c and Box 8.6 on Ayres 1983, World Bank 1983,
tute of the World Bank. The discussion of public and recent unpublished World Bank studies. Box
spending is based on the background paper by 8.7 profited from discussions with Robert Hindle,
Gallagher and the analysis of employment Shlomo Reutlinger, and Cornelis Tuinenburg.
schemes in Bolivia, Chile, and Peru on Newman David Beckmann, Michael Carter, and Robert
1988, Universidad de Chile 1987, and Graham Liebenthal provided valuable comments.
1990. The discussion of compensation for laid-off
workers draws on Collier 1988. Box 7.1 is based on Background papers
Moser 1989, Box 7.2 on the background paper by
de Janvry, Fargeix, and Sadoulet, and Box 7.3 on Ahmad, Ehtisham, and Christine Allison. "Pov-
the background paper by Milanovic. Box 7.4 is erty, Growth, and Public Policy in Pakistan."
drawn from Blejer and Guerrero forthcoming and Ahmad, Ehtisham, and Stephen Ludlow. "Pov-
Box 7.5 from African Farmer 1990. Box 7.6 was pre- erty, Inequality, and Growth in Pakistan."
pared by Stephen Haggard, drawing on Nelson Ahmad, Ehtisham, and Yan Wang. "Inequality
1990. Box 7.7 benefited from work for the Social and Poverty in China: Institutional Change and
Dimensions of Adjustment project undertaken by Public Policy, 1978-1988."
the African Development Bank, the United Na- Ahmad, Ehtisham, and Gang Zou. "Deprivation
tions Development Programme, and the World and Prosperity in Chinese History."
146
Ahmed, Sadiq, and R. Kyle Peters, Jr. "Adjust- Lau, Lawrence, Dean Jamison, and Frédéric Louat.
ment with Poverty Alleviation: Indonesia's Ex- "Education and Productivity in Developing
perience." Countries: An Aggregate Production Function
Anand, Sudhir, and Christopher Harris. "On the Approach."
Choice of Welfare Indicator in the Analysis of Milanovic, Branko. "Poverty in Eastern Europe in
Poverty: An Illustration Using Sri Lankan the Years of Crisis: Poland, Hungary, and Yugo-
Data." slavia."
Atkinson, Anthony B. "Poverty, Economic Perfor- "Poverty in Poland in the Years of Crisis,
mance, and Income Transfer Policy in OECD 1978-87."
Countries." Papanek, Gustav F. "Growth, Poverty, and Real
Bell, Clive. "Credit and Saving." Wages in Labor-Abundant Countries."
"Land Reform, Tenancy, Productivity, Pissarides, Christopher A. "Macroeconomic Ad-
and Employment as Aspects of Property Rights justment and Poverty in Selected Developed
in Land." Countries."
Bell, Clive, and Robert Rich. "Rural Poverty and Posarac, Aleksandra. "Poverty in Yugoslavia,
Agricultural Performance in India between 1956- 1978-87."
57 and 1983-84." Ravallion, Martin. "The Challenging Arithmetic of
Berry, Albert. "The Effects of Stabilization and Ad- Poverty in Bangladesh."
justment on Poverty and Income Distribution: Ravallion, Martin, Gaurav Datt, Dominique van de
Aspects of the Latin American Experience." Wale, and Elaine K. Chan. "Quantifying the
Bhalla, Surjit S. "The Role of Welfare Policies and Magnitude and Severity of Absolute Poverty in
Income Growth in Improving Living Standards the Developing World in the Mid-1980s."
in India and Sri Lanka." Roubini, Nouriel. "The Interactions between Mac-
Bourguignon, Francois. "Optimal Poverty Reduc- roeconomic Performance and Political Structures
tion, Adjustment, and Growth: An Applied and Institutions: The Political Economy of Pov-
Framework." erty, Growth, and Development."
Bourguignon, Francois, Jaime de Melo, and A. Szalai, Julia. "Poverty in Hungary during the Pe-
Suwa. "Distributional Effects of Adjustment Pol- riod of Economic Crisis."
icies: Simulations for Two Archetype Econo- Teklu, Tesfaye, and Joachim von Braun. "Coping
flues. with Disastrous Drought: The 1984/85 Famine in
Chan, Elaine K. "A Compendium of Data on Pov- North Kordofan, Sudan."
erty and Income Distribution." Wffliamson, Jeffrey G., and Ben Polak. "Poverty,
"Decomposing Changes in Poverty." Policy, and Industrialization: Lessons from the
de Janvry, Alain, André Fargeix, and Elisabeth Distant Past."
Sadoulet. "Economic, Welfare, and Political
Consequences of Stabilization Policies: A Gen- Selected bibliography
eral Equilibrium Approach."
Drèze, Jean P., Peter Lanjouw, and Nicholas Stern. The word "processed" describes works that are
"Identifying and Reaching the Poor: Principles reproduced from typescript by mimeography,
and Some Examples from a North Indian Vil- xerography, or similar means; such works may not
lage." be cataloged or commonly available through li-
Fox, M. Louise, and Samuel A. Morley. "Who braries, or may be subject to restricted circulation.
Paid the Bill? Adjustment and Poverty in Brazil,
1980-1995." Acharya, Meena, and Lynn Bennett. 1983. Women
Gallagher, Mark. "Fiscal Duress and the Social and the Subsistence Sector: Economic Participation
Sectors in Developing Countries." and Household Decisionmaking in Nepal. World
Hill, Kenneth. "Demographic Response to Eco- Bank Staff Working Paper 526. Washington,
nomic Shock." D.C.
Holt, Sharon L. "The Role of Institutions in Pov- Acharya, Sarthi, and V. C. Panwalkar. 1988. "The
erty Reduction: A Focus on the Productive Sec- Maharashtra Employment Guarantee Scheme:
tors." Impacts on Male and Female Labour." Prepared
International Fund for Agricultural Development. for the Population Council Program on Women's
"Poverty Alleviation: An IFAD Perspective." Roles and Gender Differences. Tata Institute of
Social Sciences, Bombay, India. Processed.
147
Adelman, Irma. 1984. "Beyond Export-Led Beaudry, P., and N. K. Sowa. 1989. "Labor Mar-
Growth." World Development 12, 9 (September): kets in an Era of Adjustment: A Case Study of
937-49. Ghana." Prepared for the Conference on Labor
African Development Bank, United Nations Devel- Markets in an Era of Adjustment, Coventry,
opment Programme, and World Bank. 1990. The U.K., August 7-10, sponsored by the Develop-
Social Dimensions of Adjustment in Africa: A Policy ment Economics Research Centre, University of
Agenda. Washington, D.C.: World Bank. Warwick, and the Economic Development Insti-
African Farmer. 1990. "Farmers Adjust to Economic tute of the World Bank. Processed.
Reform." Issue 3 (April): 5-16. New York: The Berg, Alan. 1987. Malnutrition: What Can Be Done?
Hunger Project. Lessons from World Bank Experience. Baltimore,
Ahmed, Raisuddin, and Mahabub Hossain. 1988. Md.: Johns Hopkins University Press.
"Developmental Impact of Rural Infrastructure: Besley, Timothy, and Ravi Kanbur. 1988. "The
Bangladesh." International Food Policy Re- Principles of Targeting." Discussion Paper 85.
search Institute, Washington, D.C., in collabora- Development Economics Research Centre, Uni-
tion with the Bangladesh Institute of Develop- versity of Warwick, Coventry, U.K. Processed.
ment Studies. Processed. Bevan, David, Paul Collier, and Jan Gunning.
Alderman, Harold. Forthcoming. "Food Subsidies 1988. "The Political Economy of Poverty, Equity
and the Poor." World Bank, Latin America Tech- and Growth: Indonesia and Nigeria." World
nical Department, Washington, D.C. In George Bank, Latin America Technical Department,
Psacharopoulos, ed., Essays on Poverty, Equity Washington, D.C. Processed.
and Growth. Bhende, M. J. 1986. "Credit Markets in Rural
Alderman, Harold, and Valerie Kozel. 1989. Formal South India." Economic and Political Weekly 21,
and Informal Sector Wage Determination in Urban 38-39 (September).
Low-Income Neighborhoods in Pakistan. Living Biggs, Tyler S., Donald R. Snodgrass, and Pradeep
Standards Measurement Study Working Paper Srivastava. 1990. "On Minimalist Credit Pro-
65. Washington, D.C.: World Bank. grams." Development Discussion Paper 331,
Alderman, Harold, and Joachim von Braun. 1984. Economic Policy Series. Harvard Institute for In-
The Effects of the Egyptian Food Ration and Subsidy ternational Development, Cambridge, Mass.
System on Income Distribution and Consumption. Processed.
Research Report 45. Washington, D.C.: Interna- Binswanger, Hans P. 1989. "Brazilian Policies That
tional Food Policy Research Institute. Encourage Deforestation in the Amazon." Envi-
Allison, Christine. 1989. "Poverty, Inequality, and ronment Department Working Paper 16. World
Public Policy in Pakistan." World Bank, Eastern Bank, Washington, D.C. Processed.
Africa Department, Washington, D.C. Pro- 1990. "The Policy Response of Agricul-
cessed. ture." In Proceedings of the World Bank Annual
Anderson, Jock R. 1985. Research and Agricultural Conference on Development Economics 1989. Wash-
Progress. Armidale, N.S.W, Australia: Univer- ington, D.C.: World Bank.
sity of New England. Binswanger, Hans P., and Miranda Elgin. 1988.
Anderson, Jock R., Robert W. Herdt, and Grant M. "What Are the Prospects for Land Reform?"
Scobie. 1988. Science and Food: The CGIAR and Its Prepared for the Twentieth International Confer-
Partners. Washington, D.C.: World Bank. ence of Agricultural Economists, Buenos Aires,
Ascher, William. 1984. Scheming for the Poor: The August 24-31. Processed.
Politics of Redistribution in Latin America. Cam- Binswanger, Hans P., Shahidur R. Khandker, and
bridge, Mass.: Harvard University Press. Mark R. Rosenzweig. 1989. "How Infrastructure
Atkinson, A. B. 1987. "On the Measurement of and Financial Institutions Affect Agricultural
Poverty." Econometrica 55: 749-64. Output and Investment in India." Policy, Plan-
Ayres, Robert L. 1983. Banking on the Poor: The ning, and Research Working Paper 163. World
World Bank and World Poverty. Cambridge, Mass.: Bank, Latin America and the Caribbean Country
MIT Press. Department II, Washington, D.C. Processed.
Beaton, George H., and Hossein Ghassemi. 1982. Binswagner, Hans P., and Jaime B. Quizon. 1986.
"Supplementary Feeding Programs for Young "What Can Agriculture do for the Poorest Rural
Children in Developing Countries." American Groups?" Agriculture and Rural Development
Journal of Clinical Nutrition 35 (April): 864-916. Department Discussion Paper ARU 57. World
Bank, Washington, D.C. Processed.
148
Birdsall, Nancy. 1980. "Population Growth and Bulatao, Rodolfo, and others. 1989. "Africa Region
Poverty in the Developing World." Population Population Projections: 1989-90 Edition." Policy,
Bulletin 35, 5 (December): 3-38. Planning, and Research Working Paper 330.
Birdsall, Nancy, and Estelle James. 1989. "Effi- World Bank, Population and Human Resources
ciency and Equity in Social Spending: How and Department, Washington, D.C. Processed.
Why Governments Misbehave." Policy, Re- Cassen, Robert, and associates. 1986. Does Aid
search, and External Affairs Working Paper 274. Work? Report to an Intergovernmental Task Force.
World Bank, Latin America and the Caribbean New York: Clarendon Press.
Country Department I, Washington, D.C. Pro- Castaneda, Tarsicio. Forthcoming. Innovative Social
cessed. Policies for Reducing Poverty: Chile in the 1980s.
Blarel, Benoit, and Frank Place. 1990. "Land Ten- Cavallo, Domingo, and Yair Mundlak. 1982. Agri-
ure and Agricultural Production under Land culture and Economic Growth in an Open Economy:
Scarcity: The Case of Rwanda." World Bank, The Case of Argentina. Research Report 36. Wash-
Agriculture and Rural Development Depart- ington, D.C.: International Food Policy Research
ment, Washington, D.C. Processed. Institute.
Blejer, Mario I., and Isabel Guerrero. Forthcoming. Cernea, Michael M., ed. 1985. Putting People First:
"The Impact of Macroeconomic Policies on In- Sociological Variables in Rural Development. New
come Distribution: An Empirical Study of the York: Oxford University Press.
Philippines." Review of Economics and Statistics. 1990. Population Displacement in Water Re-
Boateng, E. Oti, and others. 1989. "A Poverty Pro- sources Development: Current Practice, Issues, and
file for Ghana, 1987-88." World Bank, Occiden- Policy Responses. Harvard Institute for Interna-
tal and Central Africa Department, Washington, tional Development Discussion Paper. Cam-
D.C. Processed. bridge, Mass.
Booth, Anne. 1989. "Indonesian Agricultural De- Chambers, Robert. 1983. Rural Development: Put-
velopment in Comparative Perspective." World ting the Last First. London: Longman.
Development 17, 8 (August): 1235-54. Chenery, Hollis, and others. 1974. Redistribution
Borrell, Brent, and Maw-Cheng Yang. 1990. "EC with Growth. New York: Oxford University
Bananarama 1992." International Commodity Press.
Markets Paper. World Bank, International Eco- Chhibber, Ajay. 1986. "Financing Public Expendi-
nomics Department, Washington D.C. Pro- tures in Developing Countries: Direct versus In-
cessed. direct Taxation." Resource Mobilization Depart-
Boulier, Bryan L. 1985. Evaluating Unmet Needs for ment Working Paper. World Bank, Washington,
Contraceptives: Estimates for Thirty-six Developing D.C. Processed.
Countries. World Bank Staff Working Paper 678. Collier, Paul. 1988. "African Public Sector Re-
Washington, D.C. trenchment: An Analytical Survey." World Em-
Bourguignon, Francois, Wffliam H. Branson, and ployment Programme Research Working Paper
Jaime de Melo. 1989. "Adjustment and Income 27. International Labour Office, Geneva. Pro-
Distribution: A Counterfactual Analysis." Pol- cessed.
icy, Planning, and Research Working Paper 215. Forthcoming. "Aid and Economic Perfor-
World Bank, Country Economics Department, mance in Tanzania." In Uma Lele and Ijaz Nabi,
Washington, D.C. Processed. eds., Transitions in Development: The Role of Aid
Braverman, Avishay, and J. Luis Guasch. 1989. and Commercial Flows. San Francisco, Calif.: Insti-
"Rural Credit in Developing Countries." Policy, tute for Contemporary Studies.
Planning, and Research Working Paper 219. Collier, Paul, and Deepak Lal. 1986. Labour and Pov-
World Bank, Agriculture and Rural Develop- erty in Kenya, 1900-1980. Oxford: Clarendon
ment Department, Washington, D.C. Processed. Press.
Brown, L. David, and David C. Korten. 1989. "Un- Collier, Paul, Samir Radwan, and Samuel
derstanding Voluntary Organizations: Guide- Wangwe, with Albert Wagner. 1986. Labour and
lines for Donors." Policy, Planning, and Re- Poverty in Rural Tanzania: Ujamaa and Rural Devel-
search Working Paper 258. World Bank, Country opment in the United Republic of Tanzania. Oxford:
Economics Department, Washington, D.C. Pro- Clarendon Press.
cessed. Cornia, Giovanni Andrea, Richard Jolly, and Fran-
cis Stewart, eds. 1987. Adjustment with a Human
Face. 2 vols. Oxford: Clarendon Press.
149
Cox, Donald, and Emmanuel Jimenez. 1990. "Pri- Echeverri-Gent, John. 1988. "Guaranteed Employ-
vate Transfers and Public Policy in Developing ment in an Indian State." Asian Survey 28, 12:
Countries: A Case Study for Peru." Policy, Plan- 1294-1310.
ning, and Research Working Paper 345. World Edirisinghe, Nevile. 1987. The Food Stamp Scheme
Bank, Country Economics Department, Wash- in Sri Lanka: Costs, Benefits and Options for Modifi-
ington, D.C. Processed. cation. Research Report 58. Washington, D.C.:
Dandekar, Kumudini. 1983. Employment Guarantee International Food Policy Research Institute.
Scheme: An Employment Opportunity for Women. Erzan, Refik, Junichi Goto, and Paula Holmes.
Gokhale Institute Study 67. Bombay, India: Ori- 1989. "Effects of the Multi-Fibre Arrangement on
ent Longman. Developing Countries' Trade: An Empirical In-
Datt, Gaurav, and Martin Ravallion. 1990. "Re- vestigation." Policy, Planning, and Research
gional Disparities, Targeting, and Poverty in In- Working Paper 297. World Bank, International
dia." Policy, Research, and External Affairs Economics Department, Washington, D.C. Pro-
Working Paper 375. World Bank, Agriculture cessed.
and Rural Development Department, Washing- Erzan, Refik, and Peter Svedberg. 1989. "Protec-
ton, D.C. Processed. tion Facing Exports from Sub-Saharan Africa in
Deaton, Angus, and Anne Case. 1988. Analysis of the EC, Japan, and the United States." Policy,
Household Expenditures. Living Standards Mea- Planning, and Research Working Paper 320.
surement Study Working Paper 28. Washington, World Bank, International Economics Depart-
D.C.: World Bank. ment, Washington, D.C. Processed.
de Janvry, Alain, and Elisabeth Sadoulet. 1988. Evans, Timothy. 1989. "The Impact of Permanent
"Investment Strategies to Combat Rural Pov- Disability on Rural Households: River Blindness
erty: A Proposal for Latin America." Working in Guinea." IDS Bulletin 20, 2 (April): 41-48.
Paper 459. Giannini Foundation of Agricultural Fardoust, Shahrokh, and Ashok Dhareshwar.
Economics, University of California, Berkeley. 1990. "Long-Term Outlook for the World Econ-
Processed. omy: Issues and Projections for the 1990s." Pol-
de Janvry, Alain, and others. 1989. Rural Develop- icy, Research, and External Affairs Working Pa-
ment in Latin America: An Evaluation and a Pro- per 372. World Bank, International Economics
posal. IICA Program Paper 12. San Jose, Costa Department, Washington, D.C. Processed.
Rica: Instituto Interamericano de Cooperación Finger, J. Michael, and Patrick A. Messerlin. 1989.
para la Agricultura. The Effects of Industrial Countries' Policies on Devel-
de Soto, Hernando. 1989. The Other Path: The Invis- oping Countries. Policy and Research Series 3.
ible Revolution in the Third World. New York: Washington, D.C.: World Bank.
Harper and Row. Finger, J. Michael, and Andrzej Olechowski, eds.
Donaldson, Graham. 1989. "Government- 1987. The Uruguay Round: A Handbook on the Mul-
Sponsored Rural Development: Experience of tilateral Trade Negotiations. Washington, D.C.:
the World Bank." World Bank, Operations Eval- World Bank.
uation Department, Washington, D.C. Pro- Fitchett, Delbert A. 1988. Agricultural Trade Protec-
cessed. tionism in Japan: A Survey. World Bank Discus-
Dornbusch, Rudiger, and Sebastian Edwards. sion Paper 28. Washington, D.C.
1989. "Macroeconomic Populism in Latin Amer- Food and Agriculture Organization. 1987. The Fifth
ica." NEER Working Paper 2986. National Bu- World Food Survey. Rome.
reau of Economic Research, Cambridge, Mass. 1988. Rural Poverty in Latin America and the
Processed. Caribbean. Prepared for the FAO Regional Con-
Drabek, Anne Gordon, ed. 1987. "Development ference for Latin America and the Caribbean, Re-
Alternatives: The Challenges for NGOs." World cife, Brazil. Rome. Processed.
Development 15 suppi. (Autumn). Gallagher, Mark, and Osita M. Ogbu. 1989. "Pub-
Drèze, Jean P. Forthcoming. "Famine Prevention lic Expenditures, Resource Use and Social Ser-
in Africa." In Jean P. Drèze and Amartya K. Sen, vices in Sub-Saharan Africa." World Bank, Af-
eds., The Political Economy of Hunger. Oxford: Ox- rica Technical Department, Washington, D.C.
ford University Press. Processed.
Drèze, Jean, and Amartya Sen. 1990. Hunger and Galway, Katrina, Brent Wolff, and Richard Sturgis.
Public Action. Oxford: Clarendon Press. 1987. Child Survival: Risks and the Road to Health.
Columbia, Md.: Institute for Resource
Development/Westinghouse.
150
Garcia, Marito H., and Per Pinstrup-Andersen. Greene, Joshua. 1989. "The External Debt Problem
1987. The Pilot Food Price Subsidy Scheme in the of Sub-Saharan Africa." IMF Staff Papers 36, 4
Philippines: Its Impact on Income, Food Consump- (December): 836-74.
tion, and Nutritional Status. Research Report 61. Greer, Joel, and Erik Thorbecke. 1986. "Food Pov-
Washington, D.C.: International Food Policy Re- erty Profile Applied to Kenyan Smallholders."
search Institute. Economic Development and Cultural Change 35 (Oc-
George, P. S. 1985. Some Aspects of Procurement and tober): 115-41.
Distribution of Foodgrains in India. Working Paper Griffin, Charles C. 1988. User Charges for Health
on Food Subsidies 1. Washington, D.C.: Interna- Care in Principle and Practice. EDI Seminar Paper
tional Food Policy Research Institute. 37. Washington, D.C.: World Bank.
Gertler, Paul, and Jacques van der Gaag. Forth- Haddad, Lawrence, and Ravi Kanbur. 1989. "How
coming. The Willingness to Pay for Medical Care: Serious Is the Neglect of Intrahousehold In-
Evidence from Developing Countries. Baltimore, equality?" Policy, Planning, and Research Work-
Md.: Johns Hopkins University Press. ing Paper 296. World Bank, Office of the Vice
Ghai, Dharam, and Samir Radwan, eds. 1983. President, Development Economics, Washing-
Agrarian Policies and Rural Poverty in Africa. Ge- ton, D.C. Processed.
neva: International Labour Office. Haggblade, Steven, and Peter Hazell. 1989. "Agri-
Glewwe, Paul. 1988. The Distribution of Welfare in cultural Technology and Farm-Nonfarm Growth
Côte d'Ivoire in 1985. Living Standards Measure- Linkages." Agricultural Economics 3: 345-64.
ment Study Working Paper 29. Washington, Haggblade, Steven, Peter Hazell, and James
D.C.: World Bank. Brown. 1989. "Farm-Nonfarm Linkages in Rural
1988. The Distribution of Welfare in Peru in Sub-Saharan Africa." World Development 17, 8
1 985-86. Living Standards Measurement Study (August): 1173-1201.
Working Paper 42. Washington, D.C.: World Halstead, Scott B., and others, eds. 1985. Good
Bank. Health at Low Cost. New York: Rockefeller Foun-
Glewwe, Paul, and Dennis de Tray. 1988. The Poor dation.
During Adjustment: A Case Study of Côte d'Ivoire. Hart, Gfflian. 1986. Power, Labor, and Livelihood:
Living Standards Measurement Study Working Processes of Change in Rural Java. Berkeley: Uni-
Paper 47. Washington, D.C.: World Bank. versity of California Press.
1989. The Poor in Latin America during Ad- Hartmann, Betsy, and James K. Boyce. 1983. A
justment: A Case Study of Peru. Living Standards Quiet Violence: View from a Bangladesh Village.
Measurement Study Working Paper 56. Wash- London: Zed Press.
ington, D.C.: World Bank. Hayami, Yujiro, and Vernon W. Ruttan. 1985. Agri-
Glewwe, Paul, and Kwaku A. Twum-Baah. 1989. cultural Development: An International Perspective.
"The Distribution of Welfare in Ghana." World Baltimore, Md.: Johns Hopkins University Press.
Bank, Population and Human Resources Depart- Hazell, Peter B. R., and C. Ramasamy. 1988.
ment, Washington, D.C. Processed. "Green Revolution Reconsidered: The Impact of
Glewwe, Paul, and Jacques van der Gaag. 1988. the High-Yielding Rice Varieties in South India."
Confronting Poverty in Developing Countries: Defi- World Bank, Agriculture and Rural Develop-
nitions, Infonnation, and Policies. Living Standards ment Department, Washington, D.C. Processed.
Measurement Study Working Paper 48. Wash- Herz, Barbara, and Anthony R. Measham. 1987.
ington, D.C.: World Bank. The Safe Motherhood Initiative: Proposals for Action.
Golub, Stephen S., and 1. M. Finger. 1979. "The World Bank Discussion Paper 9. Washington,
Processing of Primary Commodities: Effects of D.C.
Developed-Country Tariff Escalation and Hifi, Kenneth, and Ann R. Pebley. 1988. "Levels,
Developing-Country Export Taxes." Journal of Trends and Patterns of Child Mortality in the De-
Political Economy 87, 3 (June): 559-77. veloping World." Prepared for the Workshop on
Goto, Junichi. 1989. "The Multifibre Arrangement Child Survival Programs: Issues for the 1990s,
and Its Effects on Developing Countries." World School of Hygiene and Public Health, Johns
Bank Research Observer 4, 2 (July): 203-27. Hopkins University, Baltimore, Md., November
Graham, Carol. 1990. "The APRA Government 20-21. Processed.
and the Urban Poor: The PAIT Programme in Hinchliffe, Keith. 1986. "The Monetary and Non-
Lima's Pueblos Jovenes." Duke University, Dur- Monetary Returns to Education in Africa." Edu-
ham, N.C. Processed. cation and Training Discussion Paper EDT46.
World Bank, Washington, D.C. Processed.
151
Hossain, Mahabub. 1988. Credit for Alleviation of Islam, Rizwanul, ed. 1987. Rural Industrialisation
Rural Poverty: The Grameen Bank in Bangladesh. and Employment in Asia. New Delhi: International
Research Report 65. Washington, D.C.: Interna- Labour Organisation, Asian Employment Pro-
tional Food Policy Research Institute in collabo- gramme.
ration with the Bangladesh Institute of Develop- Jamaica Statistical Institute and World Bank. 1988.
ment Studies. "Living Conditions Survey, Jamaica." Kingston.
Hossain, Mahabub, and Rita Afsar. 1989. Credit for Processed.
Women's Involvement in Economic Activities in Rural 1989. "Survey of Living Conditions, Ja-
Bangladesh. Research Report 105. Dhaka: Bangla- maica." Kingston. Processed.
desh Institute of Development Studies. Jamison, Dean T., and Lawrence J. Lau. 1982.
Hossain, Mosharaff. 1987. The Assault That Failed: Farmer Education and Farm Efficiency. Baltimore,
A Profile of Absolute Poverty in Six Villages of Md.: Johns Hopkins University Press.
Bangladesh. Geneva: United Nations Research Jimenez, Emmanuel. 1987. Pricing Policy in the So-
Institute for Social Development. cial Sectors: Cost Recovery for Education and Health
Howard, Donald 5. 1943. The WPA and Federal Re- in Developing Countries. Baltimore, Md.: Johns
lief Policy. New York: Russell Sage Foundation. Hopkins University Press.
Humphreys, Charles, and John Underwood. 1989. Jodha, N. 5. 1985. "Market Forces and Erosion of
"The External Debt Difficulties of Low-Income Common Property Resources." In Agricultural
Africa." In Ishrat Husain and Ishac Diwan, eds., Markets in the Semi-Arid Tropics: Proceedings of the
Dealing with the Debt Crisis. Washington, D.C.: International Workshop Held at ICRISAT Center, In-
World Bank. dia, 24-28 October 1983. Patancheru, India: Inter-
Huppi, Monika, and Gershon Feder. 1989. "The national Crops Research Institute for the Semi-
Role of Groups and Credit Cooperatives in Rural Arid Tropics.
Lending." World Bank, Agriculture and Rural 1986. "Common Property Resources and
Development Department, Washington, D.C. Rural Poor in Dry Regions of India." Economic
Processed. and Political Weekly 21, 27 (July): 1169-81.
Iliffe, John. 1987. The African Poor. Cambridge: Kahnert, Friedrich. 1989. "Assisting Poor Rural
Cambridge University Press. Areas through Groundwater Irrigation: Explor-
India, Government of. 1987. Family Welfare Pro- atory Proposals for East India, Bangladesh and
gramme in India: Year Book 1986-87. New Delhi: Nepal." Asia Regional Series IDP44. World
Ministry of Health and Family Welfare, Depart- Bank, Washington, D.C. Processed.
ment of Family Welfare. Kanbur, Ravi S. 1990. Poverty and the Social Dimen-
Institute of Population Studies, Chulalongkorn sions of Structural Adjustment in Côte d'Ivoire. So-
University. 1989. Thailand: Health and Population cial Dimensions of Adjustment Working Paper 2.
Studies Based on the 1987 Thailand Demographic and Washington, D.C.: World Bank.
Health Survey. Demographic and Health Surveys Karsenty, Guy, and Sam Laird. 1987. "The GSP,
Further Analysis Series 1. New York: Population Policy Options and the New Round." Weltwirts-
Council. chaftliches Archiv 123, 2: 262-96.
International Labour Organisation. 1985. Informal Keja, Ko, and others. 1986. "Effectiveness of the
Sector in Africa. Addis Ababa: Jobs and Skills Pro- Expanded Programme on Immunization." World
gramme for Africa. Health Statistics Quarterly 39: 161-70.
1988. Employment Promotion in the Informal Keyfitz, Nathan. 1985. "An East Javanese Village
Sector in Africa. Addis Ababa: Jobs and Skills Pro- in 1953 and 1985: Observations on Develop-
gramme for Africa. ment." Population and Development Review 11 (De-
International Monetary Fund. 1989. World Economic cember): 695-719.
Outlook: A Survey by the Staff of the International Khandker, Shahidur R. 1989. "Improving Rural
Monetary Fund. Washington, D.C. Wages in India." Policy, Planning, and Research
Various years. International Financial Statis- Working Paper 276. World Bank, Population and
tics Yearbook. Washington, D.C. Human Resources Department, Washington,
International Rice Research Institute. 1983. Women D.C. Processed.
in Rice Farming. Manila. Korten, David. 1980. "Community Organization
and Rural Development: A Learning Process Ap-
proach." Public Administration Review 40, 5
(September-October): 480-511.
152
Korten, David C., ed. 1986. Community Manage- Lele, Uma, and Steven W. Stone. 1989. Population
ment: Asian Experience and Perspectives. West Hart- Pressure, the Environment, and Agricultural Inten-
ford, Conn.: Kumarian Press. sification: Variations on the Boserup Hypothesis.
Korten, Frances F., and Robert Y. Siy, Jr. 1988. MADIA Discussion Paper 4. Washington, D.C.:
Transforming a Bureaucracy: The Experience of the World Bank.
Philippine National Irrigation Administration. West Leonard, Jeffrey H., and others. 1989. Environment
Hartford, Conn.: Kumarian Press. and the Poor: Development Strategies for a Common
Krueger, Anne 0. 1983. Trade and Employment in Agenda. U.S.-Third World Policy Perspectives
Developing Countries: Synthesis and Conclusions. 11. New Brunswick, N.J.: Transaction Books.
Chicago: University of Chicago Press. Levy, Victor. 1988. "Aid and Growth in Sub-
Krueger, Anne 0., and Vernon W. Ruttan. 1989. Saharan Africa: The Recent Experience." Euro-
"Assistance to Korea." In Anne 0. Krueger, pean Economic Review 32: 1777-95.
Constantine Michalopoulos, and Vernon W. Rut- Lewis, John P., and others. 1988. Strengthening the
tan, eds. Aid and Development. Baltimore, Md.: Poor: What Have We Learned? U.S.-Third World
Johns Hopkins University Press. Policy Perspectives 10. Washington, D.C.: Over-
Krueger, Anne 0., Maurice Schiff, and Alberto seas Development Council.
Valdés. 1988. "Agricultural Incentives in Devel- Liedhoim, Carl, and Donald Mead. 1987. Small
oping Countries: Measuring the Effect of Sec- Scale Industries in Developing Countries: Empirical
toral and Economywide Policies." World Bank Evidence and Policy Implications. MSU Interna-
Economic Review 2, 3 (September): 255-71. tional Development Paper 9. East Lansing: De-
Laird, Sam, and Alexander Yeats. 1990. Quantita- partment of Agricultural Economics, Michigan
tive Methods for Trade Barrier Analysis. London: State University.
Macmillan. Lipton, Michael. 1983. Labor and Poverty. World
Lal, Deepak, and Hla Myint. 1989. "The Political Bank Staff Working Paper 616. Washington,
Economy of Poverty, Equity, and Growth in D.C.
Some Developing Countries: A Comparative 1985. Land Assets and Rural Poverty. World
Study." World Bank, Latin America Technical Bank Staff Working Paper 744. Washington,
Department, Washington, D.C. Processed. D.C.
Ledesma, Antonio J. 1982. Landless Workers and Rice 1987. Improving the Impact of Aid for Rural
Farmers: Peasant Subclasses under Agrarian Reform Development. Discussion Paper 233. Brighton,
in Two Philippine Villages. Los Baños, Philippines: U.K.: Institute of Development Studies, Univer-
International Rice Research Institute. sity of Sussex.
Lee, Kyu Sik. 1989. The Location of Jobs in a Develop- Lipton, Michael, with Richard Longhurst. 1989.
ing Metropolis: Patterns of Growth in Bogota and New Seeds and Poor People. London: Unwin Hy-
Call, Colombia. New York: Oxford University man.
Press. Lipton, Michael, and John Toye. 1990. Does Aid
Lee, Kyu Sik, and Alex Anas. 1989. "Manufactur- Work in India? A Country Study of the Impact of
ers' Responses to Infrastructure Deficiencies in Official Development Assistance. London:
Nigeria: Private Alternatives and Policy Op- Routledge.
tions." Infrastructure and Urban Development Little, Ian M. D., Dipak Mazumdar, and John M.
Department Discussion Paper INU5O. World Page, Jr. 1987. Small Manufacturing Enterprises: A
Bank, Washington, D.C. Processed. Comparative Analysis of India and Other Economies.
Lele, Uma. 1989. "Aid to African Agriculture: Les- New York: Oxford University Press.
sons from Two Decades of Donor Experience." Lockheed, Marlaine, Adriaan Verspoor, and oth-
Discussion draft prepared for the Managing Ag- ers. 1990. "Improving Primary Education in De-
ricultural Development in Africa (MADIA) Sym- veloping Countries: A Review of Policy Op-
posium. World Bank, Africa Technical Depart- tions." World Bank, Population and Human
ment, Washington, D.C. Processed. Resources Department, Washington, D.C. Pro-
Lele, Uma, and L. Richard Meyers. 1989. Growth cessed.
and Structural Change in East Africa: Domestic Poli- Londoño, Juan Luis. 1989. "Income Distribution in
cies, Agricultural Performance, and World Bank As- Colombia: Turning Points, Catching Up and
sistance, 1963-86. MADIA Discussion Paper 3. Other Kuznetsian Tales." Harvard University,
Washington, D.C.: World Bank. Cambridge, Mass. Processed.
153
Mazumdar, Dipak. 1989. Microeconomic Issues of La- Organisation for Economic Co-operation and De-
bor Markets in Developing Countries: Analysis and velopment. 1985. Twenty-Five Years of Development
Policy Implications. EDT Seminar Paper 40. Wash- Co-operation: A Review. Paris.
ington, D.C.: World Bank. 1987. National Policies and Agricultural
McGregor, J. Allister. 1988. "Credit and the Rural Trade. Paris.
Poor: The Changing Policy Environment in 1988. Voluntary Aid for Development: The
Bangladesh." Public Administration and Develop- Role of Non-Governmental Organisations. Paris.
ment 8: 467-81. 1989. Development Co-operation in the 1990s.
Mencher, Joan. 1988. "Women's Work and Pov- Paris.
erty: Women's Contribution to Household Main- 1990. Geographical Distribution of Financial
tenance in Two Regions of South India." In Flows to Developing Countries, 1985-88. Paris.
Daisy Dwyer and Judith Bruce, eds., A Home Di- Osmani, Siddiqur R., and Omar H. Chowdhury.
vided: Women and Income in the Third World. Palo 1983. "Short Run Impact of Food for Work Pro-
Alto, Calif.: Stanford University Press. gramme in Bangladesh." Bangladesh Development
Migot-Adholla, Shem, and others. n.d. "Land Studies 11: 135-90.
Tenure Reform and Agricultural Development in Paul, Samuel. 1983. Managing Development Pro-
SubSaharan Africa." World Bank, Agriculture grams: The Lessons of Success. Boulder, Cob.:
and Rural Development Department, Washing- Westview Press.
ton, D.C. Processed. Peek, Peter. 1988. "How Equitable Are Rural De-
Morgan, Richard G. Forthcoming. "Social Welfare velopment Projects?" International Labour Review
Programmes and the Reduction of Household 127, 1: 73-89.
Vulnerability in the SADCC States of Southern Piazza, Alan, and John Doolette. 1990. "The Im-
Africa." In Ehtisham S. Ahmad and others, pact of Combining Agricultural Techniques and
eds., Social Security in Developing Countries. Ox- Soil Conservation Measures in China's Loess
ford: Oxford University Press. Plateau." In Proceedings of the 1988 International
Moser, Caroline 0. 1989. "The Impact of Recession Conference on Dryland Farming. College Station:
and Structural Adjustment Policies at the Micro- Texas Agricultural Experiment Station, Texas
Level: Low Income Women and Their House- A&M University.
holds in Guayaquil, Ecuador." Department of Pinstrup-Andersen, Per. 1988. Consumer-Oriented
Social Administration, London School of Eco- Food Subsidies: Cost, Benefits and Policy Options for
nomics and Political Science. Processed. Developing Countries. Baltimore, Md.: Johns
Mosley, Paul. 1987. Overseas Aid: Its Defence and Hopkins University Press.
Reform. Brighton, U.K.: Wheatsheaf Books. Pinstrup-Andersen, Per, and Peter B. R. Hazell.
Mosley, Paul, and Rudra Prasad Dahal. 1987. 1985. "The Impact of the Green Revolution and
"Credit for the Rural Poor: A Comparison of Pol- Prospects for the Future." Reprinted from Food
icy Experiments in Nepal and Bangladesh." Reviews International 1, 1. Washington, D.C.: In-
Manchester Papers on Development 3, 2 (July): ternational Food Policy Research Institute.
45-59. Pinto, Brian. 1987. "Nigeria during and after the
Murray, Christopher J. L. 1987. "A Critical Review Oil Boom." World Bank Economic Review 1, 3
of International Mortality Data." Social Science (May): 419-45.
and Medicine 25, 7: 773-81. Platteau, Jean-Philippe. Forthcoming. "Traditional
Nelson, Joan M., ed. 1990. Economic Crises and Pol- Systems of Social Security and Hunger Insur-
icy Choice: The Politics of Adjustment in the Third ance: Past Achievements and Modern Chal-
World. Princeton, N.J.: Princeton University lenges." In Ehtisham S. Ahmad and others,
Press. eds., Social Security in Developing Countries. Ox-
Newman, John L. 1988. Labor Market Activity in ford: Oxford University Press.
Côte d 'Ivoire and Peru. Living Standards Measure- Preston, Samuel. 1980. "Causes and Conse-
ment Study Working Paper 36. Washington, quences of Mortality Declines in Less Developed
D.C.: World Bank. Countries During the Twentieth Century." In
Oberai, A. S. 1987. Migration, Llrbanisation, and De- Richard Easterlin, ed., Population and Economic
velopment. Training in Population, Human Re- Change in Developing Countries. Chicago: Univer-
sources, and Development Planning Back- sity of Chicago Press.
ground Paper 5. Geneva: International Labour
Office.
154
Psacharopoulos, George. 1985. "Returns to Educa- Rodgers, Gerry, ed. 1989. Urban Poverty and the
tion: A Further International Update and Impli- Labor Market: Access to Jobs and Incomes in Asian
cations." Journal of Human Resources 20 (Fall): and Latin American Cities. Geneva: International
583-604. Labour Office.
Pulley, Robert V. 1989. Making the Poor Credit- Rodgers, Gerry, and Guy Standing, eds. 1981.
worthy: A Case Study of the Integrated Rural Devel- Child Work, Poverty and Llnderdevelopment. Ge-
opment Program in India. World Bank Discussion neva: International Labour Office.
Paper 58. Washington, D.C. Rodrik, Dani. 1989. "Policy Uncertainty and Pri-
Ravallion, Martin. 1987. Markets and Famines. Ox- vate Investment in Developing Countries."
ford: Clarendon Press. NBER Working Paper 2999. National Bureau of
1988. "Expected Poverty under Risk In- Economic Research, Cambridge, Mass. Pro-
duced Welfare Variability." Economic Journal 98, cessed.
393 (December): 1171-82. Sachs, Jeffrey D. 1989. "Social Conflict and Popu-
1989a. "Income Effects on Undernutri- list Policies in Latin America." NBER Working
tion." Economic Development and Cultural Change Paper 2897. National Bureau of Economic Re-
38 (April): 489-516. search, Cambridge, Mass. Processed.
1989b. "Land-Contingent Poverty Allevia- Salmen, Lawrence F. 1987. Listen to the People:
tion Schemes." World Development 17, 8 (Au- Participant-Observer Evaluation of Development
gust): 1223-33. Projects. New York: Oxford University Press.
1990. "Reaching the Poor through Rural Sarvekshana, Journal of the National Sample Survey
Public Works: Arguments, Evidence, and Les- Organization. 1988. Report of the Third Quinquen-
sons from South Asia." World Bank, Agriculture nial Survey on Employment and Unemployment.
and Rural Development Department, Washing- New Delhi: Department of Statistics, Ministry of
ton, D.C. Processed. Planning.
Forthcoming. "Rural Welfare Effects of Schiff, Maurice, and Alberto Valdés. Forthcoming.
Food Price Changes under Induced Wage Re- The Economics of Agricultural Pricing Incentives in
sponses: Theory and Evidence for Bangladesh." Developing Countries. Vol. 4 of Anne 0. Krueger,
Oxford Economic Papers. Maurice Schiff, and Alberto Valdés, eds., A Com-
Ravallion, Martin, and Lorraine Dearden. 1988. parative Study of the Political Economy of Agricul-
"Social Security in a 'Moral Economy': An Em- tural Pricing Policies. Baltimore, Md.: Johns
pirical Analysis for Java." Review of Economics and Hopkins University Press.
Statistics 70, 1 (February): 36-44. Sen, Amartya. 1981. Poverty and Famines: An Essay
Ravallion, Martin, and Monika Huppi. 1989. "Pov- on Entitlement and Deprivation. Oxford: Claren-
erty and Undernutrition in Indonesia during the don Press.
1980s." Agriculture and Rural Development De- Siamwalla, Ammar, and Suthad Setboonsarng.
partment Working Paper 286. World Bank, 1989. Trade, Exchange Rate, and Agricultural Pric-
Washington, D.C. Processed. ing Policies in Thailand. World Bank Comparative
Ravallion, Martin, and Dominique van de Walle. Studies. Washington, D.C.
1988. "Poverty Orderings of Food Pricing Re- Singh, K. P., and Niva Bara. 1988. "Integrating
forms." Discussion Paper 86. Development Eco- Women's Concerns in Farming Systems
nomics Research Centre, University of Warwick, Research/Extension: A Project in the Tribal Areas
Coventry, U.K. Processed. of Bihar (India)." Presented at the Farming Sys-
Riddell, Roger C. 1987. Foreign Aid Reconsidered. tems Symposium, University of Arkansas, Octo-
Baltimore, Md.: Johns Hopkins University Press. ber 10-12. Processed.
Riveros, Luis A. 1989. "Labor Markets in an Era of Sinn, Hans-Werner. 1988. "The Sahel Problem."
Adjustment: The Chilean Case." Prepared for Kyklos 41, 2; 187-213.
the Conference on Labor Markets in an Era of Spinanger, Dean. 1987. "Wifi the Multi-Fibre Ar-
Adjustment, Coventry, U.K., August 7-10, rangement Keep Bangladesh Humble?" World
sponsored by the Development Economics Re- Economy 10, 1 (March): 75-84.
search Centre, University of Warwick, and the Srinivasan, T. N., and Pranab K. Bardhan, eds.
Economic Development Institute of the World 1988. Rural Poverty in South Asia. New York: Co-
Bank. Processed. lumbia University Press.
155
Steinberg, David I. 1985. "Foreign Aid and the De- United Nations Conference on Trade and Develop-
velopment of the Republic of Korea: The Effec- ment. 1988. Handbook of International Trade and
tiveness of Concessional Assistance." AID Spe- Development Statistics: Supplement 1987. New
cial Study 42. U.S. Agency for International York.
Development, Washington, D.C. Processed. 1989. Uruguay Round: Papers on Selected Is-
Streeten, Paul. 1987. What Price Food? Agricultural sues. New York.
Price Policies in Developing Countries. New York: United Nations, Department of International Eco-
St. Martin's Press. nomic and Social Affairs. 1988. Mortality of Chil-
Subbarao, Kalanidhi. Forthcoming. "Interven- dren under Age 5: World Estimates and Projections,
tions to Combat Household-Level Food Insecu- 1950-2025. Population Studies 105. New York.
rity: A Review of India's Experience." In S. United Nations Educational, Scientific, and Cul-
Guhan, Barbara Harriss, and Robert Cassen, tural Organization. Various years. Statistical Year-
eds., Poverty in India. Dethi: Oxford University book. Paris.
Press. Universidad de Chile. Facultad de Ciencias
Summers, Robert, and Alan Heston. 1988. "A Económicas y Administrativas, Departamento
New Set of International Comparisons of Real EconomIa. 1987. Estudio sobre los Programas Espe-
Product and Price Levels Estimates for 130 Coun- ciales de Empleo. Vols. 1-3. Santiago.
tries, 1950-1985." Review of Income and Wealth Uphoff, Norman. 1986. Local Institutional Develop-
(March): 1-24. ment: An Analytical Sourcebook with Cases. West
Tendler, Judith. 1989. "What Ever Happened to Hartford, Conn.: Kumarian Press.
Poverty Alleviation?" World Development 17, 7 U.S. Agency for International Development. 1989.
(July): 1033-44. A. I. D. Microenterprise Stock-taking: Synthesis Re-
Trela, Irene, and John Whalley. 1988. "Do Devel- port. AID Evaluation Occasional Paper. Washing-
oping Countries Lose from the MFA?" NBER ton, D.C.
Working Paper 2618. National Bureau of Eco- U.S. Arms Control and Disarmament Agency.
nomic Research, Cambridge, Mass. Processed. 1988. World Military Expenditures and Arms Trans-
Tremblay, Hélène, and Pat Capon. 1988. Families of fers 1988. Washington, D.C.: Government Print-
the World: Family Life at the Close of the Twentieth ing Office.
Century. Vol. 1: The Americas and the Caribbean. van Wijnbergen, Sweder. 1989. "Growth, External
New York: Farrar, Straus and Giroux. Debt, and the Real Exchange Rate in Mexico."
Tyers, Rodney, and Kym Anderson. 1986. "Distor- Policy, Planning, and Research Working Paper
tions in World Food Markets: A Quantitative As- 257. World Bank, Latin America and the Carib-
sessment." Background paper for World Develop- bean Country Department II, Washington, D.C.
ment Report 1986. World Bank, World Processed.
Development Report Office, Washington, D.C. 1990. "Mexico's External Debt Restruc-
Processed. turing in 1989/90: An Economic Analysis."
United Nations. 1987. "First Report on the World World Bank, Latin America and the Caribbean
Nutrition Situation." Administrative Committee Regional Office, Washington, D.C. Processed.
on Coordination, Subcommittee on Nutrition, von Braun, Joachim, Eileen Kennedy, and Ho-
Geneva. Processed. warth Bouis. 1989. "Comparative Analyses of
1989. "Update on the Nutrition Situation: the Effects of Increased Commercialization of
Recent Trends in Nutrition in 33 Countries." Ad- Subsistence Agriculture on Production, Con-
ministrative Committee on Coordination, Sub- sumption, and Nutrition." International Food
committee on Nutrition, Geneva. Processed. Policy Research Institute, Washington, D.C. Pro-
United Nations Children's Fund. 1988. Children on cessed.
the Front Line: The Impact of Apartheid, Destabiliza- Von Pischke, J. D. 1989. "Finance at the Frontier:
tion and Warfare on Children in Southern and South Debt Capacity and the Role of Credit in Develop-
Africa. New York. ing the Private Economy." World Bank, Eco-
1989. Improving Child Survival and Nutri- nomic Development Institute, Washington, D.C.
tion: The Joint WHO/UNICEF Nutrition Support Processed.
Programme in Iringa, Tanzania. Dar es Salaam. Von Pischke, J. D., Dale Adams, and Gordon
1990. The State of the World's Children. New Donald. 1983. Rural Financial Markets in Develop-
York: Oxford University Press. ing Countries: Their Use and Abuse. Baltimore,
Md.: Johns Hopkins University Press.
156
Wagao, Imanne Hamisi. 1986. "The Impact of 198Th. Financing Health Services in Develop-
Global Recession on Living Standards- ing Countries: An Agenda for Reform. A World
Tanzania." Economics Department, University Bank Policy Study. Washington, D.C.
of Dar es Salaam. Processed. 1988a. Education in Sub-Saharan Africa: Poli-
Walker, Thomas S., and John G. Ryan, with oth- cies for Adjustment, Revitalization, and Expansion. A
ers. Forthcoming. Against the Odds: Village and World Bank Policy Study. Washington, D.C.
Household Economies in India's Semi-Arid Tropics. 1988b. "The Philippines: The Challenge of
Baltimore, Md.: Johns Hopkins University Press. Poverty." Report 7144-PH. Asia Country De-
Watanabe, Barbara, and Eva Mueller. 1984. "A partment II, Washington, D.C. Processed.
Poverty Profile for Rural Botswana." World De- 1988c. Rural Development: World Bank Expe-
velopment 12, 2 (February): 115-27. rience, 1965-86. A World Bank Operations Evalu-
Webster, Leila. 1989. "World Bank Lending for ation Study. Washington, D.C.
Small and Medium Enterprises: Fifteen Years of 1988d. World Development Report 1988. New
Experience." World Bank, Industry and Energy York: Oxford University Press.
Department, Washington, D.C. Processed. 1989a. "Feeding Latin America's Chil-
Whalley, John. 1985. Trade Liberalization among Ma- dren." Latin America and the Caribbean Techni-
jor World Trading Areas. Cambridge, Mass.: MIT cal Department, Washington, D.C. Processed.
Press. 1989b. India: Poverty, Employment, and So-
Williamson, Jeffrey G. 1986. "Migration and Ur- cial Services. A World Bank Country Study.
banization in the Third World." Harvard Insti- Washington, D.C.
tute for Economic Research Discussion Paper 1989c. Sub-Saharan Africa: From Crisis to
1245. Cambridge, Mass. Processed. Sustainable Growth. A Long-Term Perspective
Wilson, Francis, and Mamphela Ramphele. 1989. Study. Washington, D.C.
Uprooting Poverty: The South African Challenge. 1989d. Trends in Developing Economies.
New York: W. W Norton. Washington, D.C.
World Bank. 1983. Learning by Doing: World Bank 1989e. Women in Pakistan: An Economic and
Lending for Urban Development, 1972-82. Wash- Social Strategy. Washington, D.C.
ington, D.C. 1989f. World Debt Tables 1989-90: External
1986a. Aid for Development: The Key Issues. Debt of DevelopingCountries. Washington, D.C.
Supporting Materials for the Report of the Task Force 1989g. World Development Report 1989. New
on Concessional Flows. Development Committee York: Oxford University Press.
pamphlet 8. Washington, D.C. World Health Organization. Various years. World
1986b. Financing Education in Developing Health Statistics Annual. Geneva.
Countries: An Exploration of Policy Options. Wash- Yeats, Alexander J. 1981. "Agricultural Protection-
ington, D.C. ism: An Analysis of Its International Economic
1986c. Poverty and Hunger: Issues and Op- Effects and Options for Institutional Reform."
tions for Food Security in Developing Countries. A Trade and Development 3 (Winter): 1-29.
World Bank Policy Study. Washington, D.C. 1989. "Shifting Patterns of Comparative
World Development Report 1986. New York: Advantage: Manufactured Exports of Develop-
Oxford University Press. ing Countries." Policy, Planning, and Research
1987a. The Aga Khan Rural Support Program Working Paper 165. World Bank, International
in Pakistan: An Interim Evaluation. A World Bank Economics Department, Washington, D.C. Pro-
Operations Evaluation Study. Washington, D.C. cessed.
157
Statistical appendix
The tables in this statistical appendix present sum- should refer to the "Definitions and data notes"
mary data on the population, national accounts, for an explanation of the country groupings and to
trade, and external debt of the low- and middle- the technical notes to the World Development In-
income economies, the high-income economies, dicators for definitions of the concepts used.
and all reporting economies as a group. Readers
Table A.1 Population growth, 1965 to 1990, and projected to 2000
1989 Average annual growth (percent)
population
Country group (millions) 1 965-73 1973-80 1 980-90 1990-2000
Low- and middle-income economies 4,037 2.5 2.1 2.1 1.9
Low-income economies 2,947 2.6 2.1 2.0 1.9
Middle-income economies 1,090 2.4 2.3 2.1 1.9
Sub-Saharan Africa 479 2.6 2.8 3.2 3.1
East Asia 1,566 2.7 1.7 1.5 1.4
South Asia 1,132 2.4 2.4 2.3 1.9
Europe, Middle East, and North Africa 404 1.9 2.1 2.1 2.1
Latin America and the Caribbean 422 2.7 2.4 2.1 1.8
Severely indebted middle-income
economies 506 2.5 2.3 2.1 1.8
High-income economies 789 1.0 0.8 0.7 0.5
OECD members 755 1.0 0.7 0.6 0.5
Total reporting economies 4,826 2.2 1.9 1.8 1.7
Oil exporters 609 2.7 2.7 2.7 2.4
159
Table A.2 Population and GNP per capita, 1980, and growth rates, 1965 to 1989
1980 GNP 1980 1980 GNP Average annual growth of GNP per capita (percent)
(billions population per capita
Country group of dollars) (millions) (dollars) 1965-73 1973 -80 1980-86 1987 1988 1989'
Low- and middle-income
economies 2,406 3,359 700 4.0 2.6 1.5 2.7 3.4 1.2
Low-income economies 784 2,459 320 3.6 2.4 4.0 3.9 6.8 1.8
Middle-income economies 1,622 900 1,760 4.6 2.4 0.1 1.8 1.1 0.8
Sub-Saharan Africa 213 362 570 3.0 0.1 -2.8 -4.4 -0.8 0.5
East Asia 586 1,363 420 5.4 4.4 6.6 8.0 8.7 3.1
South Asia 220 922 240 1.0 2.0 3.2 0.9 6.1 2.3
Europe, Middle East, and
North Africa 590 335 1,740 5.6 2.1 0.8 -0.6 0.1 0.6
Latin America and the Caribbean 716 348 2,000 4.1 2.4 -1.6 1.5 -0.8 -0.8
Severely indebted middle-
income economies 791 419 1,840 4.2 2.6 -1.5 1.2 -0.4 -0.7
High-income economies 7,923 742 10,740 3.5 2.2 1.7 2.5 3.5 3.1
OECD members 7,663 716 10,750 3.5 2.2 1.9 2.8 3.7 3.1
Totalreportingeconomies 10,329 4,101 2,520 2.7 1.5 0.9 1.8 2.7 1.7
Oil exporters 964 479 1,980 4.6 2.8 -1.6 -2.5 0.5
a. Preliminary.
Table A.3 Population and composition of GDP, selected years, 1965 to 1989
(billions of dollars, unless otherwise specified)
Country group and indicator 1965 1973 1980 1985 1986 1987 1988 1989'
Low- and middle-income economies
GDP 377 849 2,406 2,521 2,598 2,818 3,159
Domestic absorption 375 842 2,446 2,522 2,618 2,801 3,178
Net exports 1 7 -39 0 -20 17 -19
Population (millions) 2,377 2,897 3,359 3,718 3,794 3,874 3,952 4,037
Low-income economies
GDP 168 312 784 824 785 812 923 984
Domestic absorption 170 310 788 853 816 823 941
Net exports -2 2 -4 -29 -31 -11 -18
Population (millions) 1,741 2,129 2,459 2,714 2,770 2,828 2,884 2,947
Middle-income economies
GDP 209 537 1,622 1,698 1,813 2,006 2,236
Domestic absorption 206 532 1,658 1,669 1,803 1,978 2,237
Net exports 3 5 -36 29 10 28 -1
Population (millions) 635 768 900 1,003 1,025 1,046 1,068 1,090
Sub-Saha ran Africa
GDP 29 65 213 195 164 150 161 156
Domestic absorption 29 64 211 194 170 152 162
Net exports 0 2 2 0 -6 -1 -1
Population (millions) 243 299 362 422 435 449 464 479
East Asia
CDP 93 215 586 645 650 732 889 1,026
Domestic absorption 93 213 585 645 635 699 860
Net exports 0 2 1 0 14 34 29
Population (millions) 980 1,208 1,363 1,470 1,492 1,516 1,538 1,566
South Asia
GDP 69 97 220 277 294 325 348 345
Domestic absorption 72 99 236 291 308 332 358 351
Netexports -3 -2 -15 -14 -13 -7 -10 -6
Population (millions) 645 781 922 1,033 1,058 1,082 1,107 1,132
Europe, Middle East, and North Africa
GDP 73 186 590 668 741 796
Domestic absorption 73 183 611 693
Net exports 0 3 -21 -25
Population (millions) 250 290 335 372 380 388 396 404
160
Table A.3 (continued)
Country group and indicator 1965 1973 1980 1985 1986 1987 1988 1989'
Latin America and the Caribbean
GDP 99 254 716 682 698 739 852
Domestic absorption 96 253 726 650 681 720 830
Net exports 3 1 -10 31 16 19 22
Population (millions) 239 295 348 388 397 405 414 422
Severely indebted middle-income
economies
GDP 109 280 791 755 784 822 943 955
Domestic absorption 106 278 807 721 769 802 918
Net exports 3 2 -15 34 16 20 25
Population (millions) 294 357 419 466 476 486 496 506
High-income economies
GDP 1,406 3,330 7,914 8,938 10,860 12,599 14,108 15,237
Domestic absorption 1,396 3,297 7,867 8,914 10,807 12,560 14,049 15,173
Net exports 10 33 47 24 53 39 59 64
Population (millions) 647 701 742 768 773 779 784 789
OECD members
GDP 1,391 3,283 7,652 8,700 10,633 12,347 13,836 14,943
Domestic absorption 1,382 3,255 7,666 8,679 10,568 12,298 13,767 14,868
Net exports 9 28 -14 21 65 50 68 75
Population (millions) 632 681 716 738 742 746 751 755
Total reporting economies
GDP 1,783 4,180 10,320 11,459 13,458 15,417 17,267
Domestic absorption 1,772 4,139 10,313 11,435 13,425 15,361 17,227
Net exports 11 41 8 24 32 56 40
Population (millions) 3,024 3,598 4,101 4,485 4,567 4,653 4,736 4,826
Oil exporters
GDP 78 226 965 1,005 844 850 926
Domestic absorption 75 209 861 985 872 848 938
Net exports 3 17 104 19 -28 2 -12
Population (millions) 321 398 479 547 562 578 593 609
Note: Components may not sum to totals because of rounding. Domestic absorption includes private consumption, general government con-
sumption, and gross domestic investment. Net exports includes goods and nonf actor services.
a. Preliminary.
Table A.4 GDP, 1980, and growth rates, 1965 to 1989
1980
GDP Average annual growth of GDP (percent)
(billions
Country group of dollars) 1965-73 1973 -80 1980-86 1987 1988 1989'
Low- and middle-income economies 2,406 6.6 4.8 3.8 4.5 5.4 3.3
Low-income economies 784 6.0 4.6 6.1 6.1 8.9 4.2
Middle-income economies 1,622 6.9 4.9 2.5 3.5 3.0 2.8
Sub-Saharan Africa 213 5.9 2.7 0.3 -1.1 2.5 3.5
East Asia 586 7.9 6.5 7.9 9.6 10.0 5.1
South Asia 220 3.6 4.1 5.6 3.2 8.9 4.8
Europe, Middle East, and North
Africa 590 7.5 4.2 3.2 1.4 2.4
Latin America and the Caribbean 716 6.5 5.1 0.9 3.0 1.2 1.5
Severely indebted middle-income
economies 791 6.4 5.2 1.0 2.6 1.6 1.4
High-income economies 7,914 4.6 3.0 2.5 3.4 4.3 3.6
OECD members 7,652 4.5 2.9 2.6 3.4 4.3 3.6
Total reporting economies 10,319 4.9 3.4 2.8 3.7 4.6 3.4
Oil exporters 965 7.3 4.9 0.8 0.3 3.0
a. Preliminary.
161
Table A.5 GDP structure of production, selected years, 1965 to 1988
(percentage of GDP)
1965 1973 1980 1985 1986 1987 1988a
Agri- Agri- Agri- Agri- Agri- Agri- Agri-
cul- Indus- cul- Indus- cul- Indus- cul- Indus- cul- Indus- cul- Indus- cul- Indus-
Count ry group ture try ture try ture try ture try ture try ture try ture try
Low- and middle-income
economies 30 30 24 34 19 38 19 36 19 35 18 37 17 37
Low-income economies 43 27 39 32 32 37 33 33 32 32 31 33 31 34
Middle-income economies 19 32 15 35 12 38 12 37 13 37 13 38 10 39
Sub-Saharan Africa 40 17 33 24 28 32 34 26 34 23 31 25 31 25
East Asia 41 35 34 40 26 44 25 42 24 42 23 43 22 43
South Asia 41 19 43 19 35 22 31 24 29 25 29 25 30 24
Europe, Middle East, and
North Africa 22 33 17 38 14 41
Latin America and the
Caribbean 15 32 12 33 9 36 10 36 11 35 9 37 6 37
Severely indebted middle-
income economies 16 32 13 33 10 36 11 35 11 34 10 37 7 38
High-income economies 5 40 5 38 3 37 3 34 3 32 3 31
OECD members 5 40 5 38 3 36 3 34 3 32 3 31
Total reporting economies 10 38 9 37 7 37 6 35 6 33 5 32
Oil exporters 19 32 14 38 11 48 13 38 13 34 12 36
a. Preliminary.
Table A.6 Sector growth rates, 1965 to 1989
(average annual percentage change)
Agriculture Industry Services
Country group 1965-73 1973-80 1980-89 1965-73 1973-80 1980-89 1965-73 1973 -80 1980-89
Low- and middle-income
economies 3.0 2.3 3.7 8.8 4.9 5.3 7.3 6.1 3.8
Low-incomeeconomies 2.9 1.8 4.3 10.7 7.0 8.7 6.3 5.3 6.1
Middle-income economies 3.2 3.0 2.7 8.0 4.0 3.2 7.6 6.3 3.1
Sub-Saharan Africa 2.2 -0.3 1.8 13.9 4.2 -0.2 4.1 3.1 1.5
East Asia 3.2 2.5 5.3 12.7 9.2 10.3 10.5 7.3 7.9
South Asia 3.1 2.2 2.7 3.9 5.6 7.2 4.0 5.3 6.1
Europe, Middle East,
and North Africa 3.4 3.2 8.6 1.4 8.1 8.4
Latin America and the
Caribbean 3.0 3.7 2.5 6.8 5.1 1.1 7.3 5.4 1.7
Severely indebted middle-
income economies 3.1 3.6 2.7 6.8 5.4 1.0 7.2 5.4 1.7
High-income economies 1.4 0.5 2.3 3.9 2.2 1.9 4.5 3.4 3.0
OECD members 1.4 0.5 2.2 3.7 2.0 2.2 4.5 3.3 3.0
Total reporting economies 2.2 1.8 3.2 4.8 2.8 2.5 4.9 3.9 3.2
Oil exporters 3.3 2.1 2.7 9.4 3.3 -0.1 6.4 8.0 2.4
Note: Figures in italics are for years other than those specified.
162
Table A.7 Consumption, investment, and saving, selected years, 1965 to 1988
(percentage of GDP)
Count ry group and indicator 1965 1973 1980 1985 1986 1987 1988'
Low- and middle-income economies
Consumption 79.4 76.1 74.7 75.9 76.6 75.2 74.8
Investment 20.4 23.2 26.9 24.1 24.2 24.1 25.6
Saving 18.9 21.7 23.6 21.2 20.8 22.3 22.6
Low-income economies
Consumption 81.8 76.7 74.6 76.5 76.4 74.4 74.2
Investment 19.1 22.7 25.8 27.0 27.5 27.0 27.8
Saving 17.8 21.0 24.4 22.4 22.4 24.2 24.4
Middle-income economies
Consumption 77.3 75.7 74.7 75.7 76.8 75.6 75.2
Investment 21.3 23.3 27.5 22.7 22.7 23.0 24.9
Saving 19.9 22.2 23.2 20.7 20.1 21.5 21.9
Sub-Saha ran Africa
Consumption 84.0 79.1 78.6 87.8 89.0 85.5 85.0
Investment 14.3 18.3 20.5 12.2 14.6 15.4 15.4
Saving 13.8 16.9 18.2 8.8 6.7 8.1 8.3
East Asia
Consumption 77.6 71.6 69.1 69.1 67.7 65.6 65.9
Investment 22.3 27.3 30.7 30.9 30.1 29.8 30.9
Saving 22.3 25.8 29.7 29.5 31.1 33.3 33.3
South Asia
Consumption 87.5 84.7 84.8 81.1 81.8 81.0 80.8
Investment 16.8 17.0 22.1 23.9 22.8 21.3 22.1
Saving 11.9 14.9 15.3 18.0 17.2 17.9 18.0
Europe, Middle East, and North
Africa
Consumption 78.1 73.1 73.1 75.4 76.9 78.8
Investment 21.9 25.3 30.5 28.3
Saving 17.8 24.2 25.8 22.3 21.1 18.8
Latin America and the Caribbean
Consumption 76.4 78.4 77.2 77.8 80.2 77.7 75.9
Investment 20.3 21.2 24.2 17.6 17.4 19.7 21.5
Saving 21.1 19.4 20.3 16.8 14.9 17.9 19.6
Severely indebted middle-income
economies
Consumption 76.0 77.8 76.8 76.8 79.3 77.0 75.0
Investment 20.9 21.5 25.1 18.7 18.7 20.5 22.4
Saving 21.0 19.5 20.5 18.0 16.1 18.8 20.9
High-income economies
Consumption 82.6 76.3 77.4 80.0 79.6 79.3 78.0
Investment 16.7 22.7 22.0 19.7 19.9 20.4 21.6
Saving 17.7 24.1 23.2 20.4 20.7 20.9 22.2
OECD members
Consumption 79.6 76.4 78.2 80.1 79.6 79.2 78.0
Investment 19.8 22.7 22.0 19.6 19.8 20.4 21.5
Saving 20.8 24.1 22.3 20.1 20.6 20.9 22.1
Total reporting economies
Consumption 82.0 76.2 76.7 79.2 79.1 78.6 77.5
Investment 17.5 22.9 23.2 20.7 20.7 21.1 22.4
Saving 18.0 23.7 23.3 20.5 20.7 21.1 22.2
Oil exporters
Consumption 76.2 69.9 63.5 74.8 79.2 75.8 76.1
Investment 19.9 22.5 25.7 23.3 24.2 23.9 25.2
Saving 18.6 25.2 35.0 23.8 20.2 22.8 22.1
a. Preliminary.
163
Table A.8 Growth of export volume, 1965 to 1988
Average annual change in export volume (percent)
Country group and commodity 1965-73 1973-80 1980-86 1986 1987 1988
By commodity
Low- and middle-income economies 5.3 3.8 4.8 5.9 7.4 7.3
Manufactures 11.6 12.8 9.5 8.4 17.6 10.2
Food 2.4 4.2 3.1 -0.7 7.2 -0.1
Nonfood 2.1 0.4 2.2 5.3 1.4 9.8
Metals and minerals 4.8 6.5 1.2 6.4 13.4 -4.1
Fuels 5.6 -0.4 2.0 7.1 -7.0 6.6
Total reporting economies 8.7 4.6 2.9 4.9 6.0 6.3
Manufactures 10.7 6.1 4.5 2.2 7.0 8.4
Food 4.6 6.8 2.9 11.4 11.7 -2.1
Nonfood 3.1 0.9 2.5 1.4 17.2 11.1
Metals and minerals 6.8 8.6 1.5 5.9 2.1 -8.0
Fuels 8.7 0.5 -1.9 12.1 -4.3
By country group
Low- and middle-income economies 5.3 3.8 4.8 5.9 7.4 7.3
Manufactures 11.6 12.8 9.5 8.4 17.6 10.2
Primary goods 4.4 1.2 2.1 4.1 0.3 4.9
Low-income economies 9.6 2.2 2.9 9.2 3.9 7.6
Manufactures 1.8 8.5 10.0 15.5 23.5 15.8
Primary goods 11.2 1.1 0.6 6.7 -4.7 3.1
Middle-income economies 4.0 4.4 5.3 4.9 8.5 7.2
Manufactures 16.7 13.8 9.5 7.0 16.4 8.9
Primary goods 2.5 1.2 2.6 3.3 2.1 5.6
Sub-Saharan Africa 15.1 0.2 -2.0 -0.1 -1.8 1.2
Manufactures 7.6 5.6 4.8 2.9 4.3 5.3
Primary goods 15.4 0.0 -2.5 -0.4 -2.3 0.8
East Asia 9.7 8.7 9.7 15.9 14.3 9.4
Manufactures 17.5 15.5 13.3 19.2 24.9 12.8
Primary goods 7.3 4.7 5.7 11.7 -0.2 3.6
South Asia -0.7 5.8 4.1 9.1 11.2 6.7
Manufactures 0.6 8.2 3.5 10.4 20.9 7.5
Primary goods -1.8 3.1 5.0 7.7 0.0 5.5
Europe, Middle East, and
North Africa
Manufactures
Primary goods
Latin America and the Caribbean -0.9 0.9 3.4 -4.2 3.9 8.1
Manufactures 16.6 10.1 8.1 -10.6 5.5 19.3
Primary goods -1.7 -0.5 2.3 -2.2 3.5 4.8
Severely indebted middle-income
economies -0.3 1.9 3.9 -3.5 2.9 11.4
Manufactures 16.9 12.1 9.0 -8.3 6.1 20.0
Primary goods -1.1 0.3 2.5 -1.9 1.9 8.6
High-income economies 9.9 4.8 2.4 4.6 5.5 6.0
Manufactures 10.6 5.5 3.9 1.3 5.3 8.1
Primary goods 8.9 3.5 -0.6 12.9 6.0 1.3
OECD members 9.4 5.4 3.6 3.4 6.0 5.3
Manufactures 10.6 5.2 3.7 1.4 4.5 7.4
Primary goods 6.7 5.9 3.1 9.5 10.0 -0.4
Oil exporters 8.7 0.0 -4.6 12.5 -4.4 10.9
Manufactures 11.7 3.9 9.5 7.3 13.6 11.4
Primary goods 8.6 -0.1 -5.6 13.1 -6.4 10.9
164
Table A.9 Change in export prices and terms of trade, 1965 to 1988
(average annual percentage change)
Country group 1965-73 1973-80 1980-86 1986 1987 1988
Export prices
Low- and middle-income economies 6.2 14.7 -4.9 -10.6 11.0 6.1
Manufactures 6.4 8.2 -2.2 9.4 8.6 8.9
Food 5.9 8.6 -2.7 7.6 -7.0 14.4
Nonfood 4.6 10.2 -4.8 -1.1 21.7 2.8
Metals and minerals 2.5 4.7 -4.4 -4.8 11.4 28.2
Fuels 8.3 26.0 -10.1 -46.7 22.0 -15.0
High-income OECD members
Total 4.8 10.3 -1.4 12.2 11.4 8.3
Manufactures 4.5 10.9 -0.3 19.0 13.7 7.0
Terms of trade
Low- and middle-income economies 0.1 2.5 -2.7 -10.9 2.3 -0.2
Low-income economies -4.9 4.1 -3.0 -16.0 4.3 -1.1
Middle-income economies 1.8 1.9 -2.7 -9.1 1.8 0.1
Sub-Saharan Africa -8.5 4.8 -4.0 -21.5 2.9 -4.4
East Asia -0.6 1.2 -1.3 -6.1 0.5 1.1
South Asia 3.7 -3.4 1.5 3.1 1.2 2.0
Europe, Middle East, and North Africa .. .. .. ..
Latin America and the Caribbean 3.8 2.3 -3.3 -14.0 -2.3 1.7
Severely indebted middle-income
economies 4.3 1.7 -2.2 -11.4 1.6 0.2
High-income economies -1.2 -2.1 0.5 7.0 0.2 0.5
OECD members -1.0 -3.3 1.1 10.6 -0.5 0.7
Oil exporters 0.3 11.5 -7.3 -40.8 11.7 -16.5
165
Table A.10 Growth of long-term debt of low- and middle-income economies, 1970 to 1989
(average annual percentage change, nominal)
Country group 1970-73 1973 -80 1980-86 1987 1988 1989
Low- and middle-income economies
Debtoutstandinganddisbursed 18.0 22.2 13.2 13.2 -1.9 1.1
Official 15.3 17.8 15.3 21.9 0.8 5.7
Private 20.7 25.6 12.0 7.2 -4.0 -2.6
Low-income economies
Debt outstanding and disbursed 16.9 16.6 14.0 25.9 5.0 6.4
Official 14.9 14.2 13.8 23.1 4.0 8.0
Private 26.1 23.8 14.6 31.7 7.0 3.4
Middle-income economies
Debt outstanding and disbursed 18.4 24.2 13.0 9.6 -4.1 -0.7
Official 15.6 20.7 16.3 21.3 -1.1 4.3
Private 20.1 25.9 11.7 3.8 -5.9 -3.8
Sub-Saha ran Africa
Debt outstanding and disbursed 20.2 23.8 13.7 26.4 0.9 4.9
Official 17.2 22.3 17.5 27.0 1.6 11.3
Private 25.6 26.2 8.4 25.2 -0.5 -8.0
East Asia
Debt outstanding and disbursed 23.7 22.7 15.8 13.7 0.2 1.1
Official 27.0 17.9 16.4 24.8 1.5 3.3
Private 20.7 26.6 15.5 6.8 -1.4 -0.6
South Asia
Debt outstanding and disbursed 11.6 11.2 12.5 18.3 5.9 9.7
Official 12.4 10.4 9.3 16.7 4.4 5.5
Private 1.6 24.5 30.3 23.1 10.3 21.3
Europe, Middle East, and North Africa
Debt outstanding and disbursed 22.2 28.9 12.4 15.4 -2.3 4.1
Official 16.4 25.6 15.9 18.6 -2.0 5.9
Private 29.9 32.1 9.4 11.9 -2.6 2.0
Lotin America and the Caribbean
Debt outstanding and disbursed 16.8 21.6 13.1 7.4 -4.8 -3.9
Official 11.6 15.2 17.4 24.0 1.0 2.9
Private 18.9 23.5 12.2 2.6 -6.8 -6.5
Severely indebted middle-income economies
Debt outstanding and disbursed 16.9 23.9 14.4 9.4 -4.7 -3.1
Official 12.7 17.5 22.9 23.9 0.7 3.1
Private 18.5 25.7 12.5 4.4 -6.9 -5.9
166
Table A. 11 Investment, saving, and current account balance before official transfers, 1965 to 1988
(percentage of GNP)
Balance of payments: current
account balance
Gross domestic investment Gross national saving before official transfers
Country 1965-73 1973 -80 1980-88 1965-73' 1973 -80 1980-88 1965-73 1973-80 1980-88
Latin America and the Caribbean
*Argentina 19.7 23.4 14.4 20.1 22.6 10.0 0.4 -0.7 -4.5
*Bolivia 25.4 24.9 12.1 16.8 18.5 1.6 -8.6 -6.4 -10.5
*Brazil 21.3 23.9 20.7 23.1 19.3 18.5 1.9 -4.6 -2.2
*Chile 14.3 17.3 17.5 11.9 12.1 8.9 -2.4 -5.2 -8.7
Colombia 18.9 18.8 20.2 15.8 19.0 16.7 -3.2 0.2 -3.5
*Costa Rica 21.8 25.5 27.9 13.0 13.8 17.7 -8.8 -11.7 -10.2
*Ecuador 19.0 26.7 23.3 12.7 21.2 17.4 -6.2 -5.6 -5.9
Guatemala 13.3 18.7 13.4 11.6 16.4 9.3 -1.7 -2.3 -4.2
*Honduras 18.6 24.9 17.5 14.0 14.6 7.8 -4.6 -10.2 -9.7
Jamaica 32.0 20.2 23.8 23.7 13.6 13.7 -8.4 -6.6 -10.2
*Mexico 20.6 24.2 23.5 16.5 20.2 21.9 -4.0 -4.0 -1.6
*Nicaragua 20.1 18.7 21.8 13.5 9.0 -5.6 -6.5 -9.7 -27.4
*Peru 24.1 23.9 27.4 19.5 19.7 22.7 -4.6 -4.1 -4.6
*Uruguay 12.0 15.7 12.6 12.0 11.3 9.8 0.0 -4.4 -2.8
*Venezuela 31.1 34.2 22.5 31.9 35.8 24.2 0.8 1.6 1.7
Sub-Saha ran Africa
Cameroon 16.6 21.8 21.6 . . 17.0 16.9 -4.8 -4.7
*Congo, People's Rep. 29.3 34.0 38.2 4.2 10.3 24.5 -25.2 -23.6 -13.8
*Côte d'Ivoire 22.8 29.1 19.4 . . 16.8 8.0 . . -12.3 -11.5
Ethiopia 12.8 9.5 12.7 11.0 6.9 5.6 -1.8 -2.5 -7.1
Ghana 12.3 8.7 7.8 8.7 . . 2.7 -3.5 -1.8 -5.1
Kenya 22.6 26.2 25.7 17.2 16.4 18.7 -5.5 -9.8 -7.0
Liberia 19.1 28.7 15.0 27.5 6.9 -1.2 -8.1
Malawi 20.0 29.7 18.3 10.7 7.2 -19.0 -11.1
Niger 9.7 23.8 16.3 . . 9.7 1.9 . . -14.1 -14.3
Nigeria 16.3 22.8 13.9 11.8 24.4 12.3 -4.5 1.6 -1.6
*Senegal 14.7 17.5 16.0 . . 4.2 0.2 . . -13.3 -15.8
Sierra Leone 13.8 14.1 14.2 9.8 -1.0 7.8 -4.0 -15.1 -6.3
Sudan 11.9 16.2 13.4 11.0 9.6 . . -0.9 -6.6
Tanzania 19.9 23.9 19.0 17.3 13.8 10.2 -2.6 -10.0 -8.8
Zaire 13.7 15.0 15.2 29.3 8.6 5.4 15.6 -6.4 -9.8
Zambia 31.9 28.5 18.2 34.3 19.9 4.3 2.4 -8.6 -14.0
East Asia
Indonesia 15.8 24.5 27.6 13.7 24.6 24.4 -2.1 0.1 -3.2
Korea, Republic of 23.9 31.2 30.3 17.6 25.9 31.7 -6.3 -5.3 1.4
Malaysia 22.3 28.7 32.4 22.6 29.4 29.1 0.2 0.6 -3.4
Papua New Guinea 27.8 22.0 27.1 . . 11.7 5.2 . . -10.4 -21.9
*Philippines 20.6 29.1 22.1 18.7 24.3 17.9 -1.9 -4.8 -4.1
Thailand 24.3 26.9 25.6 20.5 21.9 21.4 -3.8 -5.0 -4.2
South Asia
India 17.2 21.3 24.0 14.0 21.0 21.7 -3.3 -0.3 -2.2
Pakistan 16.1 17.5 19.0 . . 11.7 15.1 . . -5.9 -3.9
Sri Lanka 15.8 20.6 26.4 11.2 13.4 15.9 -4.6 -7.2 -10.5
Europe, Middle East, and
North Africa
Algeria 32.6 44.6 36.0 30.5 39.0 35.3 -2.2 -5.6 -0.7
Egypt, Arab Republic of 14.0 29.3 27.7 9.3 18.2 15.6 -4.6 -11.1 -12.1
*Hungary . 32.0 27.9 . . .
*Morocco 15.1 25.9 26.1 13.6 16.8 19.8 -1.5 -9.0 -6.3
*Poland . 28.3 25.4 -5.6 -2.9
Portugal 26.6 29.7 30.3 . . . . 29.4 . . -0.9
Tunisia 23.3 29.9 27.9 17.8 23.2 22.1 -5.5 -6.7 -5.8
Turkey 18.5 21.8 22.9 17.5 18.1 20.2 -1.0 -3.7 -2.7
Yugoslavia 29.9 35.6 37.0 27.2 32.9 38.0 -2.6 -2.7 1.0
Note: An asterisk indicates a severely indebted middle-income economy. Figures in italics are for years other than those specified.
a. Excludes transfers, 1965-69.
167
Table A.12 Composition of debt outstanding, 1970 to 1988
(percentage of total long-term debt)
Debt from official sources Debt from private sources Debt at floating rate
Country 1970-72 1980-82 1988 1970-72 1980-82 1988 1973-75 1980-82 1988
Latin America and the Caribbean
*Argentina 12.6 9.0 17.7 87.4 91.0 82.3 6.6 29.2 80.2
*Bolivia 58.2 49.3 81.1 41.8 50.7 18.9 7.3 28.4 21.7
*Brazil 30.7 11.9 24.3 69.3 88.1 75.7 26.1 45.9 62.0
*Chile 46.0 11.1 26.9 54.0 88.9 73.1 8.3 23.4 65.5
Colombia 68.1 46.1 52.8 31.9 53.9 47.2 5.4 33.7 40.5
*Costa Rica 39.8 36.8 51.5 60.2 63.2 48.5 15.5 42.4 43.6
*Ecuador 51.4 30.6 38.7 48.6 69.4 61.3 8.2 36.5 66.7
Guatemala 47.5 71.0 74.5 52.5 29.0 25.5 3.5 5.6 10.5
*Honduras 73.8 62.2 79.7 26.3 37.8 20.4 1.8 19.1 18.2
Jamaica 7.4 68.3 82.9 92.6 31.7 17.1 4.7 17.3 25.7
*Mexico 19.5 10.9 18.1 80.5 89.1 81.9 32.0 61.7 72.0
*Nicaragua 65.3 58.0 80.2 34.7 42.0 19.8 44.2 42.1 20.6
*Peru 15.6 39.4 45.0 84.4 60.6 55.0 16.1 22.9 29.4
*Uruguay 44.2 21.1 21.7 55.8 78.9 78.3 10.1 28.5 69.6
*Venezuela 30.8 3.6 1.5 69.2 96.4 98.5 17.2 57.9 72.8
Sub-Saha ran Africa
Cameroon 82.2 56.6 69.3 17.8 43.4 30.6 1.8 11.3 8.2
*congo, People's Rep. 84.0 43.9 49.1 16.0 56.1 50.9 0.0 10.9 40.2
*Côte d'Ivoire 51.4 22.9 40.8 48.6 77.1 59.2 19.3 37.5 35.8
Ethiopia 87.3 90.9 83.5 12.7 9.1 16.5 1.5 2.1 6.7
Ghana 56.5 87.9 88.5 43.5 12.1 11.5 0.0 0.0 3.7
Kenya 58.3 55.0 73.4 41.7 45.0 26.6 2.1 10.2 3.6
Liberia 81.1 73.5 81.7 19.0 26.5 18.3 0.0 16.9 9.3
Malawi 85.8 72.2 95.6 14.2 27.8 4.4 2.3 21.9 3.8
Niger 97.0 41.0 76.9 2.9 59.0 23.1 0.0 13.4 6.4
Nigeria 68.8 14.7 37.9 31.2 85.3 62.1 0.7 48.8 40.1
*Senegal 59.2 67.8 93.0 40.8 32.2 7.0 24.6 12.0 2.5
Sierra Leone 60.6 66.4 81.4 39.4 33.6 18.6 3.8 0.1 0.6
Sudan 86.9 75.4 78.6 13.1 24.6 21.4 2.2 9.4 0.8
Tanzania 63.7 75.8 93.3 36.3 24.2 6.7 0.4 0.3 2.5
Zaire 42.5 65.9 90.0 57.5 34.1 10.0 32.8 11.9 5.4
Zambia 22.0 69.7 87.5 78.0 30.3 12.5 20.7 10.2 13.4
East Asia
Indonesia 72.1 51.9 58.3 27.9 48.1 41.7 4.9 15.0 25.8
Korea, Republic of 35.2 34.3 38.2 64.8 65.7 61.8 11.8 29.0 23.8
Malaysia 51.0 21.9 23.1 49.0 78.1 76.9 17.4 36.7 42.3
PapuaNewGuinea 6.1 23.9 33.7 93.8 76.1 66.3 0.0 23.2 17.1
*Phiippines 22.6 31.3 44.8 77.4 68.7 55.2 7.3 24.0 41.4
Thailand 40.1 39.1 46.0 59.9 60.9 54.0 0.4 22.4 27.2
South Asia
India 95.1 83.9 60.7 4.9 16.1 39.3 0.0 3.0 15.1
Pakistan 90.6 92.6 93.6 9.4 7.4 6.4 0.0 3.2 8.8
Sri Lanka 81.6 79.5 82.8 18.4 20.5 17.2 0.0 12.9 4.4
Europe, Middle East, and
North Africa
Algeria 47.2 20.5 23.7 52.8 79.5 76.3 34.0 23.5 35.8
Egypt, Arab Republic of 70.3 82.1 85.8 29.7 17.9 14.2 3.1 2.3 1.4
*Hungary 0.0 12.0 10.6 0.0 88.0 89.5 0.0 81.3 60.2
*Morocco 79.1 55.9 74.7 20.9 44.1 25.3 2.7 27.2 38.4
*Poland 0.0 0.0 65.5 0.0 0.0 34.5 0.0 0.0 63.3
Portugal 29.3 24.7 18.1 70.7 75.3 81.9 0.0 33.9 36.3
Tunisia 71.4 59.9 71.6 28.6 40.1 28.4 0.0 13.6 16.5
Turkey 92.5 63.3 54.8 7.5 36.7 45.2 0.8 23.0 33.5
Yugoslavia 37.5 23.6 34.5 62.5 76.4 65.5 3.2 10.1 46.8
Note: An asterisk indicates a severely indebted middle-income economy.
168
World Development Indicators
Contents
Key 171
Introduction and maps 172
Tables
1 Basic indicators 178
Production
2 Growth of production 180
3 Structure of production 182
4 Agriculture and food 184
5 Commercial energy 186
6 Structure of manufacturing 188
7 Manufacturing earnings and output 190
Domestic absorption
8 Growth of consumption and investment 192
9 Structure of demand 194
10 Structure of consumption 196
Fiscal and monetary accounts
11 Central government expenditure 198
12 Central government current revenue 200
13 Money and interest rates 202
Core international transactions
14 Growth of merchandise trade 204
15 Structure of merchandise imports 206
16 Structure of merchandise exports 208
17 OECD imports of manufactured goods: origin and composition 210
18 Balance of payments and reserves 212
External finance
19 Official development assistance from OECD & OPEC members 214
20 Official development assistance: receipts 216
21 Total external debt 218
22 Flow of public and private external capital 220
23 Total external public and private debt and debt service ratios 222
24 External public debt and debt service ratios 224
25 Terms of external public borrowing 226
Human resources
26 Population growth and projections 228
27 Demography and fertility 230
28 Health and nutrition 232
29 Education 234
30 Income distribution and ICP estimates of GDP 236
31Urbanization 238
32 Women in development 240
Technical notes 242
Box A.1 Basic indicators for economies with populations of less than 1 million 243
Box A.2 Selected indicators for nonreporting nonmember economies 244
Data sources 260
170
Key
In each table, economies are listed in their Figures in the colored bands are summary . = not available.
group in ascending order of GNP per cap- measures for groups of economies. The 0 and 0.0 zero or less than half the unit
ita except those for which no GNP per cap- letter w after a summary measure indicates shown.
ita can be calculated. These are italicized, that it is a weighted average; in, a median Blank means not applicable.
in alphabetical order, at the end of their value; t, a total.
group. The reference numbers below re- Figures in italics are for years or periods
All growth rates are in real terms.
flect the order in the tables. other than those specified.
Data cutoff date is April 30, 1990.
Afghanistan 37 Hong Kong 102 Papua New Guinea 51
Algeria 81 Hungary 82 Paraguay 63
Angola 77 India 22 Peru 66
Argentina 84 Indonesia 34 Philippines 44
Australia 104 Iran, Islamic Republic of 94 Poland 73
Austria 110 Iraq 95 Portugal 90
Bangladesh 5 Ireland 99 Romania 96
Belgium 108 Israel 100 Rwanda 20
Benin 28 Italy 106 Saudi Arabia 97
Bhutan 9 Jamaica 60 Senegal 46
Bolivia 43 Japan 120 Sierra Leone 41
Botswana 58 Jordan 67 Singapore 101
Brazil 76 Kampuchea, Democratic 39 Somalia 7
Burkina Faso 13 Kenya 24 South Africa 80
Burundi 15 Korea, Republic of 89 Spain 98
Cameroon 59 Kuwait 107 Sri Lanka 31
Canada 113 Lao People's Democratic Rep. 10 Sudan 36
Central African Republic 26 Lebanon 78 Sweden 117
Chad 3 Lesotho 30 Switzerland 121
Chile 68 Liberia 40 Syrian Arab Republic 69
China 21 Libya 93 Tanzania 4
Colombia 62 Madagascar 12 Thailand 57
Congo, People's Republic of the 55 Malawi 6 Togo 25
Costa Rica 70 Malaysia 74 Trinidad and Tobago 88
Côte d'Ivoire 50 Mali 14 Tunisia 64
Denmark 114 Mauritania 35 Turkey 65
Dominican Republic 49 Mauritius 72 Uganda 16
Ecuador 61 Mexico 71 United Arab Emirates 111
Egypt, Arab Republic of 48 Morocco 52 United Kingdom 105
El Salvador 56 Mozambique 1 United States 118
Ethiopia 2 Myanmar 38 Uruguay 83
Finland 116 Nepal 11 Venezuela 87
France 112 Netherlands 109 Viet Nam 42
Gabon 86 New Zealand 103 Yemen Arab Republic 45
Germany, Federal Republic of 115 Nicaragua 79 Yemen, People's Dem. Rep. of 33
Ghana 29 Niger 19 Yugoslavia 85
Greece 91 Nigeria 17 Zaire 8
Guatemala 54 Norway 119 Zambia 18
Guinea 32 Oman 92 Zimbabwe 47
Haiti 27 Pakistan 23
Honduras 53 Panama 75
Note: For economies with populations of less than 1 million, see Box Al; for nonreporting nonmember economies, see Box A.2.
171
Introduction
The World Development Indicators provide infor- of economies. Data in the other tables fall into the
mation on the main features of social and eco- following broad areas: production, domestic ab-
nomic development. Most of the data collected by sorption, fiscal and monetary accounts, core inter-
the World Bank are on the low- and middle-income national transactions, external finance, and human
economies. Because comparable data for high- resources.
income economies are readily available, these are In this edition, changes have been made to a
also included here. Additional information on number of tables. While these are described more
some of these and other countries may be found in fully in the technical notes, the objectives of the
other World Bank publications, notably the World changes may be of interest.
Bank Atlas, World Tables, World Debt Tables, and So- As an outgrowth of the World Development Re-
cial Indicators of Development. Data available for port's focus this year on poverty, the reporting of
nonreporting nonmembers are summarized in the income distribution (Table 30) has been modified.
main tables and shown by country in Box A.2. Illiteracy, female as well as overall, is reported as a
In these notes the term "country" does not im- basic indicator (Table 1).
ply political independence but may refer to any Social indicators have been rearranged. The age
territory whose authorities present for it separate structure of the population is again reported, at the
social or economic statistics. As in the past, the expense of estimates of population momentum
Bank classifies economies for certain operational (Table 26). Education (Table 29) adds indicators on
and analytical purposes according to gross na- primary net enrollment and pupil-teacher ratios
tional product (GNP) per capita. The definitions and drops male enrollment measures (which can
and data notes at the beginning of the main Report still be gauged from total and female measures).
provide a detailed description of the country Gender-relative indicators are added (Table 32) for
groups. risk of dying by age 5 as well as persistence in
Although every effort has been made to stan- primary school. To accommodate these additions,
dardize the data, full comparability cannot be en- Table 27 takes over the indicator showing when net
sured, and care must be taken in interpreting the reproduction rate will equal 1; estimates for births
indicators. The statistics are drawn from sources attended by health staff and infant mortality rates
thought to be most authoritative, but the data are are moved to Table 28.
subject to considerable margins of error. Variations Data on external debt are compiled directly by
in national statistical practices also reduce the com- the Bank on the basis of reports from developing
parability of data, which should thus be construed member countries through the Debtor Reporting
only as indicating trends and characterizing major System. Other data are drawn mainly from the
differences among economies, rather than taken as United Nations and its specialized agencies, the
precise quantitative indications of those differ- International Monetary Fund, and country reports
ences. to the World Bank. Bank staff estimates are also
The indicators in Table 1 give a summary profile used to improve currentness or consistency. For
172
most countries, national accounts estimates are ob- effort to improve the international comparability
tained from member governments through World and analytical significance of the indicators.
Bank economic missions. In some instances these As in the Report itself, the main criterion used to
are adjusted by Bank staff to conform to interna- classify economies in the World Development Indi-
tional definitions and concepts to provide better cators is GNP per capita. These income groups
consistency and to incorporate latest estimates. broadly distinguish countries at different stages of
For ease of reference, only ratios and rates of economic development. Many of the economies
growth are usually shown; absolute values are are further classified by geographical location.
generally available from other World Bank publica- Other classifications include severely indebted
tions, notably the 1989-90 edition of the World Ta- middle-income economies and all oil exporters. Se-
bles. Most growth rates are calculated for two pe- verely indebted middle-income economies is a
riods, 1965-80 and 1980-88, and are computed, group of nineteen countries that replaces previous
unless otherwise noted, by using the least-squares editions' 17 highly indebted economies, which
regression method. Because this method takes into contained a mix of low- and middle-income econo-
account all observations in a period, the resulting mies. For a list of countries in each group, see the
growth rates reflect general trends that are not un- definitions and data notes. The major classifica-
duly influenced by exceptional values, particularly tions used in the tables this year are 42 low-income
at the end points. To exclude the effects of infla- economies with per capita incomes of $545 or less
tion, constant price economic indicators are used in 1988, 54 middle-income economies with per cap-
in calculating growth rates. Details of this method- ita incomes of $546 to $5,999, including Angola, a
ology are given at the beginning of the technical new Bank member, and 25 high-income econo-
notes. Data in italics indicate that they are for years mies. For a final group of 9 nonreporting nonmem-
or periods other than those specifiedup to two her economies, paucity of data, differences in
years earlier for economic indicators and up to methods of computing national income, and diffi-
three years on either side for social indicators, culties of conversion are such that only aggregates
since the latter tend to be collected less regularly are shown in the main tables. Some selected indi-
and change less dramatically over short periods of cators for these countries, however, are included in
time. All dollar figures are U.s. dollars, unless oth- Box A.2.
erwise stated. The various methods used for con- Economies with populations of less than 1 mil-
verting from national currency figures are de- lion are not shown separately in the main tables,
scribed in the technical notes. but basic indicators for these countries and territo-
Differences between figures in this year's and ries, and for Puerto Rico, are in a separate table in
last year's edition reflect not only updating revi- Box A.1.
sions in the countries themselves but revisions to The summary measures are overall estimates.
historical series and changes in methodology. In Countries for which individual estimates are not
addition, the Bank reviews methodologies in an shown, because of size, nonreporting, or insuffi-
173
cient history, have been implicitly included by as- The table format of this edition follows that used
suming they follow the trend of reporting coun- in previous years. In each group, economies are
tries during such periods. This gives a more listed in ascending order of GNP per capita, except
consistent aggregate measure by standardizing those for which no such figure can be calculated.
country coverage for each period shown. Group These are italicized and in alphabetical order at the
aggregates also include countries with less than 1 end of the group deemed to be appropriate. This
million population, even though country-specific order is used in all tables except Table 19, which
data for these countries do not appear in the ta- covers only high-income OPEC and OECD coun-
bles. Where missing information accounts for a tries. The alphabetical list in the key shows the
third or more of the overall estimate, however, the reference number for each economy; here, too,
group measure is reported as not available. italics indicate economies with no estimates of
Throughout the World Development Indicators, GNP per capita. Economies in the high-income
the data for China do not include Taiwan, China. group marked by the symbol t are those classified
However, footnotes to Tables 14, 15, 16, and 18 by the United Nations or otherwise regarded by
provide estimates of the international transactions their authorities as developing.
for Taiwan, China. The colored bands show summary measures
Groups of economies
Countries are colored to show their income
group; for example, all low-income econo-
mies (those with a GNP per capita of $545 or
less in 1988) are colored yellow. The groups
are those used in the tables that follow.
uo,edssr
Low-income economies
Middle-income economies (UK)
High-income economies a
Data not available
UaI,rJ-
Er sIvoor
F ers Gea
(Fr)
TOeIo (RU)
rrren Sr,oa
(US)
SUrer
Nro,(NZ)
oinn
PuorIo Re ParsQsry
(US)
S Kills asS Nr
V,rRrn ISa
(US)
-. -,nrthdndlobao
174
totals, weighted averages, or median values section gives details of these sources, which con-
calculated for groups of economies. The methodol- tain comprehensive definitions and descriptions of
ogy used for computing the summary measures is concepts used. It should also be noted that country
described in the technical notes. For these num- notes to the World Tables provide additional expla-
bers, w indicates that the summary measures are nations of sources used, breaks in comparability,
weighted averages; m, median values; and t, to- and other exceptions to standard statistical prac-
tals. The coverage of economies is not uniform for tices that have been identified by Bank staff on
all indicators, and the variation from measures of national accounts and international transactions.
central tendency can be large; therefore readers
should exercise caution in comparing the summary Comments and questions relating to the World
measures for different indicators, groups, and Development Indicators should be addressed to:
years or periods. Socio-Economic Data Division
The technical notes and footnotes to tables should be International Economics Department
referred to in any use of the data. These notes outline The World Bank
the methods, concepts, definitions, and data 1818 H Street, N.W
sources used in compiling the tables. A separate Washington, D.C. 20433.
lMMnd1FealoS
Inleol Man
(UK)
CRannel Glenda UK)
NeSedanda
Bdgroei
LuMnnbouig
KeG Ra olUainnany
SinSnrland
Portapd
Gibiallar (U K) CRrea
InNennKNpe OR Alpharolair
Pakmlan
Arab Net
orAgMa
Bang ladeah
Myarrnrar (UK)
(Pent)
Mauritania
Mali Keger oern50 \
CraG MAre Guam (US)
Sudan
Ste Gaiebiat tundra
Guinne8rasaU\501
Genial
,
FaaU
.aj
Niperra
/1
Central
Minoan Rep
Uganda
Maldraea
i LaMe
'4 TraeSTinrilOM UI ten
PinibO InIMea
(US)
KrrrbAlr
EiauAodal Guinea
IITI PeAr a
GAO tonre and Prinaipe
unpe Re005a
Zaire 1ururdi
\ConrPO
SethORMa
Guleanean
Taraara
Itenda
tee
Mauriliun
/
Seunioo (F I) Caledmaa
(Fr)
eazKand
Mere
175
Population
0-15 million
15-50 million
The colors on the map show the general for each of the 121 countries in the main
50-100 million size of a country's population. For ex- tables; the technical note to that table
100 + million ample, countries with a population of gives data for an additional 55 reporting
less than 15 million are colored yellow. economies in Box Al, and 9 nonreport-
Data not available Note that Table 1 gives the population ing nonmember economies in Box A.2.
Fertility and mortality
Total fertility Infant mortality Life expectancy
Births per woman Deaths per 1,000 live births Years
8 150 80
100
60
50
2
40
0 0 0
1965 1988 2000 1965 1975 1988 1965 1975 1988
Low-income economies - High-income economies
Middle-income economies - Nonreporting nonmember economies
Note: For explanations of terms or methods, see the technical notes for Tables 27, 28, and 32.
176
Share of agriculture in GDP
0-9 percent The value added by a country's agricul-
10-19 percent
tural sector divided by the gross domes- absolute values of production. For
tic product gives the share of agriculture countries with high levels of subsistence
20-39 percent in GDP. The map classifies countries by farming, the share of agriculture in GDP
40 + percent those shares. For example, countries is difficult to measure because of diffi-
whose shares of agriculture in GDP culties in assigning subsistence farming
Data not available range from 0 to 9 percent are colored its appropriate value. For more details,
yellow. The shares say nothing about see the technical note for Table 3.
External balances of low- and middle-income countries
Percentage Low-income Middle.income Severely indebted Sub-Saharan
of GDP economies economies middle-income economies Africa
2 1970-79 1980-83 1984-88 1970-79 1980-83 1984-88 1970-79 1980-83 1984-88 1970-79 1980-83 1984-88
The current account balance (on goods, services, income, and all unrequited transfers)
represents transactions that add to or subtract from an economy's stock of foreign financial
Items. For some purposes, however, official unrequited transfers (mainly foreign aid grants,
food aid, and technical assistance) are treated as being closely akin to official capital move-
ments. A measure of the current account balance before official transfers, sometimes re-
ferred to as "total to be financed," is then appropriate. For further information, see the
technical note for Table 15 but note that the table reports dollar values for each measure in
1970 and 1988, whereas the chart traces period averages relative to GDP throughout the
ri
period.
- - - Current account balance - Current account balance Financed by net official transfers
(after official transfers) (before official transfers)
Note: For explanations of terms or methods, see the technical notes for Table 18.
177
Table 1. Basic indicators
GNP per capit& Life
Average annual Adult illiteracy
Area Average annual expectancy
rate of tnflatton (percent)
Population (thousands growth rate at birth
(percent) (Percent) (years) Female Total
(millions) of square Dollars
mid-1988 kilometers) 1988 1965-88 1965-80 1980-88 1988 1985 1985
Low-income economies 2,884.0 36,997 320 w 3.1 w 8.8w 8.9w 60 w 58 w 44w
China and India 1,904.0 12,849 340 w 4.0 w 2.8w 5.8w 63 w 56 w 42 w
Other low-income 980.0 24,149 280 w 1.5 w 18.2w 13.8w 54 w 62 w 51w
I Mozambique 14.9 802 100 . . . . 33.6 48 78 62
2 Ethiopia 47.4 1,222 120 -0.1 3.4 2.1 47 . . 38
3 Chad 5.4 1,284 160 -2.0 6.2 3.2 46 89 75
4 Tanzania 24.7 945 160 -0.5 9.9 25.7 53 .
5 Bangladesh 108.9 144 170 0.4 14.9 11.1 51 78 67
6 Malawi 8.0 118 170 1.1 7.2 12.6 47 69 59
7 Somalia 5.9 638 170 0.5 10.3 38.4 47 94 88
8 Zaire 33.4 2,345 170 -2.1 24.5 56.1 52 55 39
9 Bhutan 1.4 47 180 . . 8.9 48 .
10 La0PDR 3.9 237 180 . . . . 49 24 16
11 Nepal 18.0 141 180 . . 7.8 8.7 51 88 74
12 Madagascar 10.9 587 190 -1.8 7.7 17.3 50 38 33
13 BurkinaFaso 8.5 274 210 1.2 6.5 3.2 47 94 87
14 Mali 8.0 1,240 230 1.6 9.3 3.7 47 89 83
15 Burundi 5.1 28 240 3.0 6.4 4.0 49 74 66
16 Uganda 16.2 236 280 -3.1 21.2 100.7 48 55 43
17 Nigeria 110.1 924 290 0.9 13.7 11.6 51 69 58
18 Zambia 7.6 753 290 -2.1 6.4 33.5 53 33 24
19 Niger 7.3 1,267 300 -2.3 7.5 3.6 45 91 86
20 Rwanda 6.7 26 320 1.5 12.5 4.1 49 67 53
21 China 1,088.4 9,561 330 5.4 0.1 4.9 70 45 31
22 India 815.6 3,288 340 1.8 7.5 7.4 58 71 57
23 Pakistan 106.3 796 350 2.5 10.3 6.5 55 81 70
24 Kenya 22.4 580 370 1.9 7.3 9.6 59 51 41
25 Toga 3.4 57 370 0.0 6.9 6.1 53 72 59
26 Central African Rep. 2.9 623 380 -0.5 8.5 6.7 50 71 60
27 Haiti 6.3 28 380 0.4 7.3 7.9 55 65 62
28 Benin 4.4 113 390 0.1 7.5 8.0 51 84 74
29 Ghana 14.0 239 400 -1.6 22.8 46.1 54 57 47
30 Lesotho 1.7 30 420 5.2 8.0 12.2 56 16 26
31 SriLanka 16.6 66 420 3.0 9.4 11.0 71 17 13
32 Guinea 5.4 246 430 . . . . 43 83 72
33 Yemen, PDR 2.4 333 430 . . . . 4.5 51 75 59
34 Indonesia 174.8 1,905 440 4.3 34.2 8.5 61 35 26
35 Mauritania 1.9 1,026 480 -0.4 7.7 9.4 46
36 Sudan 23.8 2,506 480 0.0 11.5 33.5 50
37 Afghanistan . . 652 4.9
38 Myanmar 40.0 677 60
39 Kampuchea,Dem. . . 181
40 Liberia 2.4 111 6.3 ' 50 77 65
41 Sierra Leone 3.9 72 7.8 42 79 71
42 VietNam 64.2 330 66
Middle-income economies 1,068.0 t 37,352 t 1,930w 2.3 w 20.4w 66.7 w 66w 31w 26w
Lower-middle-income 741.7 t 24,451 t 1,380w 2.6 w 21.7w 80.8w 65w 32w 27w
43 Bolivia 6.9 1,099 570 -0.6 15.7 482.8 53 35 26
44 Philippines 59.9 300 630 1.6 11.7 15.6 64 15 14
45 YemenArabRep. 8.5 195 640 . . . . 11.6 47 97 86
46 Senegal 7.0 197 650 -0.8 6.5 8.1 48 81 72
47 Zimbabwe 9.3 391 650 1.0 5.8 12.1 63 33 26
48 Egypt, Arab Rep. 50.2 1,001 660 3.6 7.3 10.6 63 70 56
49 Dominican Rep. 6.9 49 720 2.7 6.8 16.8 66 23 23
50 Côted'Ivoire 11.2 322 770 0.9 9.5 3.8 53 69 57
51 PapuaNewGuinea 3.7 463 810 0.5 8.1 4.7 54 65 55
52 Morocco 24.0 447 830 2.3 6.0 7.7 61 78 67
53 Honduras 4.8 112 860 0.6 5.6 4.7 64 42 41
54 Guatemala 8.7 109 900 1.0 7.1 13.3 62 53 45
55 Congo, People's Rep. 2.1 342 910 3.5 6.7 0.8 53 45 37
56 ElSalvador 5.0 21 940 -0.5 7.0 16.8 63 31 28
57 Thailand 54.5 513 1,000 4.0 6.3 3.1 65 12 9
58 Botswana 1.2 582 1,010 8.6 8.1 10.0 67 31 29
59 Cameroon 11.2 475 1,010 3.7 8.9 7.0 56 55 44
60 Jamaica 2.4 11 1,070 -1.5 12.8 18.7 73 .
61 Ecuador 10.1 284 1,120 3.1 10.9 31.2 66 20 18
62 Colombia 31.7 1,139 1,180 2.4 17.4 24.1 68 13 12
63 Paraguay 4.0 407 1,180 3.1 9.4 22.1 67 15 12
64 Tunisia 7.8 164 1,230 3.4 6.7 7.7 66 59 46
65 Turkey 53.8 779 1,280 2.6 20.7 39.3 64 38 26
66 Peru 20.7 1,285 1,300 0.1 20.5 119.1 62 22 15
67 Jordan 3.9 89 1,500 2.2 66 37 25
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
178
GNP per capita
Average annual Adult illiteracy
Area Average annual expectancy
rate of nflatzon (percent)
Population (thousands growth rate s birth
(millions) of square Dollars (percent)
(prenit) (years) Female Male
mid-1988 kilometers) 1988 1965-88 1965-80 1980-88 1988 1985 1985
68 Chile 12.8 757 1,510 0.1 129.9 20.8 72 . . 6
69 Syrian Arab Rep. 11.6 185 1,680 2.9 8.3 12.9 65 57 40
70 Costa Rica 2.7 51 1,690 1.4 11.3 26.9 75 7 6
71 Mexico 83.7 1,958 1,760 2.3 13.0 73.8 69 12 10
72 Mauritius 1.1 2 1,800 2.9 11.8 7.8 67 23 17
73 Poland 37.9 313 1,860 . . . . 30.5 72 .
74 Malaysia 16.9 330 1,940 4.0 4.9 1.3 70 34 27
75 Panama 2.3 77 2,120 2.2 5.4 3.3 72 12 12
76 Brazil 144.4 8,512 2,160 3.6 31.5 188.7 65 24 22
77 Angola 9.4 1,247 . . . . 45 59
78 Lebanon . 10 . S 9.3 .
79 Nicaragua 3.6 130 . . -2.5 8.9 86.6 64
Upper-middle-income 326.3 t 12,901 1 3,240 w 2.3 w 18.9 w 45.0w 68w 31w 24w
80 South Africa 34.0 1,221 2,290 0.8 10.1 13.9 61 .
81 Algeria 23.8 2,382 2,360 2.7 10.5 4.4 64 63 50
82 Hungaiy 10.6 93 2,460 5.1 2.6 6.4 70 c C
83 Uruguay 3.1 177 2,470 1.3 57.8 57.0 72 4 5
84 Argentina 31.5 2,767 2,520 0.0 78.2 290.5 71 5 5
85 Yugoslavia 23.6 256 2,520 3.4 15.3 66.9 72 14 9
86 Gabon 1.1 268 2,970 0.9 12.7 0.9 53 47 38
87 Venezuela 18.8 912 3,250 -0.9 10.4 13.0 70 15 13
88 Trinidad and Tobago 1.2 5 3,350 0.9 14.0 5.3 71 5 4
89 Korea, Rep. of 42.0 99 3,600 6.8 18.7 5.0 70
90 Portugal 10.3 92 3,650 3.1 11.7 20.1 74 20 16
91 Greece 10.0 132 4,800 2.9 10.5 18.9 77 12 8
92 Oman 1.4 212 5,000 6.4 19.9 -6.5 64 .
93 Libya 4.2 1,760 5,420 -2.7 15.4 0.1 61 50 33
94 Iran, Islamic Rep. 48.6 1,648 15.6 . . 63 61 49
95 Iraq 17.6 438 . . 64 13 II
C
96 Romania 23.0 238 . . . . . . 70
Low- and middle-income 3,952.Ot 74,349t 750w 2.7w 16.5w 46.8w 62w 51w 40w
Sub-Saharan Africa 463.9 t 22,240 I 330w 0.2 w 12.5 w 15.5 w 51 w 65 w 52 w
East Asia 1,538.0 I 14,0171 540w 5.2w 8.7w 5.6w 66w 41 w 29w
South Asia 1,106.81 5,1581 320w 1.8w 8.3w 7.5 w 57w 72w 59w
Europe, M.East, & N.Africa 395.61 11,4201 2,000w 2.4w 13.2w 25.8w 64w 53 w 41 w
Latin America & Caribbean 413.61 20,293 1 1,840w 1.9w 29.4w 117.4w 67 w 19 w 17 w
Severely indebted 495.5 I 20,057 1 1,730 w 2.0w 28.3 w 107.9w 66w 23 w 20w
High-income economies 784.2 t 33,739 1 17,080 w 2.3 w 7.9w 4.9w 76 w
OECD members 751.1 I 31,057 t 17,470w 2.3 w 7.7 w 4.7 w 76 w
fOther 33.1 t 2,682 t 8,380w 3.1 w 15.9w 10.8 w 71w
97 tSaudi Arabia 14.0 2,150 6,200 3.8 17.2 -4.2 64 .
98 Spain 39.0 505 7,740 2.3 12.3 10.1 77 8 6
99 Ireland 3.5 70 7,750 2.0 12.0 8.0 74 .
100 lIsrael 4.4 21 8,650 2.7 25.2 136.6 76 7 5
101 tSingapore 2.6 1 9,070 7.2 4.9 1.2 74 21 14
102 tHong Kong 5.7 1
9,220b 6.3 8.1 6.7 77 19 12
103 New Zealand 3.3 269 10,000 0.8 10.2 11.4 75
104 Australia 16.5 7,687 12,340 1.7 9.3 7.8 76
105 United Kingdom 57.1 245 12,810 1.8 11.1 5.7 75
106 Italy 57.4 301 13,330 3.0 11.4 11.0 77 4 C
107 tKuwait 2.0 18 13,400 -4.3 16.4 -3.9 73 37 30
108 Belgium 9.9 31 14,490 2.5 6.7 4.8 75
109 Netherlands 14.8 37 14,520 1.9 7.5 2.0 77
110 Austria 7.6 84 15,470 2.9 6.0 4.0 75
111 tUnited Arab Emirates 1.5 84 15,770 0.1 71
112 France 55.9 552 16,090 2.5 8.4 7.1 76
113 Canada 26.0 9,976 16,960 2.7 7.1 4.6 77
114 Denmark 5.1 43 18,450 1.8 9.3 6.3 75
115 Germany, Fed. Rep. 61.3 249 18,480 2.5 5.2 2.8 75
116 Finland 5.0 338 18,590 3.2 10.5 7.1 75
117 Sweden 8.4 450 19,300 1.8 8.0 7.5 77
118 United States 246.3 9,373 19,840 1.6 6.5 4.0 76
119 Norway 4.2 324 19,990 3.5 7.7 5.6 77
120 Japan 122.6 378 21,020 4.3 7.7 1.3 78
121 Switzerland 6.6 41 27,500 1.5 5.3 3.8 77
Total reporting economies 4,736.2 t 108,088 I 3,470w 1.5 w 9.8w 14.1w 64w 50w 39w
Oil exporters 593.31 17,292 t 1,500w 2.0w 15.1 w 21.4w 61 w 43 w 35w
Nonreporting nonmembers 364.51 25,399 t 70w
Note: For economies with populations of less than 1 million, see Box A. 1. For nonreporting nonmember economies, see Box A.2. t Economies classified by United
Nations or otherwise regarded by their authorities as developing, a. See the technical notes. b. GNP data refer to GDP. c. According to Unesco, illiteracy is less
than 5 percent.
179
Table 2. Growth of production
Average annual growth rate (percent)
GDP Agriculture Industry (Manufacturing)a Services, etc.
1965-80 1980-88 1965-80 1980-88 1965-80 1980-88 1965-80 1980-88 1965-80 1980-88
Low-income economies 5.4w 6.4w 2.6w 4.4w 8.8w 8.7w 8.2w 9.7w 6.0w 6.0w
China and India 5.3w 8.7w 2.7w 5.4w 8.2 w 11.4w 8.0 w 10.5 w 6.3w 8.6w
Other low-income 5.5w 2.0w 2.3w 2.3w 10.0w 1.7w 9.1w 5.9w 5.7w 3.4w
1 Mozambique . . -2.8 . . -0.8 . . -7.1 . . . . . . -3.1
2 Ethiopia 2.7 1.4 1.2 -LI 3.5 3.5 5.1 3.7 5.2 3.6
3 Chad 0.1 3.9 -0.3 2.6 -0.6 7.7 . . . . 0.2 4.2
4 Tanzania 3.7 2.0 1.6 4.0 4.2 -2.0 5.6 -2.5 6.7 1.0
5 Bangladeshb 2.4 3.7 1.5 2.1 3.8 4.9 6.8 2.4 3.4 5.2
6 Malawi 5.6 2.6 4.1 2.7 6.3 3.0 6.7 2.4
7 Somalia 3.4 3.2 3.9 . . 2.3 -0.1 1.2
8 Zair&' 1.4 1.9 3.2 2.5 1.7 0.3
9 Bhutan
10 La0PDR
11 Nepal 1.9 4.7 1.1 4.4 .
12 Madagascar' 1.8 0.6 . . 2.2 -1.0
13 BurkinaFaso . . 5.5 . . 6.4 . . 3.7 5.5
14 Mali b 3.9 3.2 2.8 0.3 1.8 8.1 7.6 5.8
15 Burundi 5.6 4.3 6.7 3.1 17.4 5.8 6.0 6.1 1.4 6.3
16 Uganda 0.8 1.4 1.2 0.3 -4.1 6.4 -3.7 2.3 1.1 3.4
17 Nigeria 6.9 -1.1 1.7 1.0 13.1 -3.2 14.6 -2.9 7.6 -0.4
18 Zambia5 1.9 0.7 2.2 4.1 2.1 0.3 5.3 2.5 1.5 0.0
19 Niger 0.3 -1.2 -3.4 2.8 11.4 -4.3 . . 3.4 -8.0
20 Rwanda" 4.9 2.1 . . 0.3 3.6 . . 3.4 3.4
21 China' 6.4 10.3 2.8 6.8 10.0 12.4 95c 11.0 10.3 11.3
22 India 3.6 5.2 2.5 2.3 4.2 7.6 4.5 8.3 4.4 6.1
23 Pakistan 5.1 6.5 3.3 4.3 6.4 7.2 5.7 8.1 5.9 7.4
24 Kenya 6.4 4.2 4.9 3.3 9.8 2.8 10.5 4.6 6.4 5.5
25 Togob 4.5 0.5 1.9 4.2 6.8 0.0 -0.5 5.4 -1.7
26 CentralAfricanRep. 2.6 2.1 2.1 2.6 5.3 2.0 . . 0.2 2.0 1.7
27 Haitib 2.9 -0.2 . . . . . . .
28 Benin 2.1 2.4 . . 4.2 . . 5.8 . . 7.4 . . -1.0
29 Ghanat' 1.4 2.1 1.6 0.5 1.4 1.9 2.5 3.1 1.1 4.9
30 Lesotho 5.7 2.9 . . 1.8 . . 1.6 12.4 . . 4.1
31Sri Lanka 4.0 4.3 2.7 2.7 4.7 4.4 3.2 6.2 4.6 5.3
32 Guinea5 . . . . . . .
33 Yemen, PDR . . .. .. . . . . .. .. .. .
34 Indonesiab 8.0 5.1 4.3 3.1 11.9 5.1 12.0 13.1 7.3 6.4
35 Mauritania 2.0 1.6 -2.0 1.5 2.2 4.9 . . 6.5 -0.5
36 Sudan 3.8 2.5 2.9 2.7 3.1 3.6 5.0 4.9 2.0
37 Afghanistan 2.9
38 Myanmar
39 Kampuchea, Dem.
40 Liberia 3.3 -hi 2.2 -6.0 10.0 -5.0 2.4 -0.8
41 Sierra Leone 2.8 0.2 3.9 2.2 -0.8 -4.9 0.7 -2.0 4.3 0.7
42 VietNam
Middle-income economies 6.1w 2.9w 3.2 w 2.7w 5.9w 3.2w 8.2w 3.8w 7.2w 3.1w
Lower-middle-income 6.5w 2.6w 3.3 w 2.8 w 7.8w 2.5w 7.9w 2.4w 7.0w 2.6w
43 Bolivia1' 4.5 -1.6 3.8 2.1 3.9 -5.7 5.9 -5.6 5.4 -0.2
44 Philippines' 5.9 0.1 4.6 1.8 8.0 -1.8 7.5 -0.3 5.2 0.7
45 YemenArabRep' . . 6.5 . . 2.9 . . 11.5 . . 12.8 . . 6.2
46 Senegalb 2.0 3.3 1.3 3.2 4.8 3.8 3.5 3.4 1.3 3.2
47 Zimbabwe 5.0 2.7 . . 2.5 1.7 . . 2.1 3.4
48 Egypt, Arab Rep. 6.8 5.7 2.7 2.6 6.9 5.1 . . 5.6 9.4 7.3
49 DominicanRep.' 7.9 2.2 6.3 0.8 10.9 2.5 8.9 1.0 7.3 2.5
50 Côted'Ivoire 6.8 2.2 3.3 1.6 10.4 -2.4 9.1 8.2 8.6 4.2
51 PapuaNewGuinea' 4.1 3.2 3.2 2.7 . . 5.6 1.0 . . 2.0
52 Moroccob 5.6 4.2 2.4 6.6 6.1 2.8 . . 4.2 6.8 4.2
53 Honduras 5.0 1.7 2.0 1.1 6.8 0.8 7.5 1.9 6.2 2.4
54 Guatemala1' 5.9 -0.2 . . . . . . . . . . . . .
55 Congo,PeoplesRep.b 6.3 4.0 3.1 2.0 9.9 5.1 . . 7.1 4.7 3.5
56 ElSalvador" 4.3 0.0 3.6 -1.4 5.3 0.4 4.6 0.3 4.3 0.7
57 Thailand" 7.2 6.0 4.6 3.7 9.5 6.6 11.2 6.8 7.6 6.8
58 Botswanab 14.2 11.4 9.7 -5.9 24.0 15.1 13.5 5.0 11.5 10.3
59 Cameroont' 5.1 5.4 4.2 2.4 7.8 7.8 7.0 6.2 4.8 5.5
60 Jamaica1' 1.3 0.6 0.5 0.9 -0.1 0.0 0.4 1.6 2.7 0.9
61 Ecuador5 8.7 2.0 3.4 4.3 13.7 2.2 11.5 0.6 7.6 1.1
62 Colombia 5.8 3.4 4.5 2.4 5.7 5.1 6.4 2.9 6.4 2.7
63 Paraguay" 6.9 1.7 4.92.7 9.1 0.1 7.0 1.3 7.5 2.0
64 Tunisia 6.6 3.4 5.52.4 7.4 2.4 9.9 6.0 6.5 4.4
65 Turkey 6.3 5.3 3.23.6 7.2 6.7 7.5 7.9 7.6 5.1
66 Pens1' 3.9 1.1 1.03.6 4.4 0.4 3.8 1.6 4.3 1.2
67 Jordan .. 4.2 . 6.0 . . . 3.6 3.4 4.4
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
180
Average annual growth rate (percent)
GDP Agriculture industry (Manufacturing) Services, etc.
1965-80 1980-88 1965-80 1980-88 1965-80 1980-88 1965-80 1980-88 1965-80 1980-88
68 ChiIe' 1.9 1.9 1.6 3.8 0.8 2.2 0.6 2.0 2.7 1.3
69 Syrian Arab Rep)' 8.7 0.5 4.8 0.5 11.8 1.4 . . 9.0 0.2
70 Costa Ricat' 6.2 2.4 4.2 2.5 8.7 2.3 . . . . 6.0 2.5
71 Mexic&' 6.5 0.5 3.2 1.2 7.6 -0.1 7.4 0.2 6.6 0.7
72 Mauritius 5.2 5.7 . . 4.0 . . 9.0 11.4 4.6
73 Polandt'
74 Malaysiat' 73 46 37 61 73 36
75 Panamat' 5.5 2.6 2.4 2.5 5.9 -0.8 4.7 0.7 6.0 3.5
76 Brazil 8.8 2.9 3.8 3.5 10.1 2.6 9.8 2.2 9.5 3.1
77 Angola
78 LebanonL -1.2 .. .. .. .. .. ..
79 Nicaragua 2.6 -0.3 3.3 -0.2 4.2 0.4 5.2 0.6 1.4 -0.9
Upper-middle-income 5.6w 3.3 w 3.2w 2.5w 4.7w 3.7w 7.5 w 3.7 w
80 South Africa 3.8 1.3 . . 1.7 . . 0.2 . . 0.2 . . 2.6
81 Algeriat' 6.8 3.5 5.7 5.6 7.1 3.8 9.5 6.1 6.7 2.7
82 Hungaryt' 5.6 1.6 2.7 2.4 6.4 1.0 . . 6.2 1.9
83 Uruguay 2.4 -0.4 1.0 0.3 3.1 -1.8 . . -0.5 2.3 0.2
84 Argentinab 3.5 -0.2 1.4 1.4 3.3 -0.8 2.7 -0.2 4.0 -0.2
85 Yugoslavia 6.0 1.4 3.1 1.2 7.8 1.3 . . 5.5 1.4
86 Gabonb 9.5 -0.2 .. .. .. .. ..
87 Venezuela' 3.7 0.9 3.9 3.8 1.5 -0.1 5.8 3.3 6.3 1.4
88 Trinidad and Tobago 5.1 -6.1 0.0 4.5 5.0 -8.6 2.6 -9.5 5.8 -3.4
89 Korea, Rep. oft' 9.6 9.9 3.0 3.7 16.4 12.6 18.7 13.5 9.6 8.9
90 Portugaib 5.3 0.8 . . -0.9 . . 1.0 . . . . . . 1.3
91 Greece 5.6 1.4 2.3 -0.1 7.1 0.4 8.4 0.0 6.2 2.5
92 Oman" 13.0 12.7 . . 9.4 . . 15.1 . . 37.9 . . 12.2
93 Libya 4.2 10.7 . . 1.2 13.7 . . 15.5
94 Iran,IslamicRep. 6.2 4.5 2.4 10.0 . . 13.6
95 Iraq
96 Romania
Low- and middle-income 5.8w 4.3 w 2.8w 3.7w 6.8w 5.3w 8.2w 5.9w 6.9w 3.9w
Sub-Saharan Africa 4.8w 0.8w 1.3w 1.8w 9.4w -0.8w 8.7w 0.2w 5.0w 1.4w
East Asia 7.2 w 8.5 w 3.2 w 5.7 w 10.8 w 10.3 w 10.6 w 10.2 w 8.6 w 8.0w
South Asia 3.7w 5.1w 2.5w 2.5w 4.4w 7.3w 4.6w 7.9w 4.5 w 6.1 w
Europe, M.East, & N.Africa 6.1w .. 3.5w 4.9w .. .. .. 8.6w
Latin America & Caribbean 6.0w 1.5w 3.3w 2.5w 6.0w 1.1 w 7.0w 1.3 w 6.6w 1.6w
Severely indebted 6.0w 1.5w 3.2w 2.7w 6.2w 1.0w 7.1 w 1.3w 6.6w 1.6w
High-income economies 3.7 w 2.8 w 0.8 w 2.3 w 3.2 w 1.9 w 3.6w 3.2 w 3.7 w 3.0 w
OECD members 3.6w 2.9w 0.8 w 2.2 w 3.1 w 2.2 w 3.6 w 3.2 w 3.7 w 3.0w
tOther 8.0w -1.3 w . . 12.7 w . . -7.0w 6.0 w . . 4.6 w
97 tSaudi Arabiab 11.3 -3.3 4.1 15.2 11.6 -6.0 8.1 7.9 10.5 2.6
98 Spainb 4.6 2.5 2.6 0.9 5.1 0.4 5.9 0.4 4.1 2.1
99 Ireland 5.0 1.7 2.2 1.7 0.6
100 tlsraelb 6.8 3.2 . . . . . . . . . . . . .
101 tSingapore" 10.1 5.7 2.8 -5.1 11.9 4.5 13.2 4.8 9.4 6.6
102 tHong Kong 8.6 7.3
103 New Zealandt' 2.4 2.2 3.3 .. 4.2 3.5 2.0
104 Australiab 4.0 3.3 2.7 4.4 3.0 2.2 1.3 1.1 5.7 3.7
105 United Kingdom 2.4 2.8 -1.6' 3.4 05d 1.9 1.5 22d 2.5
106 Italyb 4.3 2.2 0.8 1.0 4.0 1.1 5.1 1.9 4.1 2.7
107 tKuwait h
1.2 -1.1 . . 23.6 . . -2.3 . . 1.4 . . -0.9
108 Belgiumb 3.8 1.4 0.4 2.5 4.4 1.1 4.6 2.3 3.7 1.2
109 Netherlands" 3.8 1.6 4.7 4.1 4.0 0.8 4.8 . . 4.4 1.6
110 Austria" 4.1 1.7 2.1 0.7 4.3 1.1 4.5 1.6 4.2 1.9
111 United Arab Emirates . . -4.5 9.3 -8.7 2.7 . . 3.7
12 France" 4.0 1.8 1.0 2.3 4.3 0.1 5.2 -0.4 4.6 2.4
113 Canada 5.1 3.3 0.8 2.7 3.5 3.0 3.9 3.6 6.9 3.2
114 Denmark 2.7 2.2 0.9 3.3 1.9 3.4 3.2 2.4 3.2 2.0
115 Germany, Fed. Rep)' 3.3 1.8 1.4 1.9 2.8 0.4 3.3 1.0 3.7 2.1
116 Finland 4.0 2.8 0.0 -1.1 4.2 2.7 4.9 3.0 4.8 3.3
117 Sweden 2.9 1.7 -0.2 1.8 2.3 2.9 2.4 2.9 3.4 0.8
118 United Slates" 2.7 3.3 1.0 3.2 1.7 2.9 2.5 3.9 3.4 3.3
119 Norway 4.4 3.8 -0.4 1.3 5.7 4.7 2.6 1.8 4.1 3.4
120 Japan" 6.5 3.9 0.8 0.8 8.5 4.9 9.4 6.7 5.2 3.1
121 Switzerlandb 2.0 1.9
Total reporting economies 4.1w 3.1w 2.0w 3.2w 3.9w 2.5w 4.3w 3.8w 4.2w 3.2w
Oil exporters 6.4w 1.0w 3.1w 2.7w 6.3 w -0.1 w 7.7w 3.7w 7.6w 2.4w
Nonreporting nonmembers
a. Because manufacturing is generally the most dynamic part of the industrial sector, its growth rate is shown separately. b. GDP and Its components are at purchaser
values. c. World Bank estimate. d. Data ivfer to the period 1973-80.
181
Table 3. Structure of production
Distribution of gross domestic product (percent)
GDP'
(millions of dollars) Agriculture Industry (Manufacturing)5 Services, etc.
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
Low-income economies 161,340 t 886,620 t 44w 33 w 28w 36w 21 w 28w 32w
China and India 117,730 t 610,250 t 44w 32w 32w 40w 24w 24w 28w
Other low-income 42,660 t 273,080 t 45 w 33 w 17 w 27 w 9w 38 w 40 w
I Mozambique 1,100 . . 62 . 20 . . . . . . 18
2 Ethiopia 1,180 4,950 58 42 14 17 7 12 28 40
3 Chudc 290 920 42 47 15 18 12 15 43 35
4 Tanzania 790 2,740 46 66 14 7 8 4 40 27
5 Bangladeshe 4,380 19,320 53 46 11 14 5 7 36 40
6 Maluwi 220 1,080 50 37 13 18 . . 37 44
7 Somalia 220 970 71 65 6 9 3 5 24 25
8 Zairec 3,140 6,470 21 31 26 34 16 7 53 35
9 BhutanC . 300 . . 44 . . 28 . . 6 . . 28
10 La0PDR . 500 59 . . 20 . . 7 . . 2!
II Nepal 730 2,860 65 56 II 17 3 6 23 27
12 MadagascarC 670 1,880 31 41 16 16 11 . . 53 43
13 BurkinaFaso 260 1,750 53 39 20 23 . . 13 27 38
14 Malic 260 1,940 65 49 9 12 5 5 25 39
15 Burundi 150 960 . . 56 . IS . . 10 . . 29
16 Uganda 1,100 3,950 52 72 13 7 8 6 35 20
17 Nigeria 5,850 29,370 54 34 13 36 6 18 33 29
18 ZambiaC 1,060 4,000 14 14 54 43 6 25 32 43
19 Niger 670 2,400 68 36 3 23 2 9 29 41
20 Rwandac 150 2,310 75 38 7 22 2 15 18 40
21 Chinac 67,200 372,320 44 32 39 46 31d 33d 17 21
22 India 50,530 237,930 44 32 22 30 16 19 34 38
23 Pakistan 5,450 34,050 40 26 20 24 14 17 40 49
24 Kenya 920 7,380 35 31 18 20 11 12 47 49
25 TogoC 190 1,360 45 34 21 21 10 8 34 45
26 CentralAfricanRep. 140 1.080 46 44 16 12 4 8 38 44
27 Haitic 350 2,500 . 31 . 38 . . 15 . . 31
28 Benin 220 1,710 59 40 8 13 . . 6 33 47
29 GhanaC 2,050 5,230 44 49 19 16 10 10 38 34
30 Lesotho 50 330 65 21 5 28 1 13 30 52
31 SriLanka 1,770 6,400 28 26 21 27 17 15 51 47
32 Guineac 2,540 30 32 .. 5 . 38
33 Yemen, PDR . . 840 . 16 . . 23 . . . . 61
34 IndonesiaC 3,840 83,220 56 24 13 36 8 19 31 40
35 Mauritania 160 900 32 38 36 21 4 . . 32 41
36 Sudan 1,330 11,240 54 33 9 15 4 8 37 52
37 Afghanistan 600 . . . . .
38 Myanmar . . . . . . . . .
39 Kampuchea. Dem. . . . 0 ' ' . . S
40 Liberia 270 990 27 37 40 28 3 5 34 35
4! SierraLeone 320 1,270 34 46 28 12 6 3 38 42
42 VietNam . . . . . . .
Middle-income economies 199,900 t 2,200,750 t 20 w 12 w 33 w 40 w 19 w 24 w 46 w 50 w
Lower-middle-income 111,840 t 1,061,910 t 22 w 14 w 28 w 38 w 19 w 25 w 50 w 50 w
43 Boliviuc 710 4,310 23 24 31 27 15 17 46 49
44 PhilippinesC 6,010 39,210 26 23 28 34 20 25 46 44
45 YemenArabRep.c . 5,910 . . 23 . . 26 . . 12 . . 50
46 SenegaF 810 4,980 25 22 18 29 14 19 56 49
47 Zimbabwe 960 5,650 18 11 35 43 20 31 47 46
48 Egypt,ArabRep. 4,550 34,330 29 21 27 25 . . 14 45 54
49 Dominican Rep.0 890 4,630 23 23 22 34 l6 16 55 43
50 COte d'Ivoire 760 7,650 47 36 19 25 II 16 33 39
51 Papua New Guineac 340 3,520 42 34 18 31 . . 9 41 36
52 MowccoC 2,950 21,990 23 17 28 34 16 18 49 49
53 Honduras 460 3,860 40 25 19 21 12 13 41 54
54 Guutemalac 1,330 8,100 . . . . . . . . . . .
55 Congo, People's Rep.° 200 2,150 19 15 19 30 . . 8 62 54
56 ElSalvadorc 800 5,470 29 14 22 22 18 18 49 65
57 Thailandc 4,390 57,950 32 17 23 35 14 24 45 48
58 Botswanuc 50 1,940 34 3 19 55 12 5 47 42
59 Cameroonc 810 12,900 33 26 20 30 10 13 47 44
60 Jamaicac 970 3,220 10 6 37 42 17 21 53 52
61 EcuadorC 1,150 10,320 27 15 22 36 18 21 50 49
62 Colombia 5,910 39,070 27 19 27 34 19 20 47 47
63 Paraguayc 440 6,040 37 30 19 25 16 17 45 46
64 Tunisia 880 8,750 22 14 24 32 9 16 54 54
65 Turkey 7,660 64,360 34 17 25 36 16 26 41 46
66 Pent 5,020 25,670 18 12 30 36 17 24 53 51
67 Jordan 3,900 10 25 12 65
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
182
GDP Distribution of gross domestic product (percent)
(millions of dollars) Agriculture Industry (Manufacturing)b Services, etc.
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
68 Chilec 5,940 22,080 9 40 24 52
69 Syrian Arab Rep.c 1,470 14,950 29 38 22 16 49 46
70 Costa Ricac 590 4,650 24 18 23 28 . . 53 54
71 Mexico' 21,640 176,700 14 9 27 35 20 26 59 56
72 Mauritius 190 1,600 16 13 23 33 14 25 61 54
73 Polandc
74 Malaysiac 3,130 34,680 28 . . 25 . 9 . 47
75 Panama 660 5,490 18 9 19 18 12 8 63 73
76 Brazil 19,450 323,610 19 9 33 43 26 29 48 49
77 Angola .. .. .. .. ..
78 Lebanonc 1,150 . . 2 . . 21 . . . . 67
79 Nicaragua 570 3,200 25 21 24 34 18 24 51 46
Upper-middle-income 88,2001 1,138,840 t 18w 39w 42w
80 South Africa 10,540 78,970 10 6 42 45 23 25 48 49
81 Algeria 3,170 51,900 15 13 34 43 11 12 51 44
82 Hungaiy . . 28,000 . 14 . . 37 . . 49
83 Uniguay 930 6,680 15 II 32 29 24 53 60
84 Argentinac 16,500 79,440 17 13 42 44 33 31 42 44
85 Yugoslavia 11,190 61,710 23 14 42 49 35 37
86 Gabonc 230 3,320 26 11 34 51 . . . . 40 38
87 Venezuelac 9,820 63,750 6 6 40 36 22 55 58
88 TnnidadandTobago 690 4,400 8 5 48 31 . . 9 44 64
89 Korca,Rep.ofc 3,000 171,310 38 11 25 43 18 32 37 46
90 Portugalc 3,740 41,700 . . 9 . . 37 . . . . 54
91 Greece 5,270 40,900 24 16 26 29 16 18 49 56
92 Omanc 60 8,150 61 3 23 43 0 6 16 54
93 Libya 1,500 5 63 3 33
94 Iran, Islamic Rep. 6,170 26 36 12 38
95 Iraq 2,430 . . 18 46 . . 8 . . 36
96 Romania
Low- and middle-income 363,680 3,060,950 31w 18w 31w 39 w 20w 38 w 44w
Sub-Saharan Africa 27,490 I 149,550 1 43 w 34 w 18w 27 w 9w 39 w 39 w
East Asia 92,420 893,410 41 w 22 w 35 w 43 w 27w 24 w 36 w
SouthAsia 64,5101 312,070 44w 33 w 21 w 27 w 15w 17w 35w 39 w
Europe, M.East, & N.Africa 69,200 24 w 34 w 40 w
Latin America & Caribbean 95,330 808,340 16w iii 33w 39 w w 27w 51w 52 w
Severely indebted 105,150t 897,3901 17w lOw 34w 39w 23w 27w 50w 52w
High-income economies 1,391,7001 13,867,530 t 5w 41w 30w 55w
OECD members 1,373,380 I 13,603,0601 5w 41 w 30 w 55 w
tOther 11,020 I 234,3701 6w 54w 11w 41w
97 rSaudi Arabiac 2,300 72,620 8 8 60 43 9 8 31 50
98 Spainc 23,750 340,320 15 6 36 37 27 49 57
99 Ireland 2,340 27,820 . . 10 . . 38 52
100 tlsraelc 3,590 44,960 . . . . . , . . . . . . .
101 tSingaporec 970 23,880 3 0 24 38 15 30 74 61
102 tHong Kong 2,150 44,830 2 0 40 29 24 22 58 70
103 New Zea1and 5,410 39,800 . 10 . . 33 . . 23 . . 57
104 Australiac 22,920 245,950 9 4 39 34 26 18 51 61
105 UnitcdKingdom 89,100 702,370 3 2 46 42 34 27 51 56
106 Italyc 72,150 828,850 10 4 37 40 25 27 53 56
107 tKuwait' 2,100 19,970 0 1 70 51 3 10 29 48
108 Belgiumc 16,840 153,810 5 2 42 34 31 24 53 64
109 Netherlandsc 19,640 228,280 . . 5 . . 37 . , 24 , . 58
110 Austriac 9,480 127,200 9 4 46 45 33 32 45 51
111 tUnited Arab Emirates . . 23,850 . . 2 . . 55 9 44
112 Francec 99,660 949,440 8 4 38 37 27 27 54 59
113 Canada 46,730 435,860 6 4 41 40 26 23 53 56
114 Denmark 8,940 90,530 9 5 36 37 23 25 55 58
115 Germany,Fed.Rep.c 114,790 1,201,820 4 2 53 51 40 44 43 47
116 Finland 7,540 91,690 16 7 37 43 23 29 47 50
117 Sweden 19,880 159,880 6 4 40 43 28 30 53 54
118 UnitedStatesc 700,970 4,847,310 3 2 38 33 28 22 59 65
119 Norwayc 7,080 91,050 8 4 33 45 21 21 59 51
120 Japan 91,110 2,843,710 9 3 43 41 32 29 48 57
121 Switzerlandc 13,920 184,830 . . . . .
Total reporting economies 1,755,990 1 17,018,400 lOw 39w 28w 52w
Oil exporters 77,910 1 921,070 19w 12w 32w 35w 14w 16w 48w 51w
Nonreporting nonmembers
a. See the technical notes. b. Because manufacturing is generally the most dynamic part of the industrial sector, its share of GDP is shown separately. c. GDP and its
components are shown at purchaser values. d. World Bank estimate. e. Services, etc. include the unallocated share of GDP.
183
Table 4. Agriculture and food
Value added Food aid Fertilizer consumption Average index of
in agriculture Cereal imports in cereals (hundreds of grams food production
(millions of (thousands of (thousands of ofplant nutrient per per capita
current dollars) metric tons) metric tons) hectare of arable land)
(1979-81 =100)
1970 1988 1974 1988 1974/75 1987/88 1970/71 1987/88 1986-88
Low-income economies 89,156 t 289,209 t 22,757 t 32,469 t 6,002 t 6,977 171w 706w 112 w
China and India 60,621t 197,397t 1l,294t 18,502t 1,582t 570t 241w 997w 117w
Other low-income 28,109 t 90,501 t 11,462 t 13,967 t 4,420 t 6,407 72w 318w 102 w
I Mozambique 679 62 527 34 466 22 21 83
2 Ethiopia 931 2,090 118 1,157 54 825 4 39 89
3 Chada 142 430 37 61 20 15 7 17 103
4 Tanzania 473 1,795 431 120 148 72 31 92 89
5 Bangladesha 3,636 8,882 1,866 3,010 2,076 1,397 157 770 92
6 Malawi 119 402 17 44 0 109 52 203 85
7 Somalia 167 636 42 236 111 152 29 40 100
8 Zairea 585 2,008 343 415 1 177 8 9 98
9 Bhutan 130 3 30 0 2 10 118
10 LaoPDR 297 53 115 8 21 2 6 123
11 Nepal 579 1,601 18 52 0 21 27 232 100
12 Madagascar a 266 775 114 217 7 76 61 21 97
13 BurkinaFaso 126 685 99 128 28 38 3 57 116
14 Malia 207 952 281 109 107 26 3! 59 97
15 Burundi 159 535 7 15 6 4 5 20 100
16 Uganda 929 2,859 36 28 29 14 2 121
17 Nigeria 5,080 10,105 389 333 7 0 2 94 103
18 Zambiaa 191 568 93 128 5 140 73 183 96
19 Niger' 420 866 155 151 73 19 1 8 83
20 Rwandaa 135 880 3 11 19 8 3 20 82
21 Chinaa 36,705 120,779 6,033 15,517 0 347 410 2,361 132
22 India 23,916 76,618 5,261 2,985 1,582 223 137 517 105
23 Pakistan 3,352 8,935 1,274 602 584 657 146 829 107
24 Kenya 484 2,265 15 86 2 119 238 421 89
25 Togoa 85 464 6 110 11 16 3 76 88
26 CentralAfrican Rep. 60 469 7 40 1 6 12 4 87
27 Haitf' 782 83 205 25 154 4 25 95
28 Benin 121 680 7 121 9 11 36 49 110
29 Ghanaa 1,030 2,577 177 228 33 110 13 38 108
30 Lesotho 23 67 48 107 14 55 10 125 80
31 SriLanka 545 1,685 951 940 271 361 555 1,094 79
32 Guineaa 761 63 222 49 26 19 6 93
33 Yemen,PDR 132 148 459 0 31 118 85
34 Indonesiaa 4,340 20,055 1,919 1,702 301 319 133 1,068 117
35 Mauritania 58 339 115 219 48 51 11 55 89
36 Sudan 757 3,716 125 702 46 604 28 40 89
37 Afghanistan 5 236 10 104 24 97
38 Myanmar 26 9 0 21 125
39 Kampuchea,Dem. 223 150 226 6 11 2
40 Liberia 91 42 103 3 56 63 94 92
41 Sierra Leone 108 581 72 119 10 58 17 3 tO!
42 VietNam 1,854 417 64 65 513 651 117
Middle-income economies 49,929 t 274,894 t 42,929 t 77,525 t 1,925 t 5,915 t 330 w 648 W 99 W
Lower-middle-income 31,837 t 149,950 t 24,538 t 38,685 t 1,631 t 5,911 t 304 w 611 w 100 w
43 Boliviaa 202 1,023 209 328 22 290 7 19 95
44 Philippinesa 1,996 9,005 817 1,322 89 471 287 612 90
45 YemenArabRep/' 118 1,387 158 754 33 160 1 58 118
46 SenegaF 208 1,100 341 461 27 109 17 40 106
47 Zimbabwe 214 615 56 93 0 14 446 505 81
48 Egypt, Arab Rep. 1,942 7,257 3,877 8,479 610 1,738 1,312 3,505 111
49 DominicanRep.a 345 1,082 252 601 16 278 334 556 95
50 Côted'Ivoire 462 2,728 172 494 4 1 74 90 104
51 PapuaNewGuineaa 240 1,196 71 180 . . 0 58 381 92
52 Moroccoa 789 3,770 891 1,643 75 340 117 376 106
53 Honduras 212 956 52 144 31 146 156 190 76
54 Guatemalaa . 138 166 9 320 298 656 92
55 Congo,PeoplesRep.a 49 331 34 113 2 1 114 25 92
56 ElSalvador' 292 740 75 217 4 177 1,043 1,262 87
57 Thailanda 1,837 9,795 97 303 0 97 59 328 101
58 Botswana' 28 57 21 150 5 53 15 7 69
59 Camemona 364 3,405 81 282 4 2 34 71 97
60 Jamaicaa 93 180 340 418 1 208 873 914 101
61 Ecuador a 401 1,547 152 563 13 33 133 232 97
62 Colombia 1,806 7,364 502 864 28 90 287 945 100
63 Paraguaya 191 1,788 71 2 10 2 98 69 106
64 Tunisia 245 1,187 307 2,116 59 393 76 222 111
65 Turkey 3,383 11,125 1,276 380 16 1 157 637 98
66 Pema 1,351 . . 637 1,857 37 355 300 622 96
67 Jonian 44 377 171 874 79 29 74 362 111
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
184
Value added Food aid Fertilizer consumption
Average index of
in agriculture Cereal imports in cereals (hundreds of grams
food production
(millions of (thousands of (thousands of ofplant nutrient per
per capita
current dollars) metric tons) metric tons) hectare of arable land)
(1979-81 = 100)
1970 /988 1974 1988 1974/75 1987/88 1970/71 1987/88 1986-88
68 Chile a -
- 558 1,737 339 323 14 313 544 105
69 Syrian Arab Rep.a 435 5,728 339 1,044 47 26 68 404 93
70 CostaRicaa 222 853 110 318 1 235 1,001 1,806 89
71 Mexicoa 4,462 15,958 2,881 5,650 32 232 753 93
72 Mauritius 30 211 160 177 22 32 2,095 3,075 106
73 Polanda 4,185 3,114 1 1,678 2,223 108
74 Malaysiaa 1,198 1,023 2,387 1 489 1,596 106
75 Panamaa 149 63 93 3 387 657 95
76 Brazil 4,392 27,849 2,485 1,387 31 21 186 485 108
77 Angola 149 313 0 103 33 29 87
78 Lebanona 136 354 537 26 54 1,354 671
79 Nicaraguaa 193 44 206 3 87 215 433 71
Upper-middle-income 18,5861 18,392 1 38,8401 2941 4t 377 w 728 w 98 w
80 South Africa 1,362 4,624 127 212 422 541 84
81 Algeriaa 492 6,546 1,816 6,130 54 4 163 320 106
82 Hungaiya 1,010 4,019 408 203 1,497 2,595 111
83 Uruguay 268 744 70 44 6 0 485 420 103
84 Argentinaa 2,250 10,089 0 3 26 45 97
85 Yugoslavia 2,212 8,518 992 297 770 1,328 100
86 Gabona 60 375 24 59 . . 46 97
87 Venezuelaa 826 3,753 1,270 3,054 . . . 170 1,580 94
88 TrinidadandTobago 40 210 208 212 880 450 71
89 Korea,Rep.of 2,311 18,561 2,679 9,369 234 2,450 3,920 98
90 PortugaF 3,180 1,861 1,383 428 1,026 103
91 Greece 1,569 6,461 1,341 859 .. 861 1,542 97
92 Omana 40 52 293 417
93 Libya 93 . 612 1,435 62 416 119
94 Iran,JslamicRep. 2,120 . . 2.076 4,644 60 658 99
95 Iraq 579 870 4,442 34 397 105
96 Romania 1,381 30 565 1,301 117
Low- and middle-income 140,632 1 558,529 65,686t 109,9941 7,928 1 12,891 238 w 680 w 109 w
Sub-Saharan Africa 15,421 50,851 4,108 1 8,214t 9101 3,5831 33 w 85 w 94 w
East Asia 50,416 1 194,504 17,259 t 37,462 9231 1,327t 380 w 1,326w 123 w
South Asia 32,884 1 103,969 9,4041 7,6451 4,522 1 2,660 135 w 586 w 100 w
Europe, M.East, & N.Africa 19,480 I 23,247 I 39,511 1 1,0101 2,880t 474 w 960 w 103 w
Latin America & Caribbean 18,622 81,703 11,5371 16,946 1 563 1 2,4411 176 w 451 w 100 w
Severely indebted 21,461 1 91,994 1 16,4961 21,2421 664 1 2,134 1 336w 613 w 100w
High-income economies 87,956 I 73,681 1 78,693 I 999 w 1,172 w 101 w
OECD members 87,148 t 70,120 1 68,319 1,001 w 1,163w 102 w
tOther 765 I 3,562 1 10,374 1 645 w 3,131 w 85 w
97 tSaudiArabiaa 219 5,526 482 5,179 54 3,678
98 Spaina 15,721 4,675 3,416 593 989 115
99 Ireland 559 2,712 640 408 3,067 6,815 98
100 tIsrae1t 295 1,176 1,799 53 2 1,401 2,237 101
101 tSingapore' 44 101 682 1,054 2,500 13,750 87
102 tHong Kong 62 174 657 690 24
103 New Zealan&' 897 2,826 92 158 7,745 7,086 108
104 Australiaa 2,173 8,227 2 27 232 286 97
105 UnitedKingdom 2,971 . 7,540 3,620 2,631 3,555 109
106 ttalya 8,465 31,062 8,101 7,502 896 1,901 101
107 (Kuwait a 8 . . 101 417 . 750
108 Belgiuma 934 3,042 4,585c 4,620c 5,648c 5,098c I 13c
109 Netherlandsa 1,827 8,456 7,199 6,354 7,493 6,877 113
110 Austriaa 992 3,844 164 135 2,426 2,214 107
111 (United Arab Emirates 453 132 458 . . 1,632
112 France a 9,366 30,780 654 951 2,435 2,990 106
113 Canada 3,280 . . 1,513 595 191 484 111
114 Denmark 882 4,130 462 211 2,234 2,330 118
115 Germany, Fed. Rep.a 5,951 16,541 7,164 4,181 4,263 4,208 112
116 Finland 1,205 5,153 222 230 1,930 2,164 99
117 Sweden 1,394 4,582 300 189 1,646 1,357 99
118 United Statesa 27,828 89,811 460 1,811 816 937 94
119 Norway 624 2,876 713 460 2,443 2,704 110
120 Japan' 12,467 65,384 19,557 28,018 3,547 4,327 104
121 Switzerlanda . 1,458 788 3,831 4,306 108
Total reporting economies 226,275 1 . . 139,368 t 188,687 1 7,981 1 12,894 I 480 w 831w 107w
Oil exporters 22,367 t 114,085 t 18,105 I 45,171 1 1,038 t 2,155 1 143 w 607w 104w
Nonreporting nonmembers 15,326 I 43,584 1 572 w 1,268 w 110 w
a. Value added in agriculture data are at purchaser values. b. Value added in agriculture data refer to net domestic product at factor cost. c. Includes Luxembourg.
185
Table 5. Conmiercial energy
Energy consumption
Average annual energy per capita Energy imports
growth rate (percent)
(kilograms as a percentage of
Energy production Energy consumption of oil equivalent) merchandise exports
1965-80 1980-88 1965-80 1980-88 1965 1988 1965 1988
Low-income economies 10.0 w 4.4 w 8.2 w 5.3 w 126 w 322 w 6w 11w
China and India 9.1 w 5.8 w 8.8 w 5.5 w 146 w 424 w 4w 6w
Other low-income 12.4 w 0.3 w 5.0 w 4.2 w 73 w 122 w 8w 17w
1 Mozambique 19.8 -39.0 2.2 2.0 81 86 13 43
2 Ethiopia 7.5 6.4 4.1 2.2 10 20 8 59
3 Chad . . . . . . . . .
. 18 22
4 Tanzania 7.3 3.4 3.7 2.4 37 36 10 44
5 Bangladesh 14.0 7.6 . 50 24
6 Malawi 18.2 4.3 8.0 0.2 25 42 7 9
7 Somalia . . . 16.7 1.8 14 66 8 20
8 Zaire 9.4 4.3 3.6 1.5 74 74 6 2
9 Bhutan . . . . . . . . .
10 LaoPDR -0.2 4.2 2.0 24 37
11 Nepal 18.4 11.6 6.2 9.4 6 23 10 29
12 Madagascar 3.9 9.3 3.5 1.5 34 39 8 45
13 BurkinaFaso . . . . 10.5 . . 7 .. 11 7
14 Mali 38.6 9.3 7.0 2.9 14 21 16 31
15 Bunindi . . 11.6 6.0 8.2 5 20 11 8
16 Uganda -0.5 4.1 -0.5 4.0 36 25 1 14
17 Nigeria 17.3 -1.4 12.9 6.6 34 150 7 2
18 Zambia 25.7 1.4 4.0 0.6 464 376 6 6
19 Niger . . 15.7 12.5 3.2 8 43 9 15
20 Rwanda 8.8 5.5 15.2 4.4 8 41 10 63
21 China 10.0 5.5 9.8 5.5 178 580 0 2
22 India 5.6 6.9 5.8 5.4 100 211 8 20
23 Pakistan 6.5 6.6 3.5 6.2 135 210 7 27
24 Kenya 13.1 8.3 4.5 0.2 110 94 13 41
25 Togo 2.9 11.4 10.7 -1.5 27 54 4 6
26 CentralAfrican Rep. 6.7 0.7 2.2 3.7 22 30 9 14
27 Haiti 4.3 8.4 1.6 24 57 6 13
28 Benin . . 7.8 9.9 4.7 21 46 10 72
29 Ghana 17.7 -6.1 7.8 -3.5 76 125 6 15
a
30 Lesotho . . . . . .
31 Sri Lanka 10.4 8.6 2.2 3.7 106 162 6 25
32 Guinea 16.5 2.3 1.1 56 78 .
33 Yemen,PDR . . . . -6.4 2.5 . . 653 .
34 Indonesia 9.9 1.0 8.4 4.5 91 229 3 14
35 Mauritania . . 9.5 0.3 48 111 3 6
36 Sudan 17.8 1.6 2.0 0.9 67 58 5 37
37 Afghanistan 15.7 0.9 5.6 11.3 30 . . 8
38 Myanmar 8.4 4.8 4.9 5.1 39 74 4 5
39 Kampuchea,Dem. . . 6.5 7.6 2.1 19 . . 7
40 Liberia 14.6 -0.9 7.9 -8.4 182 164 6 12
41 SierraLeone . . . . 0.8 -0.8 109 76 II 11
42 VietNam 5.3 1.2 -2.6 1.9 93
Middle-income economies 3.7 w 3.3 w 6.6 w 2.7 w 585 w 1,086 w 8w 9w
Lower-middle-income 7.0w 4.7 w 6.8 w 2.4 w 429 w 797 w 9w lOw
43 Bolivia 9.5 -0.2 7.7 -1.5 156 249 1 3
44 Philippines 9.0 9.0 5.8 -1.0 160 244 12 16
45 YemenArabRep. . . 21.0 10.7 7 102 .
46 Senegal . . . . 7.4 -1.2 79 155 8 18
47 Zimbabwe -0.7 -0.0 5.2 0.6 441 527 7 5
48 Egypt,ArabRep. 10.7 6.3 6.2 6.7 313 607 11 4
49 DominicanRep. 10.9 5.4 11.5 2.3 127 332 8 36
50 Côtcd'Ivoire 11.1 . . 8.6 . . 101 . . 5 12
51 PapuaNewGuinea 13.7 6.4 13.0 2.4 56 243 II 9
52 Momcco 2.5 -0.1 7.9 2.4 124 239 5 17
53 Honduras 14.0 6.7 7.6 3.2 III 203 5 14
54 Guatemala 12.5 7.4 6.8 -0.5 150 168 9 14
55 Congo,People's Rep. 41.1 7.5 7.8 4.4 90 245 10 5
56 ElSalvador 9.0 3.4 7.0 1.6 140 215 5 14
57 Thailand 9.0 35.2 10.1 6.6 82 331 II 14
58 Botswana 8.8 2.6 9.5 2.3 191 415
59 Cameroon 13.0 14.9 6.3 6.1 67 152 6
60 Jamaica -0.9 5.1 6.1 -2.8 703 855 12 22
61 Ecuador 35.0 -3.6 11.9 0.5 162 573 11 3
62 Colombia 1.0 11.1 6.0 2.2 413 755 1 4
63 Paraguay . . 11.8 9.7 4.6 84 224 16 12
64 Tunisia 20.4 -1.7 8.5 5.6 170 499 12 10
65 Turkey 4.3 8.9 8.5 7.3 258 822 12 24
66 Peru 6.6 -1.1 5.0 0.7 395 478 3
67 Joidan . . 7.5 9.3 6.9 226 723 33 42
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
186
Energy consumption
Average annual energy
per capita Energy imports
growth rate (percent)
(kilograncs ass percentage of
Energy production Energy consumption of oil equivalent) merchandise exports
1965-80 1980-88 1965-80 /980-88 1965 /988 1965 1988
68 Chile 1.8 2.8 3.0 1.6 652 832 5 4
69 Syrian Arab Rep. 56.3 4.6 12.4 4.1 212 913 13 18
70 Costa Rica 8.2 7.7 8.8 2.9 267 557 8 12
71 Mexico 9.7 2.4 7.9 0.7 605 1,305 4
72 Mauritius 2.1 6.7 7.2 2.8 160 402 6 9
73 Poland 4.0 2.1 4.8 1.1 2,027 3,453 . . 15
74 Malaysia 36.9 15.6 6.7 5.9 313 784 11 5
75 Panama 6.9 9.8 5.8 4.2 576 1,627 61 57
76 Brazil 8.6 9.0 9.9 3.6 286 813 14 13
77 Angola
78 Lebanon 2.0 -4.0 2.0 3.4 713 . . 51
79 Nicaragua 2.6 -0.2 6.5 1.8 172 252 6 42
Upper-middle-income 2.3 w 2.4 w 6.4w 2.9w 912w 1,766w 7w 8w
80 South Africa 5.1 4.9 4.3 3.3 1,744 2,439
81 Algeria 5.3 5.2 11.9 6.2 226 1,094 0 2
82 Hungary 0.8 1.4 3.8 1.0 1,825 3,068 12 15
83 Uruguay 4.7 10.6 1.3 -1.3 765 769 13 10
84 Argentina 4.5 2.6 4.3 2.1 975 1,523 8 4
85 Yugoslavia 3.5 2.3 6.0 3.1 898 2,159 7 18
86 Gabon 13.7 0.3 14.7 3.0 153 1,134 3 1
87 Venezuela -3.1 -0.9 4.6 2.1 2,319 2,354 0 0
88 Trinidad and Tobago 3.8 -3.4 6.6 0.2 2,776 5,255 60 10
89 Korea, Rep. of 4.1 9.7 12.1 5.5 238 1,515 18 10
90 Portugal 3.6 4.7 6.5 2.7 506 1,324 13 21
91 Greece 10.5 8.2 8.5 2.5 615 1,986 29 11
92 Oman 23.0 10.3 30.5 7.7 14 2,012 77 2
93 Libya 0.6 -5.0 18.2 4.8 222 2,719 2 1
94 Iran, Islamic Rep. 3.6 5.1 8.9 2.7 537 875 0
95 Iraq 6.2 5.3 7.4 5.3 399 781 0
96 Romania 4.3 0.7 6.6 0.8 1,536 3,459
Low- and middle-income 5.5 w 3.8w 7.2 w 3.8 w 253 w 525 w 7w 9w
Sub-Saharan Africa 15.3 w 0.1 w 5.6w 2.7 w 71w 95 w 7w 11 w
East Asia 9.8 w 5.2 w 9.4 w 5.3 w 168 w 474 w 6w 8w
South Asia 5.8w 5.6w 5.7 w 5.5 w 99 w 182 w 7w 22 w
Europe, M.East, & N.Africa 4.4 w 3.2w 6.2 w 2.8 w 746 w 1,343 w 9w 16w
Latin America & Caribbean 1.9w 2.6w 6.9 w 1.9w 515 w 952 w 8w 7w
Severely indebted 2.4w 2.2w 6.1w 1.6w 675w 1,084w 7w 9w
High-income economies 3.1 w 0.1 w 3.1w 1.1 w 3,707 w 5,098 w 11w 9w
OECD members 2.1 w 1.7w 3.0 w 1.0w 3,748 w 5,181 w 11w 9w
'(Other 7.7 w -6.6 w 5.7 w 3.2w 1,943w 3,028w 7w 8w
97 '(Saudi Arabia 11.5 -10.3 7.2 5.0 1,759 3,098 0 I
98 Spain 3.6 7.0 6.5 1.5 901 1,902 31 17
99 Ireland 0.1 5.9 3.9 1.5 1,504 2,610 14 5
100 '(Israel -15.2 -14.2 4.4 1.4 1,574 1,972 14 13
101 tSingapore . . 10.8 -0.6 670 4,464 17 16
102 '(Hong Kong . . . . 8.4 3.9 413 1,544 4 2
103 New Zealand 4.7 6.9 3.6 3.5 2,622 4,339 7 5
104 Australia 10.5 5.7 5.0 1.6 3,287 5,157 11 6
105 UnitedKingdom 3.6 1.7 0.9 0,8 3,481 3,756 13 6
106 Italy 1.3 0.9 3.7 -0.3 1,568 2,608 16 9
107 '(Kuwait -1.6 -0.0 2.1 3.7 . . 4,637 0 0
108 Belgiumb -3.9 9.1 2.9 -0.0 3,402 4,781 .
109 Netherlands 15,4 -1.8 5.0 1.3 3,134 5,235 12 9
110 Austria 0.8 -0.7 4.0 0.6 2,060 3,396 10 7
Ill '(United Arab Emirates 14.7 -0.1 36.6 8.4 105 6,481 4 2
112 France -0.9 7.8 3.7 0.4 2,468 3,704 16 9
113 Canada 5.7 4.2 4.5 1.6 6,007 9,683 8 4
114 Denmark 2.6 57.6 2.4 0.9 2,911 3,902 13 6
115 Germany, Fed. Rep. -0.1 0.4 3.0 -0.2 3,197 4,421 8 6
116 Finland 3.8 8.1 5.1 2.6 2.233 5,550 II 9
117 Sweden 4.9 6.6 2.5 2.4 4,162 6,617 12 6
118 United States 1.1 0.4 2.3 0.9 6,535 7,655 8 14
119 Norway 12.4 6.6 4.1 3.2 4,650 9,516 11 4
120 Japan -0.4 3.7 6.1 1.9 1,474 3,306 19 14
121 Switzerland 3.7 1.7 3.1 2.0 2,501 4,193 8 4
Total reporting economies 4.0w 1.4w 4.0w 1.9w 1,010w 1,289w lOw 9w
Oil exporters 5.8w -1.0w 7.4w 3.0w 389w 790w 5w Sw
Nonreporting nonmembers 4.6 w 2.8 w 4.4w 2.8 w 2,509 w 4,777 w
a. Figures for the South African Customs Union. comprising South Africa, Namibia, Lesotho, Botswana, and Swaziland. are included in South African data. Trade
among the component territories is excluded. h. Includes Luxembourg.
187
Table 6. Structure of manufacturing
Value added Distribution of manufacturing value added (percent; current prices)
in manufacturing Machinery and
(millions of Food, beverages, Textiles and transport
current dollars) and tobacco clothing equipment Chemicals Other'
1970 1987 1970 1987 1970 1987 1970 1987 1970 1987 1970 1987
Low-income economies 45,816 1
China and India 38,394 t
Other low-income 6,285 t
IMozambique 51 13 5 3 28
2 Ethiopia 149 564 46 49 31 19 0 2 2 4 21 27
3 Chadb 51 106 .. .. .. .. .. .. .. ..
4 Tanzania 116 146 36 32 28 23 5 9 4 6 26 30
5 Bang1adesh' 387 1,313 30 26 47 32 3 5 11 16 10 21
6 Malawi . . .
. 51 33 17 21 3 3 10 17 20 26
7 Somalia 26 51 88 59 6 13 0 2 1 13 6 13
8 Zaire1' 286 374 38 16 7 10 . . 29
9 Bhulan . . 16 . . . . . . . .
10 La0PDR 47 . S
11 Nepal 32 165 . . . . . . . . . . . . . . . .
12 Madagascar1' 118 . . 36 49 28 25 6 5 7 9 23 /2
13 BurkinuFaso . . 220 69 9 2 . . 1 19
14 Mali1' 25 100 36 . . 40 4 5 . 14
15 Burundi 16 96 53 . . 25 . . 0 6 . . 16
16 Uganda 158 162 40 20 . . 2 . 4 34
17 Nigeria 543 5,196 36 . . 26 . . I . . 6 . . 31
18 Zambia1' 181 568 49 46 9 Ii 5 14 10 6 27 23
19 Niger' 30 189 . . . . . . , . ' ' ' . .
20 Rwanda1' 8 314 86 65 0 3 3 0 2 5 8 28
21 China" 30,466c 92,800c . 12 . . 14 . 25 . . 11 . , 38
22 India 7,928 43,331 13 12 21 15 20 26 14 15 32 32
23 Pakistan 1,462 5,001 24 34 38 19 6 9 9 14 23 24
24 Kenya 174 839 31 38 9 11 18 13 7 II 35 27
25 Toga" 25 94 . . .
26 Central African Rep. 12 79 . . . , , . , . . .
27 Haiti" . . . . . . . , ,
28 Benin 19 83 . . . . . .
29 Ghana" 252 501 34 16 . 4 4 41
30 Lesotho 3 37 . . . . . . . . . . , . ,
31 SriLanka 321 967 26 . . 19 . . 10 11 . . 33
32 Guinea' 117 .
33 Yemen,PDR
34 Indonesia1' 994 12,876 Ii 1 48
35 Mauritania 10
36 Sudan 140 1,111 39 . . 34 , . 3 . . 5 19
37 Afghanistan
38 Myanmar
39 Kampuchea, Dem. .
40 Liberia 15 47
41 Sierra Leone 22 24 65 . . 1 0 . . 4 30
42 VietNam . . . . . . . . . . ,
Middle-income economies 63,448 t 451,574
Lower-middle-income 36,839 t 225,539
43 Bolivia1' 135 675 33 30 34 7 I I 6 4 26 58
44 Philippines" 1,622 8,424 39 43 8 8 8 8 13 10 32 30
45 YemenArabRep)' 10 578 20 . . 50 . . . . . . / . . 28
46 Senegal" 141 868 51 48 19 15 2 6 6 7 22 24
47 Zimbabwe 293 1,637 24 34 16 16 9 9 11 9 40 32
48 Egypt, Arab Rep. . . 4,244 17 35 9 . . 12 . . 27
49 Dominican Rep.b 275 843 74 5 I . . 6 . . 14
50 Côted'Ivoire 149 1,191 27 16 10 5 42
51 Papua New Guinea" 35 227 25 37 . . 5 33
52 Morocco" 641 3,398
53 Honduras 91 515 58 . . 10 . . I . . 4 . . 28
54 Guatemala1' .
. 42 45 14 9 4 3 12 13 27 30
55 Congo, People's Rep." . . 147 65 42 4 10 1 4 8 8 22 35
56 ElSalvador 194 809 40 37 30 14 3 5 8 16 18 28
57 Thailand" 1,130 11,543 43 29 13 18 9 13 6 7 29 33
58 Botswana" 5 82 . . 52 . . 7 . . 0 . . 6 . . 36
59 Camemon" 119 1,632 47 16 5 4 . . 28
60 Jamaica1' 221 639 46 . . 7 . . . . . . 10 . . 36
61 Ecuador1' 305 2,073 43 32 14 13 3 6 8 8 32 41
62 Colombia 1,487 7,244 31 36 20 14 8 8 11 13 29 30
63 Paraguay1' 99 735 56 . . 16 . . I . . 5 . . 21
64 Tunisia 121 1,265 29 20 18 19 4 4 13 9 36 47
65 Turkey 1,930 15,863 26 17 15 15 8 15 7 11 45 43
66 Peru" 1,430 6,232 25 25 14 12 7 8 7 1/ 47 43
67 Jordan 32 552 21 22 14 3 7 I 6 7 52 67
Note: For data comparability and coverage, see the technical notes. Figures in italics are foryears other than those specified.
188
Value added Distribution of manufacturing value added (percent; current prices)
in manufacturing Machinery and
(millions of Food, beverages, Textiles and transport
current dollars) and tobacco clothing equipment Chemicals 0ther
1970 1987 1970 1987 1970 1987 1970 1987 1970 1987 1970 1987
68 Chileb 2,092 17 26 12 7 11 4 5 8 55 56
69 Syrian Arab Rep.b 37 24 40 10 3 3 2 15 20 48
70 Costa Rica" 48 12 6 7 28
71 Mexico" 8,449 36,381 28 24 15 12 13 14 11 12 34 39
72 Mauritius 26 358 75 27 6 52 5 2 3 4 12 15
73 Poland" 20 14 19 16 24 31 8 6 28 34
74 Malaysia" 500 26 21 3 6 8 22 9 15 54 37
75 Panama" 127 422 41 47 9 7 1 3 5 8 44 34
76 Brazil 10,429 78,995 16 15 13 10 22 21 10 12 39 42
77 Angola
78 Lebanonb 27 19 1 3 49
79 Nicaragua" 159 759 53 54 14 12 2 2 8 10 23 22
Upper-middle-income 26,419
80 South Africa 3,914 17,790 15 14 13 9 17 19 10 12 45 47
81 Algeriab 682 7,196 32 20 20 17 9 13 4 3 35 47
82 Hungaryb 12 7 13 10 28 35 8 12 39 37
83 Uruguay 1,690 34 29 21 19 7 11 6 9 32 32
84 Argentina" 5,750 22,024 20 21 18 12 17 15 7 11 38 41
85 Yugoslavia 10 14 15 17 23 25 7 7 45 37
86 Gabon" 37 7 6 6 44
87 Venezuelab 2,140 10,779 30 18 13 7 9 8 8 8 39 59
88 TrinidadandTobago 198 416 18 43 3 6 7 6 2 4 70 42
89 Korea,Rep.of" 1,880 42,286 26 12 17 17 11 28 11 8 36 35
90 Portugal"' 18 17 19 25 13 13 10 10 39 34
91 Greece 1,642 7,170 20 20 20 25 13 10 7 8 40 36
92 Omanb 464 S
93 Libya 81 64 5 0 12 20
94 Iran, Islamic Rep. 1,501 30 20 18 6 26
95 Iraq 325 26 14 14 9 7 10 3 16 50 50
96 Romania 0 S
L ow- and middle-income 110,929
Sub-Saharan Africa 3,376
East Asia 38,947
South Asia 10,359 t 51,621
Europe, M.East, & N.Africa
Latin America & Caribbean 34,698 t 180,987
Severely indebted 38,028 t 202,164
High-income economies 608,635 t 2,895,002
OECD members 604,270 2,855,538
tOther 2,387 t 32,313
97 tSaudi Arabia" 372 6,085 S
98 Spain" 66,408 13 18 IS 9 16 21 II 11 45 40
99 Ireland 785 31 27 19 6 13 26 7 17 30 25
100 lIsrael" . . . . 15 12 14 8 23 32 7 8 41 39
101 tSingapore" 379 5,741 12 5 5 4 28 52 4 12 51 27
102 tHong Kong 1,013 9,825 4 6 41 40 16 19 2 2 36 33
103 New Zealand" 1,777 7,101 24 26 13 10 15 16 4 6 43 43
104 Australia" 9,051 31,547 16 18 9 7 24 21 7 8 43 45
105 United Kingdom 36,044 116,553 13 14 9 6 31 32 10 II 37 36
106 Italy" 30,942 175,443 10 8 13 14 24 32 13 10 40 36
107 tKuwait" 120 1,902 5 10 4 7 1 4 4 6 86 73
108 Belgium" 8,226 32,303 17 19 12 8 22 23 9 13 40 36
109 Netherlands"' 8,545 39,759 17 19 8 4 27 27 13 13 36 37
110 Austria" 4,873 30,879 17 17 12 8 19 25 6 6 45 43
Ill ±United Arab Emirates . . 2,155 . 14 . I . . . . . . 84
112 France" 38,861 191,692 12 13 10 8 26 30 8 9 44 41
113 Canada 17,001 . . 16 15 8 6 23 26 7 9 46 45
114 Denmark 2,929 17,230 20 22 8 5 24 23 8 10 40 40
115 Germany,Fed.Rep." 70,888 359,754 13 10 8 4 32 40 9 13 38 33
116 Finland 2,588 19,132 13 12 10 6 20 25 6 7 51 50
117 Sweden 8,477 33,282 10 10 6 2 30 34 5 9 49 45
118 UnitedStates" 253,863 868,233 12 12 8 5 31 35 10 10 39 38
119 Norway 2,416 12,337 15 21 7 2 23 25 7 8 49 44
120 Japan" 73,339 689,295 8 10 8 5 34 37 II 10 40 38
121 Switzerlandb . 10 . 7 . . 31 . 9 . . 42
Total reporting economies 722,228 t 3,551,267
Oil exporters 19,643 t 128,122
Nonreporting nonmembers
a. Includes unallocable data; see the technical notes. b. Value added in manufacturing data are at purchaser values. C. World Bank estimate.
189
Table 7. Manufacturing earnings and output
Earnings per employee Total earnings as a Gross output per employee
Growth rate Index (1980=100) percentage of value added (1980=100)
1970-80 1980-87 1985 1986 1987 1970 1985 1986 /987 1970 1985 1986 1987
Low-income economies
China and India
Other low-income
I Mozambique . . .
2 Ethiopia -4.7 -0.1 85 96 105 24 20 19 20 61 114 112 119
3 Chad . . . .
4 Tanzania . . -12.7 51 42 . . 42 35 34 . . 122 99 76
5 Bangladesh -3.0 -2.9 90 81 78 26 30 29 29 116 113 106 111
6 Malawi . 1.6 115 37 39 . . 120 139
7 Somalia -5.1 . 28 . . 27
8 Zaire ..
9 Bhutan .
10 Lao PDR ..
11 Nepal
12 Madagascar -o. . t; 91 55
13 BurkinaFaso
14 Mali '39
IS Burundi -78
16 Uganda
17 Nigeria 18 105
18 Zambia 34 109
Niger
19
20 Rwanda i jo jo
21 China . . 4.2 114 124 . . 13 15 . . . . 131 131
22 India 0.5 4.9 120 127 138 47 46 48 48 83 149 157 174
23 Pakistan 3.3 6.2 132 138 . 21 20 20 . . 51 145 149
24 Kenya -3.4 -2.3 79 83 87 53 48 48 48 38 107 113 108
25 Togo . .
26 Central African Rep. . . . . . 49 . . 74 85
27 Haiti -3.3 3.4 104 116 153 .
28 Benin . .
29 Ghana 23 . . 193
30 Lesotho . - 28 .
31 SriLanka . . 0.8 102 102 . . 70 135 132
32 Guinea . .
33 Yemen, PDR . . . .
34 Indonesia 5.0 6.0 139 144 26 19 19 42 141 156
35 Mauritania
36 Sudan 3'
37 Afghanistan
38 Myanmcrr
39 Kampuchea, Dem.
40 Liberia 107
41 Sierra Leone
42 VietNa,n
Middle-income economies
Lower-middle-income
43 Bolivia 0.0 -10.3 63 41 50 43 26 24 28 65 37 35 34
44 Philippines -3.7 2.6 109 120 21 22 21 20 102 lOS 112 -
45 Yemen Arab Rep. . . . 166 169
46 Senegal -4.9 -0.1 44 44 ..
47 Zimbabwe 1.6 -0.4 105 104 103 43 42 40 40 98 92 89 90
48 Egypt, Arab Rep. 4.1 54 76
49 Dominican Rep. -1.0 79 35 63
50 Cole d'Ivoire -0.9 27 52
51 Papua New Guinea 3.0 -2.9 42
52 Morocco 73 98 92
53 Honduras . . . .
54 Guatemala -3.2 -1.4 99 93 . . . 23 27 .
55 Congo, People's Rep. . . -0.1 34 49 . .
56 El Salvador 2.4 -9.3 63 . . 28 20 71 87
57 Thailand 1.0 7.0 143 146 25 24 23 24 68 138 138
58 Botswana 2.6 -4.5 85 .. 46 . . . . 60
59 Camemon . . 29 . .
60 Jamaica -0.2 . . 43
61 Ecuador 3.3 -1.5 94 99 95 27 40 36 35 83 104 109 114
62 Colombia -0.2 3.8 116 116 134 25 18 16 18 86 128 123 146
63 Paraguay . .
64 Tunisia 4.2 . 44 95
65 Turkey 6.1 -3.5 84 81 86 26 21 16 17 108 139 154 170
66 Pens -5.0 75 87 . . 13 18 . 82 64 63
67 Jordan -1.1 101 100 101 37 31 31 22 . - 155 144 161
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified
190
Earnings per employee
Total earnings as a Gross output per employee
Growth rate Index (1980=1011) percentage of value added (1980=100)
1970-80 1980-87 1985 1986 1987 1970 1985 1986 1987 1970 1985 1986 1987
68 Chile 8.2 -3.5 97 98 84 19 14 17 15 60
69 Syrian Arab Rep. 2.8 -1.4 33 33 23 72
70 Costa Rica . . 41
71 Mexico 1.2 -3.5 91 87 . 44 26 26 . . 77 112 107
72 Mauritius 1.8 -1.8 84 86 94 34 46 44 43 139 80 72 74
73 Poland . . . .
74 Malaysia 2.0 5.2 135 134 138 29 30 30 29 96
75 Panama 0.2 3.3 125 128 126 32 34 34 34 67 87 96
76 Brazil 4.0 . 22 . . 68 108 108 104
77 Angola . . . . .
78 Lebanon . . . .
79 Nicaragua . . -14.5 69 31 . . 16 22 22 . . 206 113 109
Upper-middle-income
80 South Africa 2.7 0.0 106 102 101 46 47 46 46 .
81 Algeria -1.0 . . 45 120
82 Hungaly 3.7 1.5 108 111 III 28 34 34 33 41 111 111 113
83 Uruguay . . 0.3 96 109 116 . . 22 25 26 . . 108 113 120
84 Argentina -1.5 2.5 105 111 103 30 21 21 19 71 87 103 103
85 Yugoslavia 1.3 -1.4 91 97 93 39 29 33 30 59 100 98 89
86 Gabon . . .
87 Venezuela 3.8 -0.6 110 106 107 31 26 27 31 118 109 106 110
88 Trinidad and Tobago . . 2.3 . . 79 67 . .
89 Korea,Rep.of 10.0 5.6 I25 128 145 25 27 26 27 40 140 146 165
90 Portugal 2.5 -1.9 89 95 34 39 39 . . 123
91 Greece 4.9 . 32 56
92 Oman . . .
93 Libya . . 37 45
94 Iran, Islamic Rep. . 25 85
95 Iraq . . 36 25 25
96 Romania . . . .
Low- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N.Africa
Latin America & Caribbean
Severely indebted
High-income economies
OECD members
tOther
97 tSaudi Arabia
98 Spain 4.4 0.9 100 lOt 112 52 40 38 40 127 112 139
99 Ireland 4.1 1.6 102 105 110 49 33 33 33 44 205 187
100 ttsrael 8.8 -6.4 102 65 70 36 59 .
101 (Singapore 3.6 6.2 153 148 146 36 38 32 29 73 114 Ii 121
102 tHong Kong 6.4 4.9 113 124 143 . . 63 60 61
103 New Zealand 1.2 -1.6 95 62 59 . . 121 .
104 Australia 2.9 1.9 106 113 115 53 48 52 52 . . 111 113 117
105 UnitedKingdom 1.7 2.9 112 115 123 52 43 43 43 61 137 129
106 Italy 4.1 0.7 101 100 108 41 42 42 43 51 128 126 135
107 tKuwait . 3.7 119 12 43 . . 153 .
108 Belgium 4.6 0.3 95 104 104 46 46 47 47 51 122 127 131
109 Netherlands 2.5 0.9 100 108 . 52 50 52 . . 68 116 127
110 Austria 3.4 1.5 106 110 113 47 56 56 56 64 116 111 118
Ill fUnited Arab Emirates . . . 61
112 France . . 1.2 107 . . 66 . . 112 105 107
113 Canada 1.8 -0.1 102 101 96 53 46 45 46 68 120 . .
114 Denmark 2.5 0.3 97 100 103 56 52 53 53 64 108 101 95
115 Gcrmany,Fed.Rep. 3.5 1.4 102 107 110 46 42 43 43 60 117 107 104
116 Finland 2.6 2.4 110 115 118 47 43 50 46 73 122 123 128
117 Sweden 0.4 0.4 98 100 102 52 37 37 35 75 123 120 125
118 UnitedStates 0.1 1.4 106 108 109 47 40 39 39 63 116 116 124
119 Norway 2.6 1.6 105 107 109 50 57 59 58 75 121 118 120
120 Japan 3.1 1.8 110 112 113 32 35 37 35 45 124 116 122
121 Switzerland . . . . . . . . . . . .
Total reporting economies
Oil exporters
Nonreporting nonmembers
191
Table 8. Growth of consumption and investment
Average annual growth rate (percent)
General government Gross
consumption Private consumption, etc. domestic investment
1965-80 1980-88 1965-80 1980-88 1965-80 1980-88
Low-income economies 6.3 w 7.0 w 4.1 w 5.2 w 8.8w 8.0w
China and India 5.3 w 9.2 w 4.0w 6.8 w 8.3 w 11.7w
Other low-income 8.4 w 2.4w 4.4 w 2.8 w 9.7 w -2.3 w
IMozambique -4.6 -1.0 . . -6.6
2 Ethiopia 6.4 5.6 1.3 -0.1 2.0
3 Chad
4 Tanzania a 8.4 1.8 6.2 0.3
5 Bangladesh 2.7 3.8 0.0 2.5
6 Malawi 5.6 3.8 4.0 3.1 9.0 -8.3
7 Somalia 11.7 -0.3 4.8 0.5 12.4 0.6
8 Zaire 0.7 16.3 1.2 -1.8 6.7 -1.6
9 Bhutan
10 LaoPDR
11 Nepal . . . . . . . . .
12 Madagascar 2.0 -0.8 1.1 -0.4 1.5 -0.7
13 Burkina Faso 8.7 10.0 2.1 2.9 8.8 4.3
14 Mali 1.9 3.1 4.9 3.4 1.8 2.8
15 Burundi 7.3 5.4 6.1 2.3 9.0 8.8
16 Uganda a
1.4 . . -5.7
17 Nigeria 13.9 -1.5 5.0 -0.1 14.7 -14.5
18 Zambia 5.1 -5.4 -0.9 4.5 -3.6 -4.5
19 Niger 2.9 1.5 -2.4 1.7 6.3 -10.2
20 Rwanda 6.2 3.0 4.2 1.6 9.0 7.4
21 China 5.6 9.4 5.2 7.4 10.7 14.4
22 India 4.7 8.8 2.7 5.8 4.5 4.3
23 Pakistan 4.7 11.0 4.8 4.3 2.4 6.5
24 Kenya 10.6 1.1 5.7 5.1 7.2 -1.1
25 Togo 9.5 3.9 5.0 -1.8 9.0 -1.6
26 CentralAfricanRep. -1.1 -2.4 4.2 2.0 -5.4 8.8
27 Haiti 1.9 -1.4 2.3 0.4 14.8 -5.1
28 Benin 0.7 1.4 2.3 1.9 10.4 -11.5
29 Ghana 3.8 -0.3 1.4 2.3 -1.3 4.9
30 Lesotho 12.3 8.8 . . 17.3
31 SriLanka 1.1 10.0 4.0 4.6 11.5 -0.9
32 Guinea
33 Yemen, PDR . . .. .. . .
34 Indonesia 11.4 2.9 5.9 7.2 16.1 1.9
35 Mauritania 10.0 -4.3 1.9 4.3 19.2 -5.3
36 Sudan 0.2 -5.3 4.3 3.7 6.4 -5.1
37 Afghanistan . . .
38 Myanmar
39 Kampuchea, Dem. . . . . .. . . .
40 Liberia 3.4 1.3 3.2 0.8 6.4 -16.7
41 SierraLeone a -33 4.2 -3.0 -1.0 -5.1
42 VietNam .
Middle-income economies 7.7w 2.5w 6.6w 2.5w 8.6w -0.6 w
Lower-middle-income 7.4w 2.7w 6.2w 2.2w 8.5w -1.6w
43 Bolivia 8.0 -5.2 4.0 -1.1 4.4 -19.5
44 Philippines 7.7 0.8 5.0 2.2 8.5 -10.9
45 YemenArabRep. . . 4.3 . . 3.6 . . -9.1
46 Senegal 3.1 -0.9 1.7 2.7 3.9 2.0
47 Zimbabwe 10.6 6.2 5.1 -2.4 0.9 -1.4
48 Egypt, Arab Rep. a
44 5.5 4.1 11.3 1.5
49 Dominican Rep. 0.2 2.8 7.9 0.5 13.5 4.6
50 Côte d'Ivoire 13.2 -2.6 7.5 1.9 10.7 -11.4
51 Papua New Guinea 0.1 -0.6 3.9 1.9 1.4 -1.5
52 Morucco 10.9 4.7 4.7 2.7 11.4 3.7
53 Honduras 6.9 4.9 4.9 1.9 6.8 -0.6
54 Guatemala 6.2 2.0 5.1 0.1 7.4 -3.4
55 Congo, People's Rep. 5.5 11.3 1.3 7.1 4.5 0.6
56 El Salvador 7.0 3.3 4.1 -0.1 6.6 0.1
57 Thailand 9.5 5.2 6.2 5.1 8.0 5.0
58 Botswana 12.0 13.8 9.2 2.4 21.0 -1.5
59 Camemon 5.0 7.8 4.2 4.2 9.9 0.4
60 Jamaica 9.8 -1.5 2.0 2.4 -3.3 -1.2
61 Ecuador 12.2 -2.2 6.8 1.8 9.5 -3.7
62 Colombia 6.7 3.2 5.9 2.5 5.8 0.3
63 Paraguay 5.1 3.7 6.5 1.1 13.5 -2.1
64 Tunisia 7.2 3.9 8.3 3.3 4.6 -6.1
65 Turkey 6.1 2.8 5.7 5.7 8.8 4.4
66 Peru 6.3 0.3 4.9 2.0 0.3 -3.9
67 Joidan 4.7 6.6 -5.5
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified
192
Average annual growth rate (percent)
General government Gross
consumption Private con.tumption, etc. domestic investment
1965-80 1980-8.8 1965-80 1980-88 1965-80 1980-88
68 Chile 4.0 -0.5 0.9 0.6 0.5 -0.5
69 Syrian Arab Rep. 15.1 -1.3 11.9 -0.7 13.9 -1.5
70 Costa Rica 6.8 0.4 5.2 2.7 9.4 5.8
71 Mexico 8.5 2.7 5.8 0.3 8.5 -6.9
72 Mauritius 7.1 2.4 5.9 3.6 8.3 14.0
73 Poland . . 2.8 1.7 . . 1.6
74 Malaysia 8.5 2.2 6.0 0.9 10.4 0.0
75 Panama 7.4 3.5 4.6 4.3 5.9 -3.2
76 Brazil 6.7 2.9 8.9 2.5 11.3 0.0
77 Angola . . .
78 Lebanon
79 Nicaragua 6.6 16.0 2.0 -8.1 . . 4.0
Upper-middle-income 8.0 w 7.1w 2.9w 8.7w -0.5 w
80 South Africa 5.3 3.8 3.3 2.2 4.1 -5.8
81 Algeria 8.6 5.1 8.8 1.4 15.9 -0.8
82 Hungary a
1.3 5.7 1.3 7.0 -1.0
83 Umguay 3.2 1.1 2.4 -0.7 8.0 -9.2
84 Argentina 3.2 0.0 3.0 0.4 4.6 -7.7
85 Yugoslavia 3.6 0.7 7.9 0.1 6.5 -0.4
86 Gabon 10.7 3.3 . . 6.2 14.1 -4.9
87 Venezuela . . 0.4 . . 0.3 . . -4.7
88 TrinidadandTobago 8.9 -3.1 6.3 -11.3 12.1 -19.1
89 Korea, Rep. of 7.7 5.3 7.8 7.5 15.9 10.5
90 Portugal 8.1 2.4 6.7 2.1 4.6 -1.3
91 Greece 6.6 2.6 4.9 3.2 5.3 -3.9
92 Oman . . 13.6 .. 18.4
93 Libya 19.7 . . 19.1 7.3
94 Iran,IslamicRep. 14.6 . . 10.1 11.5
95 Iraq
96 Romania
Low- and middle-income 7.3 w 4.0w 5.6w 3.5 w 8.6 w 2.7 w
Sub-Saharan Africa 8.2 w 1.4 w 3.5 w 0.9w 9.1 w -7.3 w
East Asia 6.7w 7.1 w 5.8w 6.6w 11.4w 9.7w
South Asia 4.6 w 8.9w 3.0 w 5.6w 4.2 w 4.2 w
Europe, M.East, & N.Africa 9.4 w .. .. .. 8.9w
Latin America & Caribbean 6.5 w 2.0 w 6.4w 1.2w 8.3 w -3.2 w
Severely indebted 6.7w 2.0w 6.3w 1.4w 8.4w -3.1 w
High-income economies 2.7w 2.6w 3.9w 3.0w 3.4w 3.7 w
OECD members 2.7w 2.6w 3.8w 3.0w 3.3 w 3.7w
tOther 14.4w
a
97 tSaudi Arabia 20.0 . . 27.5
98 Spain 5.1 4.7 4.8 1.8 3.7 3.7
99 Ireland 6.1 0.1 4.3 -1.4 6.3 -3.3
100 ttsrael 8.8 1.0 6.0 4.8 5.9 1.4
101 tSingapore 10.2 7.9 8.0 4.7 13.3 2.6
102 1-Hong Kong 7.7 5.4 9.0 7.1 8.6 3.0
103 NewZealand 3.4 1.1 2.3 1.6 2.2 3.9
104 Australia 5.0 3.7 4.1 3.0 2.8 2.7
105 UnitedKingdom 2.3 1.1 2.2 3.8 0.6 6.4
106 Italy 3.4 3.0 4.1 2.5 3.4 1.5
107 1-Kuwait
a
39 9.3 0.8 11.9 -2.3
108 Belgium 4.6 0.6 4.3 1.3 2.9 0.5
109 Netherlands 2.9 1.0 4.8 1.2 1.8 2.3
110 Austria 3.7 1.4 4.4 2.1 4.5 1.7
Ill 1-United Arab Emirates
112 France 3.6 2.3 4.7 2.3 3.9 0.9
113 Canada 4.8 1.9 4.9 3.0 5.1 4.6
114 Denmark 4.8 1.3 2.3 2.2 1.2 4.5
115 Germany, Fed. Rep. 3.5 1.5 4.0 1.6 1.7 1.2
116 Finland 5.3 3.7 3.8 4.3 2.9 2.0
117 Sweden 4.0 1.6 2.5 1.8 0.9 2.2
118 United States 1.2 3.5 3.1 3.6 2.6 5.0
119 Norway 5.5 3.4 3.9 3.1 4.2 3.2
120 Japan 5.1 2.7 6.0 3.2 6.7 4.9
121 Switzerland 2.7 2.7 2.5 1.6 0.8 4.8
Total reporting economies 3.3w 2.9w 4.2w 3.1w 4.4w 3.4w
Oil exporters 11.2w 7.1 w 2.1 w 11.5 w -2.5w
Nonreporting nonmembers
a. General government consumption figures are not available separately; they are included in private consumption, etc.
193
Table 9. Structure of demand
Distribution of gross domestic product (percent)
General Etports of goods
government Private Gross domestic Gross domestic and nonfactor Resource
consumption consumption etc. investment savings services balance
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
Low-income economies 11 w lOw 70w 65w 19w 28w 18w 26w 7w 13 w -1 w -3 w
China and India 12 w 9w 68 w 61 w 21 w 32 w 20 w 30 w 4w 11 w 0w -2 w
Otherlow-income 9w 12w 78w 74w 14w 18w 12w 14w 17w 19w 3w 4w
I Mozambique . . 22 . . 93 . 33 . . -15 . . 15 . -47
II -1 -11
2 Ethiopia
3 Chad
11
20
24
22
77
74
72
89
13
12
16
12
12
6 12
5
4 12
19 23 6 24
-26
4 Tanzania
5 Bangladesh
10
9
12
9
74
83
93
88
15
11
21
12
16
8 3
26
10
16
8 4 9 1
6 Malawi 16 14 84 78 14 16 0 8 19 23 14 8
3 30
7
8
Somalia
Zaire
8
9
10
24
84
61
86
68
11
14
34
11 30
8 3
8
17
36
8
37 152
9 Bhutan . . . . . . . . . .
10 Lao PDR 12 66 31 21 18 -9
6
. . . .
II Nepal 20 10
12 Madagascar
BurkinaFaso
23
9
10
12
26
100
74
87
80
80
78
10
12
6
16
25
0
4
4
10
4 8 16
9
8 13
21
15
6 9
8 29
13
14 Mali 10 10 84 93 18
6
15 5 4 12 15 13 18
2 13
15 Bumndi 7 17 89 78 18 4 5 10 12
16 Uganda
17 Nigeria
10
5 12
8 78
83
45
87
73
69
11
14
25
13
13
II
12
12
40
15
14
5 26
13
49
11
25
28
2 8
15
1
2
3
18 Zambia
19 Niger
15
6
17
11 90 85 8 10 3 4 9 17
5 10
6
20 Rwanda 14 12 81 82 10 16 5 6 12 8
21 China 14 7 61 56 24 38 25 37 4 14 1 -1
22 India 9 12 76 67 17 24 15 21 4 7 -2 -3
23 Pakistan II 14 76 73 21 18 13 13 8 14 -8 -5
24 Kenya 15 19 70 59 14 26 15 22 31 19 1 -3
25 Togo 8 17 76 69 22 21 17 14 20 34 -6 -7
26 CentralAfricanRep. 22 67 90 21 12 II -1 27 17 -II -12
27 Haiti
II
8
10
11 90 86 7 10 2 4
0
13 13
18
5
-8
6
-12
28 Benin 12 87 89 11 12 3 13
29 Ghana 14 9 77 84 18 12 8 6 17 19 -10 -6
30 Lesotho 18 28 109 145 II 47 26 73 16 23 38 120
31 SriLanka 13 10 74 78 12 23 13 13 38 26 1 -10
32 Guinea 10 . . 71 . . 22 19 26 . .
33 Yemen, PDR .. . . . . .. . . .. .. . . . . .
34 Indonesia 5 9 87 65 8 22 8 25 5 25 0 4
35 Mauritania 19 14 54 76 14 18 27 10 42 51 13 -8
36 Sudan
37 Afghanistan
12
a
9 79
99
85
. .
10
11
10 9
1
7 15
II
4
-10
3
38 Myanmar . . . . 0 . . . . . .
39 Kampuchea,Dem. 16 . 71 . 13 . . 12 12
40 Liberia 12 . 61 17 . . 27 . . 50 10
3
.
41 SierraLeone 8 6 83 77 12 11 8 17 30 31 6
42 VietNam . . . . . . .
Middle-income economies 11w 15w 67w 59w 21w 25w 21w 27w 17w 27w Ow 3w
Lower-middle-income lOw 12w 71w 64w 19w 23w 18w 24w 14w 21w -1w 2w
43 Bolivia
44 Philippines
9
9
20
9
74
70
73
73
22
21
11
17
17
21 18
6 21
17
16
25
5 0
-4
45 YemenArabRep. 0 -13
46 Senegal 17
. . 20
16 75
. . 80
74
.
12
. 13
15
.
8
.
9 24
. . 16
26
27
.
-4
.
5
47 Zimbabwe 12 19 65 57 15 21 23 24 8 3
48 Egypt,ArabRep.
49 DominicanRep.
19
19
14
6
67
75
78
78
18
10
20
24
14
6 16
8 18
16
22
34
4
-4
12
-7
50 Côted'Ivoire 11 19 61 59 22 14 29 22 37 33 7 8
51 PapuaNewGuinea 34 21 64 58 22 26 2 21 18 45 20 5
52 Morocco 12 15 76 62 10 24 12 23 18 25 1 0
53 Honduras
54 Guatemala
10
7 8
17 75
82
72
84
15
13
16
14
15
10
11
8
27
17
23
16 -3
0 5
-6
55 Congo, People's Rep. 14 22 80 58 22 22 5 20 36 42 -17 -2
56 ElSalvador 9 13 79 81 15 13 12 6 27 16 -2 -6
57 Thailand 10 11 72 63 20 28 19 26 16 34 -1 -2
58 Botswana 24 . . 89 . . 6 . . -13 . . 32 . . 19
59 Camemon 13 10 75 76 13 17 12 14 24 16 -1 -3
60 Jamaica 8 15 69 66 27 27 23 19 33 49 -4 -8
61 Ecuador 9 11 80 68 14 23 11 21 16 27 -3 -2
62 Colombia 8 11 75 67 16 21 17 22 11 16 1 2
63 Paraguay 7 6 79 72 15 24 14 23 15 26 -1 -2
64 Tunisia 15 16 71 65 28 19 14 19 19 43 -13 0
65 Turkey 12 9 74 65 15 24 13 26 6 24 -1 2
66 Peni 10 8 59 68 34 29 31 24 16 10 -3 -4
67 Jonlan 27 76 26 -3 54 -29
Note. For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
194
Distribution of gross domestic product (percent)
General Erports of goods
government Private Gross domestic Gross domestic and nonfactor Resource
consumption consumption, etc. investment savings services balance
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
68 Chile 11 II 73 65 15 17 16 24 14 37 7
69 SyrianArabRep.
70 CostaRica
14
13
16
15
76
78
72
59
10
20
17
26
10 13
26
17
23 36
11
10
1
0 4 0
9
71 Mexico
72 Mauritius 13
6 10
II
75
74
67
64
20
17
20
25
19
13
23
25 36
8
72
16 2
4 2
0
73 Poland . . 8 . . 56 . - 33 35 . . 23 . . 3
74 Malaysia 49 20 26 36 42 67
75 Panama
76 Brazil
15
11
11
14
.
12
61
73
67
.
60
18
20
.
23
24
16
22
.
28
. 36
8
. .
10
2 4
2
10
5
77 Angola . 0 . . . .
78 Lebanon 10 81 22 . . 9 . . 36 . . 13
79 Nicaragua 8 74 21 . . 18 . . 29 . . -3
Upper-middle-income 12 w 62w 24w .. 25w 20w 1w
80 SouthAfrica 62 56 28 20 27 25 26 29 0 5
81 Algeria
11
15
18
16 66
75
53 22
26
31 19 31
28
22 16 3
-1
82 Hungary 11 61 25 25 . . 38 3
83 Uruguay IS 13 68 73 11 10 18 14 19 22 7 4
84 Argentina 8 11 69 71 19 14 22 18 8 10 3 4
85 Yugoslavia 18 14 52 47 30 39 30 40 22 24 0 1
86 Gabon II 22 52 45 31 27 37 33 43 44 6 6
87 Venezuela 10 10 56 66 25 30 34 25 26 22 9 -6
88 TrinidadandTobago 12 22 67 57 26 18 21 21 65 39 -5 3
89 Korea,Rep.of 9 10 83 51 15 30 8 38 9 41 -7 9
Portugal 66 30 27 35 -5 -10
90
91
92 Oman
Greece
12
12
.
14
21
68
73
. .
68
25
26
. .
18
. .
20
15
. .
21
II 9
.
23 11 7
93 Libya 14 36 29 50 53 . . 21 - -
94 Iran, IslamicRep. 13 63 - 17 . . 24 20 . . 6 -
95 Iraq
96 Romania
20
. 0
. 50 .
. .
16
. . .
.
.
31
. .
.
.
38 . . 15
Low- and middle-income 11w 13w 68 w 61 w 20 w 26 w 20 w 27 w 13w 23 w -1w 2w
Sub-Saharan Africa lOw 15w 73 w 72 w 14w 15w 14w 12w 23 w 23 w 1w -4w
East Asia 13w 10 w 64w 56 w 22 w 31 w 23 w 34 w 8w 30 w Ow 3w
South Asia 8w 12 w 77 w 70 w 17w 22 w 14 w 18 w 6w 8w -3w -4w
Europe, M.East, & N.Africa 13w 65 w 22 w 20 w 19 w -2w
Latin America & Caribbean 9w 11w 69 w w 20 w w 21 w w 13 w iS w 1w 2w
Severely indebted 9w 11w 69w 64w 21w 22w 22w 25w 14w 16w 1w 3w
High-income economies 17 w 17 w 63 w 61 w 20 w 22 w 21 w 22 w 12 w 21 w 1w Ow
OECD members 17 w 17 w 63 w 61 w 20 w 22 w 20w 22 w 12 w 20w 1w Ow
tOther 14 w 22 w 50 w 53 w 24 w 25 w 34 w 25 w 54 w 63 w 10 w -2w
97 tSaudi Arabia
98 Spain
18 33 34 47 14
28
27 48 20
22
60
10
38
19 334 6
99 Ireland
100 lIsrael 20
15
8 15
16
31
68
68
65
63
55
59
26
29
23
18
17
24
17
15
27
10
35
19
64
32
9
13
10
6
101 tSingapore 10 II 80 48 22 37 10 41 123 . . 12 4
102 iHong Kong
103 New Zealand
7 7 64
62
60
60
36
26
28
22
29
26
33
26
71
22
136
28
7 5
3
104 Australia
13
13
15
19 69 58
64
20 24 18 23 15 17 2
-1 -4
105 United Kingdom 16 19 63 21 21 20 17 18 23
106 Italy 14 15 63 62 23 23 24 23 13 18 I 0
107 tKuwait 13 25 26 60 16 20 60 15 68 41 45 -5
108 Belgium 13 16 64 63 23 18 23 21 36 68 0 3
109 Netherlands 15 18 70 59 16 19 15 23 43 55 -1 4
110 Austria 13 17 57 56 30 27 30 27 25 37 -1 0
Ill lUnited Arab Emirates . . 21 44 26 . . 36 . . 55 10
112 France 16 19 57 60 26 21 27 21 13 22 1 0
26 0
113
114
Canada
Denmark
14
16
19
25
60
72
59
54
26
13
22
18
26
12
23
21
26
29
19
32
33
2 0
3
115
116
Germany,Fed Rep.
Finland
15
14
19
19
67
58
55
54 30
18 21
27
18
29 27 20
19
25 2 6
0
117 Sweden 26 72 II 21 22 32
118 UnitedStates
18
19 20
20
63
53
67
52 30
17
19
15
28
10
18
29
13
28 41
6 Il
36 -1
1 20
119 Norway 15 56
120 Japan 8 9 64 57 27 31 28 33 11 13 1 3
121 Switzerland II 12 60 58 30 31 30 31 29 36 0
Total reporting economies 15w 16w 64w 61w 20w 22w 20w 23w 12w 21w Ow 1w
Oil exporters 11w 16w 66w 59w 20w 25w 24w 24w 23w 25w 5w Ow
Nonreporting nonmembers
a. General government consumption figures are not available separately; they are included in private conaumption, etc.
195
Table 10. Structure of consumption
Percentage share of total household consumption a
Food Gross rents, frel Other consumption
and power Transport and Other
Cereals Clothing
communication consumer
and and Fuel and Medical
Total tubers footwear Total power care Education Total Motor cars Total durables
Low-income economies
China and India
Other low-income
I Mozambique S
2 Ethiopia 50 24 6 14 7 3 2 8 1 17 2
3 Chad .. .. .. ..
4 Tanzania 64 32 10 8 3 3 3 2 0 10 3
5 Bangladesh 59 36 8 17 7 2 1 3 0 10 3
6 Malawi 55 28 5 12 2 3 4 7 2 15 3
7 Somalia . . . . . . . . . . . . . . . . . . .
8 Zaire 55 15 10 11 3 3 1 6 0 14 3
9 Bhutan . . . .
10 La0PDR .. . . .. .
11 Nepal 57 38 12 14 6 3 1 1 0 13 2
12 Madagascar 59 26 6 12 6 2 4 4 I 14 1
13 BurkinaFaso . . . . . . . . . . . . .
14 Mali 57 22 6 8 6 2 4 10 1 13 1
15 Bunindi . . . . .
16 Uganda . . . . . . . . . . .
17 Nigeria 52 18 7 10 2 3 4 4 I 20
18 Zambia 37 8 10 11 5 7 13 5 1 16
19 Niger .. . .. .. . .
20 Rwanda 30 II 11 16 6 3 4 9 28
61b 13 3 15
21 China 8 1 1 1 . .
22 India 52 18 Il 10 3 3 4 7 0 13 3
23 Pakistan 54 17 9 15 6 3 3 1 0 15 5
24 Kenya 39 16 7 12 2 3 9 8 1 22 6
25 Togo . . .
26 Central African Rep. . . .
27 Haiti .. .. .. .. .. .. .. ..
28 Benin 37 12 14 12 2 5 4 14 2 15 5
5C
29 Ghana 50 . . 13 11 3 3 J5
30 Lesotho . . . . . . .
31 SriLanka 43 18 7 6 3 2 3 15 1 25 5
32 Guinea . . . . .
33 Yemen, PDR .. .. .. .. .. . . . . . .
34 Indonesia 48 2! 7 13 7 2 4 4 0 22 5
35 Mauritania . . . . . .
36 Sudan 60 5 15 4 5 3 2 11
37 Afghanistan . . .
38 Myanmar . . . . . . 0
39 Kampuchea, Dem. . . . . . . 0
40 Liberia . S S S . .
41 Sierra Leone 56 22 4 15 6 2 3 12 9
42 VietNam
Middle-income economieu
Lower-middle-income
43 Bolivia 33 . . 9 12 1 5 7 12 . . 22
44 Philippines 51 20 4 19 5 2 4 4 2 16 2
45 Yemen Arab Rep. . . . . . . . . S . S S
46 Senegal 50 15 11 12 4 2 5 6 0 14 2
47 Zimbabwe 40 9 11 13 5 4 7 6 1 20 3
48 Egypt,ArabRep. 50 10 11 9 3 3 6 4 1 18 3
49 Dominican Rep. 46 13 3 15 5 8 3 4 0 21 8
50 Côted'Ivoire 40 14 10 5 1 9 4 10 23 3
51 Papua New Guinea . . S S S S S S S
52 Morocco 40 12 11 9 2 4 6 8 1 22 5
53 Honduras 39 . . 9 21 . 8 5 3 . . 15
54 Guatemala 36 10 10 14 5 13 4 3 0 20 5
55 Congo,People'sRep. 42 19 6 11 4 3 1 17 1 20 4
56 El Salvador 33 12 9 7 2 8 5 10 1 28 7
57 Thailand 30 7 16 7 3 5 5 13 0 24 5
58 Botswana 35 13 8 15 5 4 9 8 2 22 7
59 Cameroon 24 8 7 17 3 II 9 12 1 21 3
31 22
60 Jamaica 39 4 15 7 . 17
61 Ecuador 30 . . 10 7e i 5 6C l2 30
62 Colombia 29 6 12 2 7 6 13 27
63 Paraguay 30 6 12 21 4 2 3 10 I 22 3
64 Tunisia 37 7 10 13 4 6 9 7 1 18 5
65 Turkey 40 8 15 13 7 4 1 5 . 22
66 Peru 35 8 7 15 3 4 6 10 0 24 7
67 Jordan 35 5 6 5 8 6 35
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
196
Percentage share of total household conoumption
Food Gross rents, fi4el Other consumption
and power Transport and
Cereals Clothing Other
Communication
and and Fuel and Medical consumer
Total tubers footwear Total power care Education Total Motor cars Total durables
68 Chile 29 7 8 13 2 5 6 Il 0 29 5
69 Syrian Arab Rep. ..
70 Costa Rica 33 8 8 9 1 7 8 8 0 28 9
71 Mexico 35i 10 8 5 5 12 25
72 Mauritius 24 7 5 19 3 5 7 11 1 29 4
73 Poland 29 9 7 2 6 7 8 2 34 9
74 Malaysia 30 . . 5 9 . ' 5 8 16 . . 27
75 Panama 38 7 3 11 3 8 9 7 0 24 6
76 Brazil 35 9 10 11 2 6 5 8 1 27 8
77 Angola . . . .
78 Lebanon
79 Nicaragua
Upper-middle-income
80 South Africa 26 . . 7 12 4' . 17 34
81 Algeria . .. .. .. .. .. .. ..
82 Hungaiy 25 . . 9 10 5 5 7 9 2 35 8
83 Uniguay 31 7 7 12 2 6 4 13 0 27 5
84 Argentina 35 4 6 9 2 4 6 13 0 26 6
85 Yugoslavia 27 . 10 9 4 6 5 11 2 32 9
86 Gabon . . . . . . . . . . . 0
.
7C
87 Venezuela 38 . . 4 8 . . 8 10 25
88 Trinidad and Tobago . . . . . . . . . . . . .
89 Korea, Rep. of 35 14 6 II 5 5 9 9 25 5
90 Portugal 34 . . 10 8 3 6 5 13 3 24 7
91 Greece 30 8 12 3 6 5 13 2 26 5
92 Oman . . , , . . . . . . .
93 Libya .. .. .. .. .. .. ..
94 Iran, Islamic Rep. 37 10 9 23 2 6 5 6 1 14 5
95 Iraq
96 Romania
Low- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N.Africa
Latin America & Caribbean
Severely indebted
High-income economies
OECD members
fOther
97 tSaudi Arabia . . . . . . . . . . . . . . .
98 Spain 24 3 7 16 3 7 5 13 3 28 6
99 Ireland 22 4 5 11 5 10 7 11 3 33 5
100 tlsrael 22 . . 4 20 2 10 14 10 . . 20
101 tSingapore 19 . . 8 11 . . 7 12 13 . . 30
102 tHongKong 12 1 9 15 2 6 5 9 1 44 15
103 New Zealand 12 2 6 14 2 9 6 19 6 34 9
104 Australia 13 2 5 21 2 10 8 13 4 31 7
105 United Kingdom 12 2 6 17 4 8 6 14 4 36 7
106 Italy 19 2 8 14 4 10 7 11 3 31 7
107 f Kuwait
108 Belgium
..
15
..
2
..
6 17
.. . 7
..
10
..
9
..
11 3 31 7
109 Netherlands 13 2 6 18 6 11 8 10 3 33 8
110 Austria 16 2 9 17 5 10 8 15 3 26 7
Ill f United Arab Emirates . . . . . . . .
112 France 16 2 6 17 5 13 7 13 3 29 7
113 Canada 11 2 6 21 4 5 12 14 5 32 8
114 Denmark 13 2 5 19 5 8 9 13 5 33 7
115 Germany, Fed. Rep. 12 2 7 18 5 13 6 13 4 31 9
116 Finland 16 3 4 15 4 9 8 14 4 34 6
117 Sweden 13 2 5 19 4 11 8 11 2 32 7
118 UnitedSlates 13 2 6 18 4 14 8 14 5 27 7
119 Norway 15 2 6 14 5 10 8 14 6 32 7
120 Japan 16 4 6 17 3 10 8 9 1 34 6
121 Switzerland 17 4 17 6 15 . . 9 . . 38
Total reporting economies
Oil exporters
Nonreporting nonmembers
a. Data refer to either 1980 or 1985. b. tncludes beverages and tobacco. c. Refers to government expenditure. d. Excludes government expenditure. e. Excludes
fuel. f. Includes fuel.
197
Table 11. Central government expenditure
Percentage of total expenditure
Housing, Total
amenities, expenditure as Overall
social security Economic a percentage surplus/defictt as a
Defense Education Health and welfare a services Other a of GNP percentage of GNP
1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988
Low-income economies
China and India
Other low-income :: 1o:w 20.5w 9.0w 5.5w 2:8w 6.2w :: 51:;w 24.1w -3.0w
I Mozambique
2 Ethiopia 14.3 . . 14.4 10.6 5.7 3.6 4.4 9.3 22.9 30.1 52.6 46J 13.7 35.2 -1.4 -6.8
3 Chad 24.6 . 14.8 . . 4.4 1.7 . . 21.8 . . 32.7 . . 14.9 . . -2.7
4 Tanzania 11.9 . . 17.3 . . 7.2 2.1 . . 39.0 . . 22.6 . . 19.7 . . -5.0
5 Bang1adesh' 5.1 14.8 . . 5.0 9.8 . . 39.3 . . 25.9 . . 9.4 -1.9
6 Ma1awit 3.1 5.6 15.8 10.0 5.5 5.9 5.8 2.0 33.1 27.0 36.7 49.4 22.1 32.0 -6.2 -8.6
7 Somaliab 23.3 . . 5.5 . . 7.2 . 1.9 . . 21.6 . . 40.5 . 13.5 0.6
8 Zaire 11.1 . . 15.2 . . 2.3 2.0 . . 13.3 . . 56.1 . . 19.8 -3.8
9 Bhutan .. .
10 LaoPDR
II Nepal 7.2 5.6 7.2 10.9 4.7 4.3 0.7 3.3 57.2 51.2 23.0 24.7 8.5 19.7 -1.2 -6.2
12 Madagascar 3.6 . . 9.1 . . 4.2 . . 9.9 . . 40.5 . 32.7 . . 20.8 . . -2.5
13 BurkinaFaso 11.5 17.9 20.6 14.0 8.2 5.2 6.6 . . 15.5 7.0 37.6 55.9 11.1 16.8 0.3 0.4
14 Mali . . 8.4 . . 9.8 . . 2.6 . . 3.3 . . 18.4 . . 57.5 . . 28.2 . . -5.5
15 Burundi 10.3 . . 23.4 . . 6.0 . . 2.7 . . 33.9 . . 23.8 . . /9.9 . . 0.0
16 Uganda 23.1 26.3 15.3 15.0 5.3 2.4 7.3 2.9 12.4 14.8 36.6 38.6 21.8 10.3 -8.1 -3.0
17 Nigeriab 40.2 2.8 4.5 2.8 3.6 0.8 0.8 1.5 19.6 35.9 31.4 56.2 8.3 27.8 -10.3
18 Zambiab 0.0 0.0 19.0 8.3 7.4 4.7 1.3 2.3 26.7 21.0 45.7 63.7 34.0 26.0 -13.8 -9.8
19 Niger .. .. .. .. .. .. .. .. .. ..
20 Rwanda 25.6 . . 22.2 . . 5.7 . . 2.6 . . 22.0 . . 21.9 . . 12.5 -2.7
21 China .. .. .. .. .. .. .. .. .. .. .. .. ..
22 India 26.2 19.3 2.3 2.9 1.5 1.8 3.2 5.4 19.9 21.7 46.9 49.0 10.5 17.8 -3.2 -7.9
23 Pakistan 39.9 29.5 1.2 2.6 1.1 0.9 3.2 8.7 21.4 34.5 33.2 23.8 16.9 21.7 -6.9 -7.0
24 Kenyab 6.0 9.2 21.9 21.5 7.9 6.1 3.9 3.5 30.1 19.8 30.2 39.9 21.0 28.6 -3.9 -6.6
25 Togo . . 11.1 . . 19.9 . . 5.2 . . 8.5 . . 31.2 . . 24.1 . . 32.5 -2.6
26 Central African Rep. . . . . . . 25.7
27 Haiti . . . . . . .
. 14.5
28 Benin .. .. .. .. .. .. .. .. .. .. .. ..
29 Ghana' 7.9 3.2 20.1 25.7 6.3 9.0 4.1 11.9 15.1 19.2 46.6 31.1 19.5 14.0 -5.8 0..'
30 Lesotho 0.0 . . 22.4 . . 7.4 . . 6.0 . . 21.6 . . 42.7 . . 14.5 . . 3.5
31 SriLanka 3.1 9.6 13.0 7.8 6.4 5.4 19.5 11.7 20.2 29.2 37.7 36.3 25.4 31.4 -5.3 -12.8
32 Guinea .
33 Yemen, PDR . . . . . . . . .. . . . . . . . . .. .. .. .. ..
34 Indonesia 18.6 8.3 7.4 10.0 1.4 1.8 0.9 1.7 30.5 . . 41.3 78.2 15.1 22.7 -2.5 -3.3
35 Mauritania . . . . . . . . . . . . . . . . . . . . .
36 Suda&' 24.1 . . 9.3 . . 5.4 1.4 . . 15.8 . . 44.1 . . 19.2 . . -0.8
37 Afghanistan . . . . . . . . . . . . . . . . . . . . . . .
38 Myanmar 31.6 12.9 15.0 13.4 6.1 4.9 7.5 13.2 20.1 38.6 19.7 16.9
39 Kampuchea, Dem. . . . . . . .. . . . . .. . . .. . . .. . . .
40 Liberia 5.3 . . 15.2 . . 9.8 3.5 . . 25.8 . . 40.5 .. 16.7 27.1 1.1
41 Sierra Leone 3.6 . . 15.5 . . 5.3 2.7 . . 24.6 . . 48.3 . . 23.9 -4.4
42 VietNam S
Middle-incomeeconomies 12.2w 12.9w 12.6w . . 6.1 w . . 16.7w 16.6w 25.7w . . 26.7w . . 18.5w 19.9w -2.8w -3.9w
Lower-middle-income 11.3 w 12.9w 17.5w 13.3w 5.7w 4.0w 16.2w 10.6w 23.1 w 16.1 w 26.2w 43.1 w 14.9w 15.4w -3.4w -3.7w
43 Bolivia 18.8 /4.5 31.3 18.4 6.3 1.9 0.0 25.6 12.5 17.1 31.3 22.5 9.6 15.8 -1.8 -0.1
44 Philippine&' 10.9 /1.6 16.3 15.7 3.2 4.6 4.3 2.2 17.6 31.9 47.7 34.1 13.4 15.6 -2.0 -2.8
45 YemenArabRep. 33.8 31.2 4.0 17.6 2.9 3.6 0.0 0.0 0.9 6.3 58.4 41.3 13.4 31.8 -2.2 -13.1
46 Senegal . . . . . . . . . . . . . . . . . . . . . . 18.8 . . -2.8
47 Zimbabwe . . 16.3 . . 22.0 7.5 . . 3.8 . . 23.1 . . 27.3 38.7 -9.1
48 Egypt, Arab Rep.
49 Dominican Rep. 8.5 14.2 .. 11.7 .. 11.8 .. 35.4 .. 18.3 17.7
50 Côte d'Ivoire
51 Papua New Guineab 45 159 96 17 212 471 317 -19
52 Morocco 12.3 15.1 19.2 17.0 4.8 3.0 8.4 7.3 25.6 21.4 29.7 36.0 22.8 29.2 -3.9 -4.6
53 Honduras 12.4 . . 22.3 . . 10.2 8.7 . . 28.3 . . 18.1 . . 16.1 . . -2.9
54 Guatemala . . . . . . 9.9 12.1 -2.2 -1.1
55 Congo, People's Rep. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
56 ElSalvador b 6.6 25.7 21.4 17.1 10.9 7.1 7.6 4.4 14.4 17.6 39.0 28.1 12.8 11.3 -1.0 -0.3
57 Thailand 20.2 18.7 19.9 19.3 3.7 6.2 7.0 5.4 25.6 19.5 23.5 30.9 16.7 16.4 -4.2 1.0
58 Botswanab 0.0 12.1 10.0 18.1 6.0 74 21.7 11.0 28.0 21.2 34.5 30.2 33.7 50.9 -23.8 21.9
59 Caineroon
60 Jamaica . . . . .. . . . . . . . . . . . . . . . . . . . . .
61 Ecuador' 15.7 . . 27.5 . . 4.5 0.8 .. 28.9 . . 22.6 . . 13.4 17.1 0.2 -2.4
62 Colombia . . . . . . . . . . . . . . . . . . . . 13.1 14.7 -2.5 -0.7
63 Paraguay 13.8 .. .. 18.3
12.1 .. .. 19.6 .. 32.7
3.5 .. 13.1 .. -1.7
64 Tunisia 4.9 5.7 5.9
30.5 8.814.622.0 23.3 24.4 25.17.4 27.3 23.1 37.1 -0.9 -4.8
65 Turkey 15.5 10.4 18.1 12.7 2.4 3.1 3.1 42.0 22.1 3.2 18.1 49.3 22.7 22.0 -2.2 -4.0
66 Perub 14.5 20.0 23.6 15.3 5.8 1.8 . 30.9 18.9 5.523.6 . 40.1 16.1 14.6 -0.9 -5.7
67 Jordan 33.5 26.5 9.4 13.0 3.8 5.4 10.5 9.5 26.6 15.7 16.2 30.0 52.3 49.9 -7.6 -15.7
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
198
Percentage of total expenditure
Housing, Total
amenities; expenditure ax Overall
social security Economic a percentage surplus/deficit as a
Defense Education Health and welfare a services Other a of GNP percentage of GNP
1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988
68 Chile 6.1 10.4 14.3 12.0 8.2 6.3 39.8 39.2 15.3 11.2 16.3 20.9 43.2 33.4 -13.0 -0.2
69 Syrian Arab Rep. 37.2 40.4 11.3 10.4 1.4 1.5 3.6 4.5 39.9 25.0 6.7 18.2 28.8 28.3 -3.5 -2.7
70 Costa Rica 2.8 2.2 28.3 16.2 3.8 19.3 26.7 26.7 21.8 12.3 16.7 23.3 18.9 28.0 -4.5 -4.7
71 Mexico 4.5 1.4 16.4 7.4 4.5 1.1 25.4 9.3 35.8 11.4 13.4 69.3 11.4 27.9 -2.9 -10.0
72 Mauritius 0.8 0.8 13.5 12.7 10.3 7.6 18.0 16.6 13.9 23.8 43.4 38.6 16.3 24.8 -1.2 0.3
73 Poland . . . . . . . . . . . . . . . . . . . . . . . . .
. 40.4 . . -2.4
74 Malaysia 18.5 . . 23.4 . . 6.8 . . 4.4 . . 14.2 . . 32.7 . . 26.5 31.3 -9.4 -8.0
75 Panama 0.0 5.9 20.7 15.6 15.1 16.7 10.8 16.0 24.2 6.1 29.1 39.7 27.6 34.4 -6.5 -4.4
76 Brazil 8.3 4.0 8.3 4.8 6.7 9.5 35.0 24.2 23.3 12.1 18.3 45.4 17.4 25.1 -0.3 -12.2
77 Angola . . . . . . .
78 Lebanon . . . . . . . . . . . . . . . . . . . . . . .
79 Nicaragua 12.3 . . 16.6 4.0 . . 16.4 . . 27.2 . 23.4 . . 15.5 58.0 -3.9 -16.3
Upper-middle-income 13.6w . 7.5 w . . . . 17.1 w .. .. . 24.3 w -1.9w
80 SouthAfrica . . . . . . . . . . . . 21.8 33.1 -4.2 -5.7
81 Algeria .. .. .. .. .. .. .. .. .. .. .. ..
82 Hungaiy . . 4.8 . . 2.1 . . 1.7 . . 28.7 . . 34.8 . . 27.8 . . 58.3 . . -0.2
83 Uruguay 5.6 10.2 9.5 7.1 1.6 4.8 52.3 49.5 9.8 8.3 21.2 20.1 25.0 23.7 -2.5 -0.7
84 Argentina 10.0 6.9 20.0 6.9 . . 2.1 20.0 32.0 30.0 17.7 20.0 34.4 19.6 21.6 -4.9 -4.1
85 Yugoslavia 20.5 55.1 0.0 0.0 24.8 0.0 35.6 11.2 12.0 16.3 7.0 17.3 21.1 7.5 -0.4 0.0
86 Gabon5 . . . . . . . . . . . . . . . . . . . 40.1 . . -12.9
87 Venezuela 10.3 5.8 18.6 19.6 11.7 10.0 9.2 11.7 25.4 17.3 24.8 35.6 18.1 21.8 -0.2 -2.1
88 Trinidad and Tobago . . . . . . . . . . . . . . . . . . . . . . . . . .
89 Korea, Rep. of 25.8 27.1 15.8 19.0 1.2 2.2 5.9 8.5 25.6 17.1 25.7 26.0 18.0 15.7 -3.9 1.6
90 Portugal . . 5.4 . . 9.5 . . 7.8 . . 25.7 . . 9.4 . . 42.3 . . 45.3 . . -11.0
91 Greece 14.9 . . 9.1 . 7.4 . . 30.6 . . 26.4 . . 11.7 . . 27.5 . . -1.7
92 Oman 39.3 38.2 3.7 10.7 5.9 4.8 3.0 8.3 24.4 12.9 23.6 25.1 62.1 49.3 -15.3 -12.6
93 Libya .. .. .. .. .. .. .. .. .. .. .. ..
94 Iran, Islamic Rep. 24.1 14.2 10.4 19.6 3.6 6.0 6.1 17.4 30.6 15.7 25.2 27.1 30.8 -4.6
95 Iraq . . . . . . . . . . . . . . .
96 Romasnia 5.4 2.9 0.5 . . 16.2 . . 61.8 . . 13.1
Low- and middle-income 14.1 w 13.6w 12.9 w 5.7w 14.4w . .. 39.7w 15.9w 20.9w -4.7w
Sub-Saharan Africa
East Asia ..
South Asia 20.3w 3.7w 2.1 w 7.2w .. 24.3w 42.4w .. 18.8w .. -8.0w
Europe,M.East,&N.Africa .. .. .. -4.2w
Latin America & Caribbean 6.7w 5.8w 15.4w 11.0w 6.5w .. 20.3 w 16.6w 17.0w 13.5w 34.1 w 53.1 w 11.9w 13.0w -1.9w -3.2w
Severely indebted 7.3 w 6.4w 15.6w 10.8w 5.9w 4.4w 18.9 w 17.2w 16.9w 19.2 w 35.4 w 42.0w 12.2w 14.4w -2.2 w -3.8w
High-income economies 21.8w 13.4w . . 4.8w 11.1 w 12.4w 41.9w 36.5w 13.0w 9.0w 12.2w 23.9w 22.7w 28.9w -1.9w -3.3w
OECD members 21.7w 13.3w . . 4.7w 11.2w 12.6w 42.3 w 36.9w 13.0w 8.9w 11.8w 23.6w 22.3w 28.6w -1.8w -3.4w
tOther
97 tSaudi Arabia . . . . . . . . . . . . . . . . . . . . . . . . . . .
98 Spain 6.5 . . 8.3 . . 0.9 . . 49.8 . . 17.5 . . 17.0 . . 19.6 34.1 -0.5 -4.5
99 Ireland . . 2.8 . . 11.8 . . 12.4 . . 30.3 . . 15.4 . . 27.3 32.7 58.1 -5.5 -10.7
100 lIsrael 42.9 27.2 7.1 9.6 3.6 3.7 7.1 21.2 7.1 11.7 32.2 26.6 43.9 50.6 -15.7 -9.9
101 tSingapore 35.3 14.6 15.7 14.4 7.8 3.6 3.9 11.0 9.9 15.6 27.3 40.8 16.7 35.0 1.3 -2.7
102 tHong Kong . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
103 NewZealand" 5.8 4.7 16.9 11.1 14.8 12.4 25.6 29.7 16.5 9.2 20.4 32.9 31.1 49.1 -4.2 0.7
104 Australia 14.2 9.2 4.2 7.0 7.0 9.6 20.3 28.6 14.4 7.0 39.9 38.6 20.2 28.7 0.3 -1.3
105 UnitedKingdom 16.7 12.6 2.6 2.2 12.2 13.6 26.5 30.9 11.1 6.9 30.8 33.8 31.8 37.6 -2.7 -0.8
106 Italy 6.3 3.3 16.1 7.6 13.5 10.4 44.8 35.4 /8.4 10.5 0.9 32.7 29.5 51.3 -8.7 -14.2
107 tKuwait 8.4 13.9 15.0 14.2 5.5 7.7 14.2 20.1 16.6 18.1 40.1 26.1 34.4 35.7 17.4 23.5
108 Belgium 6.7 4.9 15.5 12.2 1.5 1.8 41.0 43.3 18.9 9.1 16.4 28.7 39.3 52.4 -4.3 -8.3
109 Netherlands 6.8 5.1 15.2 11.9 12.1 10.9 38.1 39.6 9.1 9.4 18.7 23.2 41.0 55.7 0.0 -4.3
110 Austria 3.3 2.6 10.2 9.3 10.1 12.8 53.8 47.5 11.2 11.3 11.4 16.4 29.6 40.1 -0.2 -5.1
Ill tUnitedArabEmirates5 24.4 . . 16.5 . . 4.3 . . 6.1 . . 18.3 . . 30.5 . . 4.0 0.3
112 France . . . . . . . . . . . . . . . . . . . . . . .
. 32.3 43.1 0.7 -2.3
113 Canada 7.6 7.7 3.5 3.1 7.6 5.9 35.3 37.3 19.5 11.7 26.5 34.3 20.1 23.4 -1.3 -3.0
114 Denmark 7.3 5.1 16.0 9.0 10.0 1.3 41.6 41.1 11.3 7.2 13.7 362 32.6 41.2 2.7 4.7
115 Genaany, Fed. Rep. 12.4 8.9 1.5 0.6 17.5 18.2 46.9 49.4 11.3 7.1 10.4 15.8 24.2 29.9 0.7 -1.5
116 Finland 6.1 5.3 15.3 13.9 10.6 10.6 28.4 36.1 27.9 20.1 11.6 14.1 24.3 30.2 1.2 0.3
117 Sweden 12.5 6.8 14.8 9.2 3.6 1.1 44.3 54.2 10.6 8.0 14.3 20.7 27.9 40.8 -1.2 2.2
118 UnitedStates 32.2 24.8 3.2 1.7 8.6 12.5 35.3 31.5 10.6 6.5 10.1 23.0 19.1 22.9 -1.5 -3.2
119 Norway 9.7 8.3 9.9 8.2 12.3 10.7 39.9 36.1 20.2 19.5 8.0 17.2 35.0 41.5 -1.5 0.2
120 Japan' . . . . . . . . . . . . . . . . . . . . .
. 12.7 17.0 -1.9 -3.5
121 Switzerland 15.1 4.2 . . 10.0 . . 39.5 . . 18.4 . . 12.8 . . 13.3 . . 0.9
Total reporting economies 20.8 w 13.3w . . 5.4w 10.4w . . 38.3 w 33.7w 14.4w 10.1 w 16.1 w 26.1 w 21.9w 28.2w -2.1 w -3.6w
Oil exporters 15.0w . . 14.5 w . . 31.2 w . . 24.0w . . 22.0w . . 0.2 w -4.5w
Nonreporting nonmembers
a. See the technical notes. b. Refers to budgetaiy data.
199
Table 12. Central government current revenue
Percentage of total current revenue
Tax revenue
Taxes on Domestic Taxes on Total current
income, Social taxes international revenue as
profit, and security on goods trade and Nontax a percentage
capital gain contributions and services transactions Other taxes revenue of GNP
1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988
Low-income economies
China and India
Other low-income 28.0w :: 27.4w 19.3w 21.2w :: 19.5w
1 Mozambique
2 Ethiopia 23.0 304 00 00 298 224 304 191 56 22 111 259 105 255
3 Chad 16.7 0.0 . . 12.3 . . 45.2 . . 20.5 . . 5.3 . . 10.8
4 Tanzania 29.9 25.8 0.0 0.0 29.1 57.4 21.7 8.6 0.5 3.1 18.8 5.1 15.8 15.2
5 Bangtadesh' 3.7 11.7 0.0 0.0 22.4 33.2 18.0 31.5 3.8 7.1 52.2 16.5 8.6 8.6
6 Malawi" 31.4 33.7 0.0 0.0 24.2 33.0 20.0 16.0 0.5 0.4 23.8 16.8 16.0 20.6
7 Somalia" 10.7 0.0 . . 24.7 . . 45.3 . . 5.2 . . 14.0 . . 13.7
8 Zaire 22.2 26.8 2.2 0.7 12.7 14.6 57.9 42.5 1.4 2.3 3.7 13.8 14.3 15.3
9 Bhutan
10 LaoPDR
11 Nepal 4.1 8.4 0.0 0.0 26.5 36.1 36.7 31.1 19.0 5.4 13.7 19.1 5.2 10.3
12 Madagascar 13.1 . . 7.2 . . 29.9 . . 33.6 . . 5.5 . . 10.8 . . 18.3
13 BurkinaFaso 16.8 16.2 0.0 7.4 18.0 23.3 51.8 22.6 3.2 9.8 10.2 20.7 11.4 17.7
14 Mali . . 8.1 . . 4.5 . . 21.7 . . 27.5 . . 29.2 . . 9.1 . . 16.0
15 Bunindi 18.1 . . 1.2 . . 18.3 . . 40.3 . . 15.6 . . 6.5 . . 11.5
16 Uganda 22.1 6.0 0.0 0.0 32.8 24.3 36.3 69.6 0.3 0.0 8.5 0.0 13.7 8.2
17 Nigeria" 43.0 39.9 0.0 0.0 26.3 5.1 17.5 6.6 0.2 -14.5 13.0 62.9 9.4 18.5
18 Zambiab 49.7 37.9 0.0 0.0 20.2 40.3 14.3 17.3 0.1 0.5 15.6 4.0 23.2 15.8
19 Niger . . . . . . . . . . . . . . . . .
20 Rwanda 17.9 . . 4.4 . . 14.1 . . 41.7 . . 13.8 . . 8.1 . . 9.8
21 China .. .. .. .. .. .. .. .. .. .. ..
22 India 21.3 14.5 0.0 0.0 44.5 35.3 20.1 30.3 0.9 0.3 13.2 19.5 10.2 14.0
23 Pakistan 13.6 11.9 0.0 0.0 35.9 33.0 34.2 31.0 0.5 0.3 15.8 23.8 12.5 16.5
24 Kenya" 35.6 28.5 0.0 0.0 19.9 41.2 24.3 18.9 1.4 1.4 18.8 10.1 18.0 21.5
25 Togo . . 35.7 7.8 . . 9.6 35.4 1.1 . . 10.4 . . 25.3
26 CentralAfricanRep. . . 23.9 . . 0.0 . . 13.1 . . 45.2 . . 11.4 . . 6.4 . . 13.1
27 Haiti . . 11.8 0.0 . . 42.2 21.4 10.3 . . 14.3 . . 10.8
28 Benin .. .. .. .. .. .. .. .. .. .. ..
29 Ghana" 18.4 28.7 0.0 0.0 29.4 28.3 40.6 35.2 0.2 0.1 11.5 7.8 15.1 13.8
30 Lesotho 14.3 11.] 0.0 0.0 2.0 10.3 62.9 67.8 9.5 0.1 11.3 10.5 11.7 21.7
31 SriLanka 19.1 11.1 0.0 0.0 34.7 40.8 35.4 29.9 2.1 4.3 8.7 13.9 20.1 19.0
32 Guinea . . .
33 Yemen, PDR .. .. .. .. .. .. .. .. .. .. ..
34 Indonesia 45.5 55.9 0.0 0.0 22.8 24.5 17.6 5.6 3.5 3.0 10.6 11.0 13.4 19.2
35 Mauritania .. .. .. .. ..
36 Sudan" 11.8 . . 0.0 30.4 . . 40.5 1.5 15.7 . . 18.0
37 Afghanistan . . . . . . . . . . . . . . . . . . . . . .
38 Myanniar 28.7 6.7 0.0 0.0 34.2 39.8 13.4 14.3 0.0 0.0 23.8 39.2
39 Kampuchea, Dem. . . . . . . . . . . . . . . . . . . . . . . . . .
40 Liberia 40.4 33.9 0.0 0.0 20.3 25.1 31.6 34.6 3.1 2.3 4.6 4.2 17.0 17.0
41 SierraLeone" 32.7 20.1 0.0 0.0 14.6 22.4 42.4 53.6 0.3 0.9 9.9 3.0 19.5 7.3
42 VietNwn . . . . . . . . . . . . . . . . . . .
Middle-income economies 21.1 w 25.0w 20.8 w 28.4w 13.1 w 12.2 w . . 23.5 w . 16.1 w 15.7w
Lower-middle-income 22.0w 26.5 w . . . . 27.6 w 35.9w 19.4 w 14.7w . . . . 14.8 w 19.6w 11.9 w 11.7w
43 Bolivia 15.4 2.7 0.0 9.6 24.5 59.7 30.9 20.6 23.9 1.2 3.2 6.3 16.9 11.8
44 Philippines" 13.8 21.5 0.0 0.0 24.3 37.5 23.0 24.5 29.7 2.2 9.3 14.3 12.4 14.3
45 YemenArabRep. 6.1 19.7 0.0 0.0 10.3 10.3 56.5 20.7 9.6 9.3 17.5 40.0 8.0 23.4
46 Senegal 17.5 . . 0.0 . . 24.5 . . 30.9 . . 23.9 . . 3.2 . . 16.9
47 Zimbabwe . . 47.3 . . 0.0 . . 24.9 15.0 1.1 . . 11.7 . . 32.0
48 Egypt, Arab Rep. . . 14.9 . . 14.2 . . 11.3 . . 13.2 . . 8.1 . . 38.3 . . 34.3
49 DominicanRep. 17.9 16.3 3.9 4.0 19.0 25.8 40.4 39.9 1.7 1.9 17.0 12.1 17.2 16.4
50 Côte d'Ivoire . . . . . . . . . . . . . . . . . . . . .
51 PapuaNewGuinea" . . 43.6 . . 0.0 . . 11.8 . . 25.9 . . 1.7 . . 17.1 . . 23.5
52 Monicco 16.4 19.0 5.9 5.0 45.7 46.1 13.2 12.7 6.1 7.0 12.6 10.2 18.5 24.4
53 Honduras 19.2 . . 3.0 . . 33.8 . . 28.2 . . 2.3 . . 13.5 . . 13.2
54 Guatemala 12.7 20.3 0.0 0.0 36.1 27.5 26.2 37.2 15.6 6.8 9.4 8.2 8.9 10.3
55 Congo, People's Rep. 19.4 . . 0.0 . . 40.3 . . 26.5 . . 6.3 . . 7.5 . . 18.4
56 El Salvador 15.2 21.1 0.0 0.0 25.6 43.8 36.1 21.1 17.2 7.7 6.0 6.3 11.6 10.5
57 Thailand 12.1 19.9 0.0 0.0 46.3 46.3 28.7 22.9 1.8 2.9 11.2 8.0 12.5 17.4
58 Botswanab 19.9 43.3 0.0 0.0 2.4 1.4 47.2 13.8 0.4 0.1 30.0 41.5 30.7 74.1
59 Cameroon . . 31.3 . . 5.4 .. 14.9 18.7 4.0 . . 25.8 . . 18.9
60 Jamaica . . . . . . . . . . . . . . . . . . . . . . . . .
61 Ecuador" 19.6 48.6 0.0 0.0 19.1 25.6 52.4 19.6 5.1 3.7 3.8 2.6 13.6 14.2
62 Colombia 37.1 27.0 13.7 8.6 15.2 27.7 19.8 19.1 7.1 6.2 7.1 11.5 10.6 13.8
63 Paraguay 8.8 13.8 10.4 13.3 26.1 25.3 24.8 10.9 17.0 23.0 12.9 13.6 11.5 10.6
64 Tunisia 15.9 13.3 7.1 8.8 31.6 22.1 21.8 24.7 7.8 5.7 15.7 25.4 23.6 33.2
65 Turkey 30.8 39.5 0.0 0.0 31.0 32.0 14.6 6.6 6.1 3.3 17.5 18.5 20.6 17.9
66 Peru" 17.3 19.6 0.0 0.0 32.7 52.2 15.4 21.5 21.2 -0.1 13.5 6.7 15.2 9.0
67 Jordan 9.0 9.8 0.0 0.0 14.9 14.9 34.7 31.1 7.1 7.4 34.2 36.8 26.6 29.2
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
200
Percentage of total current revenue
Tax revenue
Taxes on Domestic Taxes on Total current
income, Social taxes international revenue as
profit, and security on goods trade and Nontax a percentage
capital gain contributions and services transactions Other taxes revenue of GNP
1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988 1972 1988
68 Chile 14.3 22.6 28.6 5.8 28.6 36.0 14.3 9.5 0.0 -0.2 14.3 26.2 30.2 31.7
69 Syrian Arab Rep. 6.8 32.3 0.0 0.0 10.4 6.1 17.3 5.8 12.1 23.1 53.4 32.7 25.1 22.6
70 Costa Rica 17.7 12.7 13.4 26.5 38.1 30.2 18.1 21.1 1.6 0.1 11.0 9.5 15.7 22.3
71 Mexico 37.3 26.8 18.6 11.8 32.2 68.5 13.6 3.4 -8.5 -18.1 6.8 7.5 10.1 18.0
72 Mauritius 22.7 10.9 0.0 4.0 23.3 18.6 40.2 50.8 5.5 5.2 8.2 10.5 15.6 25.1
73 Poland . . 30.4 . . 21.4 . . 30.4 . . 6.2 . . 6.5 . . 5.1 . . 38.7
74 Malaysia 25.2 32.2 0.1 0.7 24.2 18.0 27.9 17.3 1.4 2.0 21.2 29.8 20.3 25.1
75 Panama 23.3 21.9 22.4 24.7 13.2 14.4 16.0 10.0 7.7 3.2 17.3 25.8 21.8 31.9
76 Brazil 20.0 11.5 27.7 16.6 35.4 13.6 7.7 1.7 3.1 2.8 6.2 53.8 18.9 34.4
77 Angola . . . . . . . . . . . . . . . . . . . . .
78 Lebanon
79 Nicaragua 95 125 140 122 373 501 244 69 90 102 58 81 126 407
Upper-middle-income 20.5 w 15.2w . . 8.1w 30.1 w . . 22.6w
80 SouthAfrica 54.8 52.7 1.2 1.2 21.5 31.8 4.6 2.9 5.0 2.8 12.8 8.6 21.2 27.7
81 Algeria .. .. .. .. .. .. . .. .. ..
82 Hungaly . . 15.6 . . 24.5 . . 40.2 . . 5.4 . . 2.5 . . 11.7 . . 58.2
83 Umguay 4.7 7.9 30.0 25.6 24.5 44.7 6.1 12.2 22.0 5.2 12.6 4.4 22.7 22.1
84 Argentina 0.0 8.5 33.3 26.4 0.0 35.0 33.3 10.3 0.0 11.0 33.3 8.8 14.7 19.7
85 Yugoslavia 0.0 0.0 52.3 0.0 24.5 60.1 19.5 38.4 0.0 0.0 3.7 1.5 20.7 7.5
86 Gabonb 18.2 44.2 6.0 0.0 9.5 6.5 44.9 16.2 4.2 1.9 17.2 31.2 28.3 47.1
87 Venezuela 54.2 61.7 6.0 3.2 6.7 4.8 6.1 13.7 1.1 1.9 25.9 14.7 18.5 28.2
88 Trinidad and Tobago . . . . . . . . . . . . . . . . . . . . . . . . .
89 Korea, Rep. of 29.0 30.3 0.7 3.8 41.7 37.3 10.7 14.0 5.3 4.9 12.6 9.6 13.1 18.3
90 Portugal . . 25.2 . . 22.7 . . 32.0 . . 3.0 . . 91.2 . . 8.8 . . 33.5
91 Greece 12.2 17.9 24.5 34.9 35.5 36.3 6.7 0.5 12.0 0.2 9.2 10.2 25.4 35.8
92 Oman 71.1 19.0 0.0 0.0 0.0 0.8 3.0 3.0 2.3 0.8 23.6 76.4 47.4 35.9
93 Libya .. .. .. .. .. .. .. .. .. .. ..
94 iran, islamic Rep. 7.9 13.4 2.7 9.5 6.4 8.0 14.6 9.0 4.9 5.6 63.6 54.4 26.2 19.4
95 Iraq .. .. .. .. .. .. .. .. ..
96 Ronsania 6.0 0.0 8.2 16.5 0.0 0.0 0.0 0.0 0.0 12.3 85.8 71.2
Low- and middle-income 20.6 w 23.4w 25.2 w 29.3w 15.4w 14.9w 22.1 w . . 13.5w 16.5w
Sub-Saharan Africa
East Asia
South Asia . . 13.7w 35.5w 29.6 w 20.6w :: 14.9w
Europe, M.East, & N.Africa
Latin America & Caribbean 25.5 w 24.4 w 20.1w 37.7w 12.1 w 13.4w 11.1 w 12.1 w 10.4w 10.4w
Severely indebted 23.4w 25.0w 21.9w 39.5w 12.0w 13.4w 10.9w 11.9w 10.5 w 11.2w
High-income economies 44.0 w 37.8 w 23.3 w 20.0w 2.3w 1.1w 6.5w 8.5w 22.0w 25.5w
OECD members 44.3 w 38.1 w 23.5 w 20.1 w 2.2w 1.1w 6.2w 7.8w 21.7w 25.2w
tOther
97 tSaudi Arabia . . . . . . . . . . . . . . . . . . . . .
98 Spain 15.9 24.6 38.9 41.0 23.4 19.4 10.0 4.8 0.7 1.7 11.1 8.5 19.7 27.8
99 Ireland 28.3 35.8 9.0 13.1 32.1 30.2 16.7 7.1 3.2 3.1 10.6 10.7 30.1 46.7
100 tlsmel 40.0 42.9 0.0 7.0 20.0 32.3 20.0 3.8 10.0 1.8 10.0 12.1 31.3 41.6
101 tSingapore 24.4 19.1 0.0 0.0 17.6 14.5 11.1 2.7 15.5 9.7 31.4 54.0 21.5 29.6
102 tHong Kong . . . . . . . . . . . . . . . . . . . . . . . . .
103 NewZealandh 61.4 51.4 0.0 0.0 19.9 26.3 4.1 2.8 4.5 2.0 10.0 17.5 29.8 46.6
104 Australia 58.3 61.6 0.0 0.0 21.9 22.1 5.2 4.6 2.1 0.5 12.5 11.3 22.2 27.5
105 UnitedKingdom 39.4 38.0 15.6 18.5 27.1 30.6 1.7 0.1 5.4 2.5 10.8 10.2 32.6 36.4
106 Italy 16.6 37.1 39.2 37.3 31.7 24.2 0.4 0.0 4.3 -1.0 7.7 2.3 24.9 36.9
107 tKuwait 68.8 0.9 0.0 0.0 19.7 0.7 1.5 1.6 0.2 0.1 9.9 96.7 55.2 51.9
108 Belgium 31.3 36.8 32.4 34.8 28.9 22.9 1.0 0.0 3.3 2.2 3.1 3.3 35.1 45.0
109 Netherlands 32.5 26.5 36.7 40.4 22.3 22.1 0.5 0.0 3.4 2.5 4.7 8.6 43.4 51.0
I 10 Austria 20.7 19.6 30.0 36.5 28.3 25.7 5.4 1.6 10.2 8.2 5.5 8.3 29.7 35.5
Ill tUnitedArabEmirates' 0.0 . . 0.0 . . 0.0 . . 0.0 0.0 100.0 . . 0.2
112 France 16.8 17.4 37.0 42.4 37.9 29.4 0.3 0.0 3.0 3.1 4.9 7.6 33.4 41.4
113 Canada 54.0 51.8 8.8 14.9 15.9 18.9 11.0 3.8 -0.6 0.0 10.9 10.7 21.1 20.5
114 Denmark 40.0 38.6 5.1 2.8 42.1 42.2 3.1 0.1 2.8 3.1 6.8 13.3 35.5 42.9
115 Germany, Fed. Rep. 19.7 17.7 46.6 54.9 28.1 22.6 0.8 0.0 0.8 0.1 4.0 4.6 25.3 28.5
116 Finland 30.0 28.9 7.8 10.3 47.7 46.3 3.1 0.9 5.8 3.9 5.5 9.8 26.5 29.8
117 Sweden 27.0 17.8 21.6 29.7 34.0 29.0 1.5 0.6 4.7 9.6 11.3 13.3 32.4 42.9
118 UnitedStates 59.4 51.5 23.6 34.2 7.1 3.6 1.6 1.7 2.5 0.8 5.7 8.1 17.6 19.7
119 Norway 22.6 16.1 20.6 23.2 48.0 41.2 1.6 0.6 1.0 1.3 6.2 17.7 36.8 46.1
120 Japun' . . . . . . . . . . . . . . . . . . . . .
. 11.2 13.6
121 Switzerland 13.9 . . 37.3 . . 21.5 . . 16.7 . . 2.6 . . 8.0 . . 14.5
Total reporting economies 40.6 w 35.7 w 23.2w 20.7w 3.8w 2.6w . . 8.0w 9.7w 20.9w 24.5w
Oil exporters 26.4 w 24.2w 10.9w .. .. 20.2w
Nonreporting nonmembers
a. See the technical notes. b. Refers to budgetary data.
201
Table 13. Money and interest rates
Monetary holdings, broadly defined Average
Nominal interest rates of banks
Average annual annual
(average annual percentage)
nominal growth Average outstanding as a inflation
rate (percent) percentage of GDP (GDP deflator) Deposit rate Lending rate
1965-80 1980-88 1965 1980 1988 1980-88 1980 198.8 1980 1988
Low-income economies
China and India
Other low-income
1 Mozambique . . . . . . . . . . 33.6 . .
2 Ethiopia 12.7 11.9 12.5 25.3 43.5 2.1 . 7.50 . . 6.00
3 Chad 12.5 14.4 9.3 20.0 26.4 3.2 5.50 4.31 11.00 10.79
4 Tanzania 19.7 21.5 37.2 24.8 25.7 4.00 17.46 11.50 29.63
5 Bangladesh 23.8 16.9 . . 11.1 8.25 12.00 11.33 16.00
6 Malawi 15.4 17.7 17.6 20.3 25.0 12.6 7.92 13.50 16.67 22.25
7 Somalia 20.4 42.4 12.7 17.8 17.2 38.4 4.50 16.25 7.50 33.67
8 Zaire 28.2 59.4 11.1 8.9 10.4 56.1
9 Bhutan 26.5 . . 8.9 . . . . .
10 Lao PDR 8.1 . . 7.20 14.00 4.80 15.00
11 Nepal 17.9 19.3 8.4 21.9 31.1 . . 4.00 8.50 14.00 15.00
12 Madagascar 12.2 16.3 19.6 27.6 26.1 17.3 5.63 11.50 9.50
13 Burkina Faso 17.1 12.9 9.3 18.5 24.9 3.2 13,55 5.25 9.38 7.29
14 Mali 14.4 12.3 . . 17.9 23.4 3.7 13.71 5.25 9.38 6.63
15 Bunindi 15.7 9.9 10.1 13.3 17.0 4.0 2.50 4.00 12.00 12.00
16 Uganda 23.1 77.8 . . 12.7 78 100.7 6.80 26.00 10.80 35.00
17 Nigeria 28.5 11.9 9.9 21.5 24.6 11.6 5.27 13.09 8.43 13.96
18 Zambia 12.7 28.9 . . 32.6 30.6 33.5 7.00 11.44 9.50 18.39
19 Niger 18.3 6.1 3.8 13.3 18.1 3.6 6.19 5.25 9.38 8.00
20 Rwanda 19.0 10.5 15.8 13.6 17.3 4.1 6.25 6.25 13.50 12.00
21 China . . 25.9 . . 33.5 63.4 4.9 5.40
22 India 15.3 17.0 23.7 36.2 45.4 7.4 16.50 16.50
23 Pakistan 14.7 14.4 40.7 38.7 40.2 6.5 . . . .
24 Kenya 18.6 14.9 . . 37.7 37.9 9.6 5.75 10.33 10.58 15.00
25 Togo 20.3 8.9 10.9 29.0 38.1 6.1 12.71 6.71 9.38 7.13
26 CentralAfricanRep. 12.7 6.3 13.5 18.9 18.1 6.7 5.50 7.44 10.50 12.25
27 Haiti 20.3 8.1 9.9 26.1 16.4 7.9 10.00 . . .
28 Benin 17.3 5.1 10.6 21.1 18.2 8.0 13.71 5.25 9.38 7.13
29 Ghana 25.9 45.0 20.3 16.2 12.3 46.1 11.50 16.50 19.00 25.58
30 Lesotho . . 18.8 . . 47.3 12.2 9.60 9.58 11.00 13.67
31 Sri Lanka 15.4 16.0 32.3 35.3 37.1 11.0 14.50 13.23 19.00 12.42
32 Guinea . . . . . . .
33 Yemen,PDR 15.2 11.1 . . 114.8 174.8 4.5 . . . . .
34 Indonesia 54.4 23.8 . . 13.2 28.5 8.5 6.00 17.72 . . 22.10
35 Mauritania 20.7 12.1 5.7 20.5 23.8 9.4 . . 6.00 . . 12.00
36 Sudan 21.6 28.1 14.1 32.5 23.3 33.5 6.00 . . .
37 Afghanistan 14.0 16.2 14.4 26.8 9.00 9.00 13.00 13.00
38 Myanmar 11.5 10.3 1.50 1.50 8.00 8.00
39 Kampuchea, Dem.
40 Liberia 10.30 5.43 18.40 13.36
41 Sierra Leone 15.9 50.6 11.7 20.6 14.5 9.17 16.33 11.00 28.00
42 VietNam
Middle-income economies
Lower-middle-income
43 Bolivia 24.3 589.2 10.9 16.2 21.7 482.8 18.00 . . 28.00
44 Philippines 17.7 16.1 19.9 19.0 21.1 15.6 12.25 11.32 14.00 15.92
45 YemenArabRep. . . 20.4 . . 61.8 60.3 11.6 9.33 9.50 .
46 Senegal 15.6 7.6 15.3 26.6 22.5 8.1 6.19 5.25 9.38 6.96
47 Zimbabwe . . 18.1 . . 54.6 44.0 12.1 3.52 9.68 17.54 13.00
48 Egypt, Arab Rep. 17.7 22.2 35.3 52.2 86.8 10.6 8.33 11.00 13,33 17.00
49 Dominican Rep. 18.5 22.4 18.0 21.8 29.5 16.8 . . . . .
50 Côted'Ivoire 20.4 7.1 21.8 25.8 32.3 3.8 13.55 5.25 9.38 7.13
51 PapuaNewGuinea . . 8.9 . . 32.9 31.3 4.7 6.90 9.27 11.15 12.68
52 Morocco 15.7 14.5 29.4 42.4 50.5 7.7 4.88 8.50 7.00 9.00
53 Honduras 14.8 11.7 15.4 22.8 33.0 4.7 7.00 8.63 18.50 15.38
54 Guatemala 16.3 14.7 15.2 20.5 22.2 13.3 9.00 12.17 11.00 15.17
55 Congo, People's Rep. 14.2 8.6 16.5 14.7 21.5 0.8 6.50 7.81 11.00 11.79
56 El Salvador 14.3 17.1 21.6 28.1 28.5 16.8
57 Thailand 17.9 18.0 23.6 37.6 65.0 3.1 12.00 9.50 18.00 15.00
58 Botswana . . 25.1 . . 30.7 34.3 10.0 5.00 5.00 8.48 7.83
59 Camemon 19.1 10.8 11.7 18.3 18.3 7.0 7.50 7.21 13.00 13.46
60 Jamaica 17.2 25.9 24.3 35.6 59.6 18.7 10.29 17.92 13.00 23.00
61 Ecuador 22.6 31.8 15.6 20.2 16.1 31.2 . . 34.00 12.00 26.00
62 Colombia 26.5 . . 19.8 23.7 24.1 . . 28.28 19.00 28.22
63 Paraguay 21.3 20.0 12.1 19.8 15.6 22.1 . . . . . .
64 Tunisia 17.4 15.5 30.2 42.1 . . 7.7 2.50 7.37 7.25 9.87
65 Turkey 27.5 50.3 23.0 17.2 22.8 39.3 10.95 35.40 25.67 50.00
66 Peru 25.9 100.8 18.7 16.3 . . 119.1 . .
67 Jordan 19.1 12.8 88.8 152.2 2.2 . . . . . .
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
202
Monetary holdings, broadly defined
Average
annual Nominal interest rates of banks
Average annual
(average annual percentage)
nominal growth Average oat standing as a inflation
rate (percent) percentage of GD? (GD? deflator) Deposit rate Lending rate
1965-80 1980-88 1965 1980 1988 1980-88 1980 1988 1980 1988
68 Chile 139.9 . 16.3 21.3 .. 20.8 37.46 26.60 47.14 38.28
69 SyrianArabRep. 21.9 19.8 24.6 40.9 . . 12.9 5.00 .
70 CostaRica 24.6 26.7 19.3 38.8 36.1 26.9 . 15.18 . . 28.69
71 Mexico 21.9 62.6 25.1 27.5 14.0 73.8 20.63 52.93 25.20 52.70
72 Mauritius 21.8 20.2 27.3 41.1 58.9 7.8 9.25 10.00 12.19 14.96
73 Poland 26.7 58.4 32.9 30.5 3.00 21.00 8.00 16.67
74 Malaysia 21.5 13.0 26.3 69.8 117.4 1.3 6.23 3.00 7.75 7.25
75 Panama 3.3
76 Brazil 43.4 20.6 18.0 188.7 115.00 859.43
77 Angola
78 Lebanon 16.2 42.3 83.4 . . .
79 Nicaragua 15.0 . . 15.4 21.0 86.6 7.50
Upper-middle-income
80 South Africa 14.0 15.8 56.6 49.5 52.1 13.9 5.54 13.54 9.50 15.33
81 Algeria 22.3 17.5 32.1 58.5 . . 5.6 . . . . .
82 Hungary . . 7.6 . . 46.5 43.0 6.4 3.00 9.00 9.00 13.00
83 Uruguay 65.8 57.1 28.0 31.2 36.3 57.0 50.30 67.82 66.62 101.98
84 Argentina 86.0 284.0 22.2 19.6 290.5 79.40 432.75 . . 430.38
85 Yugoslavia 25.7 67.0 43.6 59.1 50.3 66.9 5.88 279.21 11.50 455.17
86 Gabon 25.2 6.5 16.2 15.2 24.6 0.9 7.50 8.17 12.50 11.79
87 Venezuela 22.3 16.4 17.3 36.3 43.6 13.0 . . 8.95 . . 8.50
88 TnnidadandTobago 23.1 12.4 21.3 32.0 . . 5.3 6.57 6.03 10.00 12.58
89 Korea, Rep. of 35.5 19.5 11.1 31.7 44.6 5.0 19.50 10.00 18.00 10.13
90 Portugal 19.5 21.4 77.7 96.3 98.7 20.1 18.20 . . 18.50
91 Greece 21.4 25.1 35.0 61.6 80.0 18.9 14.50 17.32 21.25 22.89
92 Oman . . 17.7 . . 12.3 28.6 -6.5 . . 7.57 . . 9.40
93 Libya 29.2 2.1 14.2 34.7 0.1 5.13 5.50 7.00 7.00
94 IranislamicRep. 28.6 21.6
95 Iraq . . . . 19.7
96 Romania . . 7.5 33.2
Low- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N.Afiica
Latin America & Caribbean
Severely indebted
High-income economies
OECD members
'(Other
97 tSaudi Arabia 32.1 9.4 16.4 18.6 63.7 -4.2 . . . . .
98 Spain 19.7 9.7 59.2 75.2 65.3 10.1 13.05 9.06 16.85 12.43
99 Ireland 16.1 6.0 . . 58.1 45.0 8.0 12.00 3.63 15.96 8.29
100 '(Israel 52.7 137.9 15.3 56.9 65.0 136.6 . . 19.39 176.93 61.43
101 '(Singapore 17.6 12.1 58.4 74.4 116.1 1.2 9.37 2.74 11.72 5.96
102 tHongKong . . . . . . 69.3 . . 6.7 . . . . .
103 New Zealand 12.8 16.4 56.5 53.4 11.5 11.4 11.0 16.32 12.63 20.84
104 Australia 13,1 12.7 50.0 46.6 47.7 7.8 8.58 11.92 10.58 18.52
105 UnitedKingdom 13.8 13.2 47.8 46.1 5.7 5.7 14.13 5.35 16.17 10.29
106 Italy 18.0 12.2 69.0 82.7 74.8 11.0 12.70 6.69 19.03 13.57
107 '(Kuwait 17.8 5.3 28.1 33.1 88.2 -3.9 4.50 4.50 6.80 6.80
108 Belgium 10.4 6.8 59.2 57.0 58.0 4.8 7.69 4.54 . . 8.92
109 Netherlands 14.7 5.8 54.4 79.0 87.7 2.0 5.96 3.48 13.50 7.77
110 Austria 13.3 7.3 48.9 72.6 84.9 4.0 5.00 2.73 .
Ill '(United Arab Emirates 11.8 19.0 66.2 0.1 9.47 . . 12.13
112 France 15.0 9.9 53.7 69.7 74.6 7.1 6.25 5.01 18.73 15.65
113 Canada 15.3 7.8 40.2 64.4 64.0 4.6 12.86 9.54 14.27 10.83
114 Denmark 11.5 15.6 46.0 42.6 . . 6.3 10.80 7.75 17.20 12.59
115 Germany, Fed. Rep. 10.1 5.7 46.1 60.4 64.9 2.8 7.95 3.29 12.04 8.33
116 Finland 14.7 14.3 39.1 39.5 50.8 7.1 9.00 7.75 9.77 9.72
117 Sweden 10.7 10.7 46.8 46.5 49.8 7.5 11.25 8.88 15.12 13.32
118 United States 9.2 9.3 64.0 58.8 66.3 4.0 13.07 7.73 15.27 9.31
119 Norway 12.6 12.2 51.9 51.6 62.1 5.6 5.07 8.03 12.63 14.28
120 Japan 17.2 8.8 106.7 134.0 177.4 1.3 5.50 1.76 8.35 5.03
121 Switzerland 7.1 8.0 101.1 107.4 123.0 3.8 7.75 2.69 5.56 5.07
Total reporting economies
Oil exporters
Nonreporting nonmembers
203
Table 14. Growth of merchandise trade
Average annual growth rate'
Merchandise trade
(millions of dollars) (percent)
Terms of trade
Exports Imports Exports Imports (1980 =100)
1988 1988 1965-80 1980-88 1965-80 1980-88 1985 1988
Low-income economies 107,355 t 131,444 t 5.6w 4.1 w 4.5 w 2.6w 92 m 93 m
China and India 62,140 t 77,751 t 4.8w 10.0w 4.5 w 10.2w 104 m 101 m
Other low-income 45,215 t 53,693 t 5.9w 0.5 w 4.5 w -3.2 w 91 m 93 m
Mozambique
1 104 706
2 Ethiopia 374 1,099 -o. -o 7.2 99 iô
3 Chad 148 366
4 Tanzania 373 1,185 -4.0 1.6 0.5 90 94
5 Bangladesh 1,231 2,987 6.1 3.3 124 111
6 Malawi 301 412 4.1 3.3 3.3 -3.4 69 72
7 Somalia 58 354 3.8 -9.7 5.8 -4.1 91 91
8 Zaire 2,207 1,954 4.7 -2.9 -2.9 0.2 82 96
9 Bhutan . . . . . . . .
10 LaoPDR 58 188 .
11 Nepal 186 628 -2.3 5.5 3.0 7.0 91 93
12 Madagascar 282 382 0.7 -3.5 -0.4 -1.8 104 95
13 BurkinaFaso 249 697 6.8 6.5 5.8 2.2 80 69
14 Mali 255 513 11.0 7.0 6.2 3.7 82 88
15 Burundi 123 165 3.0 8.4 2.0 1.1 100 81
16 Uganda 298 518 -3.9 2.6 -5.3 4.6 96 78
17 Nigeria 7,390 6,324 11.4 -3.6 15.2 -13.7 90 40
18 Zambia 1,073 889 1.7 -3.7 -5.5 -4.8 71 107
19 Niger 369 430 12.8 -4.9 6.6 -4.2 109 83
20 Rwanda 113 370 7.7 1.3 8.7 5.8 102 108
21 China* 47,540 55,251 5.5 11.9 7.9 13.1 95 84
22 India 14,600 22,500 3.7 4.7 1.6 5.4 114 119
23 Pakistan 4,362 7,521 4.3 8.4 0.4 3.8 88 106
24 Kenya 1,034 1,989 0.3 0.1 1.7 -0.6 92 91
25 Togo 334 411 4.6 -0.3 8.6 -3.8 91 80
26 CentralAfncanRep. 132 236 0.4 0.1 -1.1 3.5 87 94
27 Haiti 207 300 7.0 -2.6 8.4 -2.4 97 101
28 Benin 225 413 5.2 2.4 6.7 2.7 90 94
29 Ghana 882 1,091 -1.8 1.1 -1.4 -1.4 91 78
30 Lesotho 55 534 b b b b b
31 SriLanka 1,472 2,241 0.5 5.8 -1.2 3.4 99 102
32 Guinea 584 468 . . . . . . . . .
33 Yemen,PDR 80 598 -13.7 1.9 -7.5 4.4 99 76
34 Indonesia 19,677 15,732 9.6 2.9 14.2 -2.1 94 70
35 Mauritania 433 353 2.7 9.7 5.4 2.4 112 104
36 Sudan 486 1,223 -0.3 2.7 2.3 -7.9 90 86
37 Afghanistan . . . . . . . . . . . .
38 Myanmar 299 611 -2.1 -7.0 -1.7 -8.0 70 72
39 Kampuchea, Dem. . . . . . . . . . . . . .
40 Liberia 382 308 4.5 -3.2 1.5 -9.8 91 103
41 SierraLeone 106 156 -3.8 -3.2 -2.7 -13.1 100 94
42 VietNam . . . . . . . . .
Middle-income economies 341,143 t 338,711 t 2.4w 5.8w 5.9w 0.6w 92m 86m
Lower-middle-income 172,809 t 163,123 t 5.8 w 6.0 w 5.2w -0.2w 92m 87m
43 Bolivia 541 700 2.8 -0.5 5.0 -2.6 84 57
44 Philippines 7,074 8,159 4.7 0.4 2.9 -1.7 92 110
45 YemenArabRep. 853 1,310 2.8 35.6 23.3 -10.0 93 40
46 Senegal 761 1,147 2.4 7.0 4.1 2.8 100 96
47 Zimbabwe 1,589 1,325 3.4 1.5 -1.8 -6.0 84 83
48 Egypt, Arab Rep. 4,499 10,771 2.7 6.2 6.0 1.5 84 62
49 Dominican Rep. 893 1,608 1.7 0.0 5.5 2.3 66 76
50 Côted'Ivoire 2,359 1,542 5.6 1.5 8.0 -2.2 96 92
51 PapuaNewGuinea 1,464 1,589 12.8 6.4 1.3 1.1 95 89
52 Morocco 3,624 4,818 3.7 5.0 6.5 1.8 89 103
53 Honduras 919 940 3.1 2.8 2.5 -0.3 93 102
54 Guatemala 1,074 1,548 4.8 -2.0 4.6 -3.0 87 87
55 Congo, People's Rep. 912 611 12.5 4.6 1.0 -2.1 94 49
56 ElSalvador 573 975 2.4 -4.4 2.7 -0.6 96 86
57 Thailand 15,806 17,876 8.5 11.3 4.1 6.2 74 82
5 b b I'
58 Botswana 1,418 1,031
59 Cameroon 1,639 1,484 5.2 6.8 5.6 2.5 92 64
60 Jamaica 832 1,428 -0.3 -4.5 -1.9 -0.5 95 97
61 Ecuador 2,203 1,714 15.1 5.7 6.8 -2.8 94 50
62 Colombia 5,339 4,515 1.4 8.2 5.3 -3.5 98 68
63 Paraguay 919 878 7.9 15.7 4.6 4.9 82 102
64 Tunisia 2,397 3,692 10.8 3.0 10.4 -1.6 83 77
65 Turkey 11,662 14,340 5.5 15.3 7.7 10.3 91 115
66 Peru 2,694 2,750 2.3 -2.5 -0.2 -3.2 81 80
67 Jordan 875 2,751 13.7 6.5 9.7 0.3 93 102
* Data for Taiwan, China, are: 60,382 44,584 19.0 15.1 8.5
13.9 104 105
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
204
Merchandise trade Average annual growth rate
(millions of dollars) (percent)
Terms oftrade
Exports Imports Exports Imports (1980 = 100)
1988 1988 1965-80 1980-88 1965-80 1980-88 1985 1988
68 Chile 7,052 4,833 7.9 4.5 2.6 -6.0 79 94
69 SyrianArabRep. 1,345 2,223 11.4 -0.5 8.5 -8.2 97 56
70 Costa Rica 1,270 1,409 7.0 2.9 5.7 -0.3 95 98
71 Mexico 20,658 18,903 7.6 5.5 5.7 -4.9 98 67
72 Mauritius 1,110 1,115 3.1 12.1 6.4 8.7 90 117
73 Poland 13,211 12,064 . . 4.7 . . 2.3 106 116
74 Malaysia 20,848 16,584 4.4 9.4 2.9 0.4 87 74
75 Panama 2,352 2,815 . . 1.2 . . -5.6 94 104
76 Brazil 33,689 14,691 9.3 6.0 8.2 -2.9 89 117
77 Angola . . .
78 Lebanon . . . . . . . . . . . . .
79 Nicaragua 236 791 2.3 -6.3 1.3 0.4 85 84
Upper-middle-income 168,333 t 175,588 t 0.9w 4.4w 6.8 w 1.4 w 91 m 75 m
80 South Africa 19,714 16,664 6.1' 02b 01b 66b 75b 735
81 Algeria 7,674 7,432 1.5 3.4 13.0 -5.9 97 41
82 Hungaiy 9,922 9,326 . . 5.4 . . 1.7 92 75
83 Uniguay 1,402 1,177 4.6 2.1 1.2 -4.2 87 99
84 Argentina 9,134 5,324 4.7 0.1 1.8 -8.0 90 86
85 Yugoslavia 12,779 13,329 5.6 0.9 6.6 -1.3 Ill 120
86 Gabon 1,204 998 8.1 -2.2 10.5 0.8 90 54
87 Venezuela 10,234 11,581 -9.5 0.4 8.7 -3.9 93 41
88 TrinidadandTobago 1,160 1,247 -5.5 -6.0 -5.8 -15.0 96 55
89 Korea,Rep.of 60,696 51,811 27.2 14.7 15.2 9.9 106 108
90 Portugal 10,218 16,038 3.4 11.6 3.7 6.3 85 107
91 Greece 5,400 11,978 11.9 4.6 5.2 3.4 88 89
92 Oman 3,941 1,822 . . . . . . . . .
93 Libya 5,640 6,386 3.3 -5.4 15.8 -14.8 91 47
94 Iran,IslamicRep. 9,454
95 Iraq 9,014 10,268
96 Romania
Low- and middle-income 448,498 470,155 3.2w 5.4w 5.6w 1.1 w 92 m 89 m
Sub-Saharan Africa 28,871 32,738 6.6w -0.7w 4.9w -5.0w 91 m 92 m
East Asia 173,653 167,930 9.7 w 10.4 w 8.6 w 6.9 w 94 m 84m
South Asia 21,712 35,950 1.7w 5.4w 0.6w 4.4w 95 m 104 m
Europe, M.East, & N.Africa 102,798 138,333 0.3w 92 m 83 m
Latin America & Caribbean 101,750 78,540 -2.0w 3.2w 4.4w -4.1 w 90 m 86 m
Severely indebted 127,659t 101,688t -1.0w 3.6w 5.6w -2.3w 92m 92m
High-income economies 2,178,528 I 2,265,978 t 7.0 w 3.4w 4.4 w 4.9w 95 m 98 m
OECD members 2,024,259 t 2,110,250 t 7.2 w 4.1 w 4.2 w 5.1 w 94 m 103 m
tOther 154,2691 155,728 t 6.0 w -4.2 w 10.4w 0.6w 96 m 64 m
97 tSaudi Arabia 23,138 20,465 8.8 -16.3 25.9 -9.3 95 54
98 Spain 40,458 60,434 12.4 7.7 4.4 7.2 90 103
99 Ireland 18,736 15,558 9.8 7.8 4.8 3.2 107 112
100 tlsrael 9,605 15,030 8.9 7.6 6.3 4.4 94 92
101 tSingapore 39,205 43,765 4.7 7.3 7.0 4.9 101 101
102 tHong Kong 63,161 63,894 9.5 12.3 8.3 10.4 103 105
103 NewZealand 8,785 7,304 4.2 3.9 1.1 3.5 97 110
104 Australia 25,283 29,318 5.5 5.8 0.9 3.0 89 74
105 UnitedKingdom 145,076 189,466 4.8 3.1 1.4 4.9 96 93
106 Italy 128,534 135,514 7.7 3.8 3.5 4.3 95 108
107 tKuwait 7,160 5,348 -1.9 -2.9 11.8 -5.5 95 54
108 Belgiumc 88,953 9l,098 7.8 4.7 5.2 2.8 87 89
109 Netherlands 103,206 99,743 8.0 4.7 4.4 3.4 91 91
110 Austria 28,111 36,579 8.2 4.3 6.1 4.0 90 98
111 tUnitedArabEmirates 12,000 7,226 10.9 0.1 20.5 -7.1 91 54
112 France 161,702 176,745 8.5 3.4 4.3 2.6 94 101
113 Canada 111,364 112,180 5.4 6.4 2.6 8.4 122 119
114 Denmark 27,816 26,458 5.4 5.8 1.7 5.2 96 107
115 Germany,Fed.Rep. 322,555 248,999 7.2 4.6 5.3 3.3 88 106
116 Finland 21,639 20,911 5.9 3.5 3.1 4.2 96 114
117 Sweden 49,867 45,793 4.9 5.5 1.8 3.3 89 95
118 UnitedStates 315,313 458,682 6.4 1.2 5.5 8.9 114 118
119 Norway 22,503 23,212 8.2 6.3 3.0 3.5 97 67
120 Japan 264,772 183,252 11.4 5.3 4.9 5.0 112 157
121 Switzerland 50,633 56,325 6.2 4.3 4.5 4.5 88 103
Total reporting economies 2,627,026 1 2,736,133 1 6.1 w 3.8 w 4.6w 4.1w 93m 92m
Oil exporters 161,995 t 163,458 1 3.0 w -3.6 w 9.3w -5.5w 94m 54m
Nonreporting nonmembers
a. See the technical notes. b. Figures for the South African Customs Union, comprising South Africa, Namibia, Lesotho, Botswana, and Swaziland, are included in
South African data. Trade among the component teeritones is excluded. c. Includes Luxembourg.
205
Table 15. Structure of merchandise imports
Percentage share of merchandise imports
Other Machinery
primary and transport Other
Food Fuels commodities equipment manufactures
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
Low-income economies 22w 7w 5w 9w 10 w 6w 28w 34w 34w 44w
China and India 28w 3w 3w 5w 19 w 7w 26w 35w 24w SOw
Other low-income 17w 11w 7w 14 w 4w 6w 29w 31w 42w 37w
I Mozambique 17 21 8 6 7 9 24 45
2 Ethiopia 6 5 6 23 6 3 37 34 44 36
3 Chad 13 19 3 23 42
4 Tanzania 10 6 9 14 2 2 34 46 45 32
5 Bangladesh 23 10 9 20 38
6 Malawi 15 6 5 6 3 2 21 44 57 41
7 Somalia 31 22 5 4 8 8 24 39 33 27
8 Zaire 18 23 7 3 5 6 33 31 37 36
9 Bhutan .. .
10 Lao PDR 27 15 6 18 33
11 Nepal 22 6 5 8 14 6 37 22 22 57
12 Madagascar 19 5 5 33 2 3 25 37 48 22
13 BurkinaFaso 23 19 4 3 14 4 19 33 40 41
14 Mali 20 13 6 16 5 1 23 43 47 26
15 Bunindi 16 11 6 5 9 4 15 19 55 62
16 Uganda 7 6 1 9 3 2 38 45 51 38
17 Nigeria 9 18 6 2 3 6 34 31 48 43
18 Zambia 9 7 10 9 3 1 33 38 45 46
19 Niger 12 21 6 8 6 10 21 30 55 32
20 Rwanda 12 11 7 21 5 7 28 28 50 33
21 China* 36 2 0 2 25 6 12 41 27 49
22 India 22 5 5 13 14 9 37 20 22 53
23 Pakistan 20 14 3 19 5 13 38 28 34 25
24 Kenya 10 11 11 22 3 4 34 31 42 31
25 Togo 15 34 3 5 5 10 31 21 45 31
26 CentralAfncanRep. 13 11 7 8 2 3 29 36 49 43
27 Haiti 25 14 6 8 6 2 14 25 48 51
28 Benin 18 14 6 31 7 3 17 16 53 37
29 Ghana 12 8 4 17 3 3 33 36 48 37
30 Lesothoa . . . . . . . . .
31 Sri Lanka 41 17 8 17 4 3 12 27 34 37
32 Guinea . . . . . . . . . . . . . . .
33 Yemen, PDR 19 16 40 40 5 2 10 21 26 19
34 Indonesia 6 3 3 18 2 5 39 39 50 36
35 Mauritania 9 21 4 5 1 2 56 36 30 36
36 Sudan 23 7 5 19 4 2 21 36 47 36
37 Afghanisean 17 . 4 . . I . . 8 . 69
38 Myanmar 15 5 4 2 5 2 18 43 58 48
39 Kampuchea,Dem. 6 . . 7 . . 2 . . 26 . . 58
40 Liberia 16 20 8 22 3 2 34 28 39 28
41 SierraLeone 17 23 9 8 3 4 30 19 41 46
42 VieeNam . .
Middle-income economies 15w 11 w 8w lOw 11 w lOw 31 w 34w 36w 35 w
Lower-middle-income 16 w 12 w 8w 11 w 8w 8w 33 w 33 w 35 w 37 w
43 Bolivia 19 15 1 3 3 3 35 52 42 27
44 Philippines 20 10 10 13 7 7 33 20 30 50
45 YemenArabRep. 41 28 6 8 6 6 26 22 21 35
46 Senegal 36 40 6 12 4 3 15 16 38 30
47 Zimbabwe 13 8 8 6 3 3 31 35 46 48
48 Egypt,ArabRep. 26 19 7 2 12 8 23 29 31 41
49 Dominican Rep. 23 16 10 17 4 4 24 26 40 38
50 Côted'Ivoire 18 18 6 15 3 3 28 27 46 37
51 Papua New Guinea 23 24 5 10 3 2 25 34 45 30
52 Monicco 36 12 5 13 10 17 18 24 31 34
53 Honduras II 8 6 15 1 2 26 30 56 45
54 Guatemala II 6 7 10 2 3 29 27 50 54
55 Congo, People's Rep. 15 24 6 7 1 3 34 25 44 42
56 ElSalvador 15 15 5 8 4 4 28 19 48 53
57 Thailand 6 4 9 II 6 12 31 37 49 36
58 Botswanaa . . . . . . . . . . . . .
59 Cameroon 11 23 5 1 4 2 28 35 51 39
60 Jamaica 20 14 9 11 5 5 23 22 43 48
61 Ecuador 10 5 9 3 4 4 33 62 44 26
62 Colombia 8 9 1 4 10 7 45 37 35 43
63 Paraguay 24 15 14 18 4 8 31 35 28 25
64 Tunisia 16 18 6 7 7 14 31 22 41 39
65 Thrkey 6 2 10 20 10 8 37 34 37 36
66 Pent 17 19 3 1 5 4 41 44 34 33
67 Jordan 28 19 6 16 6 5 18 23 42 38
* Data for Taiwan, China, are: 13 6 5 9 25 14 29 36 29 35
Note: For data comparability and coverage, see the technical notes. Figures in italica are for years other than those specified.
206
Percentage share of merchandise imports
Other Machinery
primary and transport Other
Food Fuels commodities equipment manufactures
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
68 Chile 20 2 6 6 10 2 35 46 30 44
69 SyrianArabRep. 22 17 10 II 9 5 16 28 43 39
70 Costa Rica 9 5 5 12 2 2 29 28 54 52
71 Mexico 5 16 2 1 10 8 50 36 33 38
72 Mauritius 34 17 5 9 3 4 16 16 43 53
73 Poland 11 17 11 32 29
74 Malaysia 25 15 12 6 10 5 22 47 32 28
75 Panama 11 9 21 21 2 I 21 24 45 44
76 Brazil 20 14 21 28 9 7 22 25 28 26
77 Angola .. S S
78 Lebanon 28 . . a . . a- 17 . . 36
79 Nicaragua 12 25 5 11 2 3 30 17 51 44
Upper-middle-income 13 w 10 w 8w lOw 14w 12w 29w 35w 35w 34w
80 SouthAfrica 5 2 5 0 11 3 42 47 37 47
81 Algeria 26 30 0 2 6 5 15 27 52 35
82 Hungary 12 7 12 16 22 10 27 30 28 36
83 Uniguay 7 9 17 13 16 8 24 34 36 36
84 Argentina 6 4 10 8 21 7 25 43 38 38
85 Yugoslavia 16 7 6 18 19 12 28 27 32 36
86 Gabon 16 22 5 2 2 3 38 36 40 37
87 Venezuela 12 11 1 0 5 3 44 40 39 45
88 TrinidadandTobago It 19 50 12 2 5 16 26 22 38
89 Korea, Rep. of 15 6 7 12 26 17 13 35 38 30
90 Portugal 16 13 8 14 19 9 27 30 30 34
91 Greece 15 17 8 5 II 8 35 30 30 40
92 Oman 27 15 19 3 4 2 15 41 34 39
93 Libya 13 15 4 1 3 2 36 33 43 49
94 Iran, Islamic Rep. 16 0 6 . . 36 42
95 Iraq 24 0 7 , . 25 . . 44
96 Ro,nania
Low- and middle-income 17 w lOw 7w 10 w 11w 9w 30 w 35 w 35 w 37 w
Sub-Saharan Africa 14 w 16 w 6w 9w 4w 5w 30 w 30 w 44w 38 w
East Asia 21w 6w 6w 8w 15w 11w 23 w 38 w 34 w 38 w
South Asia 29 w 9w 4w 14 w 11w 9w 32 w 22 w 26 w 45 w
Europe, M.East, & N.Africa 14 w 13 w 9w 32 w 32 w
Latin America & Caribbean ii 12w 9w 9w 8w 6w 34 w 35w 36 w 38 w
Severely indebted 14 w 12 w 7w 11 w 14 w 7w 32w 33w 33w 37w
High-income economies 19 w 10 w 11 w 9w 19w 8w 20w 34w 31w 40w
OECD members 19 w 10 w 11 w 9w 20 w 9w 20w 34w 31w 39w
tOther 21w 9w lOw 7w 12 w 5w 19w 33w 37w 46w
97 tSaudi Arabia 29 17 1 1 5 2 27 34 38 46
98 Spain 19 II 10 11 16 9 27 38 28 30
99 Ireland 18 12 8 6 10 5 25 34 39 44
100 tlsrael 16 8 6 8 12 5 28 30 38 49
101 tSingapore 23 7 13 14 19 6 14 42 30 30
102 tHongKong 25 7 3 2 13 6 13 27 46 58
103 New Zealand 7 7 7 5 10 5 33 37 43 45
104 Australia 5 5 8 5 10 4 37 39 41 47
105 UnitedKingdom 30 10 11 5 25 8 II 37 23 41
106 Italy 24 13 16 9 24 12 15 29 21 37
107 lKuwait 21 17 1 1 7 2 33 39 39 41
108 Belgium' 14 Ii 9 7 21 9 24 24 32 48
109 Netherlands 15 15 10 9 13 6 25 28 37 41
110 Austria 14 6 7 6 13 8 31 36 35 44
Ill tUnited Arab Emirates 15 4 3 3 7 1 34 43 41 49
112 France 19 10 15 8 18 8 20 33 27 41
113 Canada tO 6 7 4 9 5 40 55 34 31
114 Denmark 14 12 It 6 11 7 25 28 39 47
115 Germany,Fed.Rep. 22 It 8 8 21 9 13 29 35 43
116 Finland 10 6 10 tO 12 8 35 39 34 38
117 Sweden 12 6 11 7 12 7 30 39 36 40
118 UnitedStates 19 6 10 10 20 5 14 43 36 36
119 Norway 10 6 7 4 12 8 38 39 32 43
120 Japan 22 17 20 21 38 20 9 13 11 30
121 Switzerland 16 7 6 4 11 5 24 32 43 52
Total reporting economies 18w lOw lOw 9w 18w 8w 22w 34w 32w 40w
Oil exporters 14w 13w 7w 4w 8w Sw 34w 37w 39w 42w
Nonreporting nonmembers
a. Figures for the South African Customs Union, comprising South Africa, Namibia, Lesotho, Botswana, and Swaziland, are included in South African data. Trade
among the component territories is excluded. b. Includes Luxembourg.
207
Table 16. Structure of merchandise exports
Percentage share of merchandise exports
Fuels Other Machinery
minerals, primary and transport Other (Textiles
and metals commodities equipment manufactures and cloth,ng)a
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
Low-income economies 21w 25w 53w 23w 1w 4w 23w 49w 11w
China and India 8w lOw 45w 18w 2w 6w 45w 67w
Other low-income 30w 44w 60w 29w 1w 1w 8w 24w 4w
1 Mozambique 14 3 83 89 0 . . 2 . .
2 Ethiopia 1 3 98 96 1 0 0 1 0
3Chad 4 .. 93 .. 0 .. 4
4 Tanzania 4 6 83 75 0 3 13 15
5 Bangladesh 2 29 0 69 67
6 Malawi 0 0 99 83 0 5 1 12 0
7 Somalia 6 0 80 95 4 0 10 5
8 Zaire 72 64 20 29 0 1 8 6
9 Bhutan . .
10 LaoPDR 25 65 0 . . tO
11 Nepal 0 4 78 29 0 2 22 65 . . 27
12 Madagascar 4 14 90 70 1 3 4 13 1 3
13 BurkinaFaso 1 0 94 98 I 1 4 1 0
14 Mali 1 0 96 70 1 1 2 29 0
15 Burundi 1 0 94 83 0 0 6 16 0
16 Uganda 14 4 86 96 0 0 1 0 0
17 Nigeria 32 88 65 10 0 0 2 2 0
18 Zambia 97 95 3 3 0 1 0 1 0
19 Niger 0 76 95 20 1 I 4 3 0
20 Rwanda 40 9 60 90 0 0 1 1
21 China* 6 10 48 17 3 4 43 69 . . 24
22 India 10 9 41 18 1 II 48 62 36 25
23 Pakistan 2 1 62 30 1 2 35 67 29 54
24 Kenya 13 20 81 63 0 2 6 15 0
25 Togo 49 69 48 23 1 1 3 8 0
26 Central African Rep. 1 0 45 60 0 0 54 40 0
27 Haiti 14 4 61 23 2 13 23 61
28 Benin I 36 94 38 2 7 3 19 0
29 Ghana 13 32 85 65 1 0 2 3 0
30 LesothOb
31 SriLanka 2 7 97 50 0 3 1 40 0 32
32 Guinea . . . . . . . . . .
33 Yemen, PDR 80 90 14 9 2 1 4 0 2
34 Indonesia 43 49 53 22 3 1 1 28 0 8
35 Mauritania 94 32 5 65 1 0 0 2 0
36 Sudan 1 13 98 80 1 3 0 4 0
37 Afghanistan 0 . . 86 . . 0 . . 13 . .
38 Myanmar 5 3 94 86 0 7 0 3 0
39 Kampuchea,Dem. 0 . . 99 . . 0 . . 1 .
40 Liberia 72 54 25 44 I 0 3 1 0
41 Sierra Leone 25 21 14 21 0 1 60 58 0
42 VietNam
Middle-income economies 35w 21w SOw 21w 2w 19w 13w 40w 3w
Lower-middle-income 27w 24w 58w 30w 1w 16w 12w 30w 2w
43 Bolivia 92 89 3 8 0 1 4 2 0 1
44 Philippines 11 12 84 26 0 10 6 52 1 7
45 Yemen Arab Rep. 9 88 91 1 0 9 0 2 .
46 Senegal 9 25 88 50 1 6 2 18 1 3
47 Zimbabwe 45 13 40 47 1 3 15 37 6 3
48 Egypt,Arab Rep. 8 64 72 10 0 0 20 25 15
49 Dominican Rep. 10 21 88 53 0 6 2 20 0
50 Cole d'Ivoire 2 3 93 85 1 3 4 9 0
51 Papua New Guinea 1 63 89 32 0 1 10 4
52 Morocco 40 21 55 29 0 2 5 48 1 18
53 Honduras 7 10 89 79 0 0 4 II
54 Guatemala 0 3 86 59 l 3 13 35 4
55 Congo,People's Rep. 5 72 32 17 2 1 61 10 0
56 ElSalvador 2 3 81 68 1 3 16 26 6
57 Thailand 11 3 84 45 0 II 4 41 0 17
58 Botswana5 . . . . . . . . . . . .
59 Camemon 17 53 77 35 3 2 2 10 0 1
60 Jamaica 28 15 41 26 0 3 31 56 4 14
61 Ecuador 2 45 96 48 0 2 2 4 1 0
62 Colombia 18 26 75 49 0 1 6 24 2 6
63 Paraguay 0 0 92 88 0 0 8 12 0 0
64 Tunisia 31 19 51 17 0 6 19 58 3 30
65 Turkey 9 6 89 30 0 7 2 57 1 32
66 Pem 45 58 54 20 0 3 1 18 0 7
67 Jordan 27 43 54 10 II 6 7 41 0 5
* Data for Taiwan, China, are: 2 1 56 7 4 34 37 58 5 15
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
208
Percentage share of merchandise exports
Fuels Other Machinery
minerals, primary and transport Other (Textiles
and metals commodities equipment manufactures and clothtng)
1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
68 Chile 89 67 7 18 1 3 4 12 0 0
69 Syrian Arab Rep. 7 50 83 25 1 3 9 22 7 6
70 Costa Rica 0 1 84 59 1 7 15 33 3
71 Mexico 22 38 62 7 1 33 15 22 3 2
72 Mauritius 0 0 100 39 0 4 0 58 0 47
73 Poland . 19 . . 14 . . 33 .. 34 . . 5
74 Malaysia 35 18 59 37 2 26 4 19 0 4
75 Panama 35 13 63 66 0 1 2 20 0 5
76 Brazil 9 21 83 31 2 18 7 30 1 3
77 Angola
78 Lebanon 13 . . 53 . . 14 . . 20 . . 2
79 Nicaragua 4 2 90 89 0 0 6 9 1
Upper-middle-income 41 w 18 w 37w 13w 3w 25w 20w 45w 4w
80 South Africab 24 II 44 9 3 3 29 77 1
81 Algeria 58 96 38 1 2 1 2 3 0 0
82 Hungary 5 8 25 22 32 31 37 38 9 7
83 Uruguay 0 I 95 52 0 3 5 44 2 13
84 Argentina 1 5 93 70 1 5 5 20 0 3
85 Yugoslavia 11 9 33 13 24 31 33 48 8 8
86 Gabon 50 59 39 27 1 3 10 11 0
87 Venezuela 97 90 1 1 0 3 2 6 0 0
88 Trinidad and Tobago 84 61 9 6 0 1 7 32 1 0
89 Korea, Rep. of 15 2 25 5 3 39 56 54 27 22
90 Portugal 4 4 34 16 3 17 58 64 24 30
91 Greece 8 15 78 30 2 3 11 52 3 31
92 Oman 90 91 10 2 0 5 0 2 . 0
93 Libya 99 99 1 0 1 1 0 1 0
94 Iran, Islamic Rep. 87 . . 8 . . 0 . . 4 . . 4
95 Iraq 95 4 .. 0 1 .. 0
96 Romania
Low- and middle-income 30 w 22 w 53 w 21w 1w 16 w 17 w 42 w 5w
Sub-Saharan Africa 34 w 45 w 58 w 38 w 1w 2w 6w 14 w Ow
East Asia 17 w 10 w 58 w 16 w 2w 23 w 21 w 51 w 2w
South Asia 6w 7w 57 w 24 w 1w 8w 36 w 61 w 27 w
Europe, M.East, & N.Africa
Latin America & Caribbean 43w 35w 50w 29w 1w 14w 6w 23 w 1w
Severely indebted 35 w 30 w 45 w 26w 5w 17 w 15 w 27 w
High-income economies 12 w 9w 20 w 12 w 29 w 40 w 38 w 40 w 7w 4w
OECD members 9w 7w 21 w 12 w 31 w 41 w 39 w 40w 7w 5w
(Other 57 w 31 w 13 w 6w 4w 23 w 26 w 40w 11w 14w
97 tSaudi Arabia 98 90 1 1 1 4 1 5 0
98 Spain 9 7 51 20 10 34 29 39 6 4
99 Ireland 3 2 63 28 5 32 29 39 7 5
100 (Israel 6 2 28 13 2 18 63 67 9 6
101 ISingapore 21 15 44 11 11 47 24 28 6 5
102 tHong Kong 2 2 11 6 6 25 81 66 43 29
103 New Zealand 1 8 94 68 0 6 5 19 0 2
104 Australia 13 37 73 38 5 8 10 17 1 1
105 United Kingdom 7 10 10 8 41 39 41 43 7 4
106 Italy 8 3 14 8 30 36 47 53 15 13
107 (Kuwait 84 90 9 1 4 4 3 5 0
108 Belgiumc 13 8 11 12 20 26 55 53 12 7
109 Netherlands 12 12 32 25 21 21 35 42 9 4
110 Austria 8 5 16 8 20 34 55 53 12 9
Ill tUnited Arab Emirates 99 79 1 4 0 0 0 16
112 France 8 5 21 19 26 35 45 41 10 5
113 Canada 28 18 35 20 15 38 22 24 1 1
114 Denmark 2 4 55 32 22 25 21 39 4 5
115 Germany, Fed. Rep. 7 4 5 6 46 48 42 42 5 5
116 Finland 3 5 40 14 12 28 45 53 2 4
117 Sweden 9 6 23 10 35 43 33 41 2 2
118 United States 8 6 27 17 37 47 28 31 3 2
119 Norway 21 51 28 II 17 15 34 23 2 1
120 Japan 2 1 7 1 31 65 60 33 17 2
121 Switzerland 3 4 7 4 30 33 60 60 10 5
Total reporting economies 15w 11w 27w 14w 25w 35w 34w 41w 7w
Oil exporters 67w 71w 25w 7w 3w 8w 7w 15w 1w
Nonreporting nonmembers
a. Textiles and clothing is a subgmup of other manufactures. b. Figures for the South Afncan Customs Union, comprising South Afnca, Namibia, Lesotho, Botswana,
and Swaziland, aix included in South African data. Trade among the component territones is excluded. c. Includes Luxembourg.
209
Table 17. OECD imports of manufactured goods: origin and composition
Composition of 1988 imports of manufactures by high-
Value of imports of income OECD countries (percent)
manufactures, by origin
Electrical
(millions of dollars) Textiles and machinery and Transport
1968 1988 clothing Chemicals electronics equipment Others
Low-income economies 1,281 37,243 43 w 7w 5w 2w 44w
China and India 759 t 26,942 42 w 7w 6w Ow 45 w
Other low-income 523 t 10,300 45 w 6w 1w 6w 42 w
I Mozambique 6 5 12 3 10 2 72
2 Ethiopia 6 53 10 10 1 1 78
3Chad 0 1 7 1 4 0 88
4 Tanzania 0 14 6 5 15 2 72
5 Bangladesh 0 774 84 0 0 0 16
bMalawi 0 17 90 0 1 0 9
7 Somalia 1 5 1 2 18 10 70
8 Zaire 38 382 0 3 0 0 96
9Bhutan 0 1 17 16 0 8 60
10 Lao PDR 0 2 77 5 4 1 13
11 Nepal 2 173 90 0 0 0 9
12 Madagascar 0 5 10 4 6 3 77
13 BurkinaFaso 30 36 62 2 1 1 34
14 Mali 7 33 66 13 0 0 20
15 Bunindi 4 2 6 2 7 14 71
16 Uganda 0 6 2 1 2 2 94
17 Nigeria 12 160 2 26 2 1 68
18 Zambia 9 32 22 0 1 4 73
19 Niger 0 367 0 98 0 0 2
20 Rwanda 0 1 3 17 11 0 69
21 China 206 19,874 42 8 8 0 42
22 India 553 7,069 41 4 1 0 54
23 Pakistan 161 2,159 79 0 0 0 21
24 Kenya 14 95 7 3 4 2 84
25Togo 1 13 2 1 3 1 93
26 CentralAfricanRep. 12 51 0 0 0 0 100
27 Haiti 12 398 50 3 17 0 30
28 Benin 0 15 I 1 3 47 48
29 Ghana 16 49 I 3 2 10 85
30 Lesotho5
31 SriLanka 6 824 74 1 0 0 24
32 Guinea 30 133 0 38 0 0 62
33 Yemen, PDR 2 3 1 0 53 12 34
34 Indonesia 10 3,339 33 3 1 0 62
35 Mauritania 0 3 25 5 3 6 61
36 Sudan 2 18 14 8 2 24 52
37 Afghanistan 10 50 91 1 0 1 8
38 Myanmar 3 22 37 8 0 2 52
39 Kampuchea,Dem. 0 1 19 0 4 0 77
40 Liberia 25 904 0 0 0 65 34
41 Sierra Leone 80 69 0 0 1 0 99
42 VietNam 0
Middle-income economies 3,371 t 183,671 22 w Sw 17w 7w 49w
Lower-middle-income l,080r 63,138t 22 w 6w 21w 8w 43w
43 Bolivia 1 24 23 7 0 6 65
44 Philippines 107 3,732 34 4 29 0 32
45 YemenArabRep. 0 31 1 0 2 3 94
46 Senegal 9 37 7 33 6 4 50
47 Zimbabwe 12 553 7 0 0 0 93
48 Egypt, Arab Rep. 28 448 62 5 2 1 31
49 Dominican Rep. 5 1,202 46 6 0 47
50 Cbte d'Ivoirc 5 199 24 3 1 2 70
51 PapuaNewGuinea 3 20 5 0 1 11 82
52 Morocco 18 1,468 62 19 5 2 12
53 Honduras 3 109 62 2 0 1 35
54 Guatemala 4 143 70 8 0 3 18
55 Congo, People's Rep. 8 97 0 0 0 0 100
56 ElSalvador 1 110 56 0 25 0 18
57 Thailand 25 5,892 28 2 14 1 55
58 Botswana5
59 Camemon '1 . 1 59
60 Jamaica 58 563 40 54 1 0 5
61 Ecuador 3 50 14 5 2 1 79
62 Colombia 34 779 24 4 0 0 72
63 Paraguay 4 53 18 13 0 0 69
64 Tunisia II 1,295 62 15 8 1 14
65 Turkey 23 4,356 71 5 2 1 21
66 Peru 8 346 52 8 4 0 36
67 Jordan 0 123 4 50 5 2 40
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
210
Composition 011988 imports of manufactures by high-
Value of imports of income OECD countries (percent)
manufactures, by origin
Electrical
(millions of dollars) Textiles and machinery and Transport
1968 1988' clothing Chemicals electronics equipment Others
68 Chile 9 335 13 31 I 4 51
69 Syrian Arab Rep. 23 32 8 2 3 55
70 Costa Rica 2 421 65 1 11 1 22
71 Mexico 295 18,557 4 5 35 13 43
72 Mauritius 0 635 83 1 0 0 16
73 Poland 222 2,637 22 16 6 11 45
74 Malaysia 28 6,047 16 3 58 0 23
75 Panamac 20 794 5 3 1 63 29
76 Brazil 107 11,472 8 9 6 14 63
77 Angola 2 202 0 0 0 0 100
78 Lebanon 18 185 17 2 3 1 78
79 Nicaragua 2 5 5 6 32 2 55
Upper-middle-income 2,291 t 120,533 23w 5w 15w 6w 52w
80 South Africa' 491 3,101 4 15 I 2 78
81 Algeria 13 239 1 21 I 18 60
82 Hungaly 119 2,310 24 21 9 3 42
83 Umguay 14 277 58 2 1 1 39
84 Argentina 86 1,539 11 17 I 2 68
85 Yugoslavia 270 6,699 27 8 9 11 45
86 Gabon 9 115 0 69 1 1 29
87 Venezuela 20 528 2 24 3 7 64
88 Trinidad and Tobago 37 304 74 0 0 25
89 Korea, Rep. of 247 42,367 25 2 19 8 46
90 Portugal 336 8,394 40 7 8 6 39
91 Greece 82 3,728 56 4 3 6 31
92 Oman 1 278 0 0 3 34 62
93 Libya 5 280 0 88 1 0 Il
94 Iran, Islamic Rep. 107 478 90 1 1 0 8
95 Iraq 4 133 1 9 4 13 73
96 Romania 87 2,150 30 8 3 3 56
Low- and middle-income 4,653 220,913 26 w 5w 15 w 6w 48 w
Sub-Saharan Africa 343 1 4,412 16 w 14w 1w 14 w 55 w
East Asia 925 t 127,935 24 w 3w 18 w 4w 51w
South Asia 724 r 11,037 54 w 3w 1w Ow 42 w
Europe, M.East, & N.Africa 1,374 35,947 41 w 10 w 7w 6w 36 w
Latin America & Caribbean 795 1 38,472 11w 9w 20 w 12w 49 w
Severely indebted 1,038t 44,1431 14w 8w 20w 11w 47w
High-income economies 85,975 I 1,226,247 6w 13w 11w 19 w 51w
OECD members 84,267 t 1,176,481 5w 13 w 11w 20 w 51w
tOther 1,708 t 49,766 22 w 6w 18 w 2w 51w
97 (Saudi Arabia 8 2,095 0 47 1 1 51
98 Spain 470 21,512 5 10 6 27 51
99 Ireland 363 11,618 8 24 12 2 55
100 tlsracl 261 6,589 8 14 11 2 64
101 (Singapore 36 13,858 6 5 33 2 54
102 tHong Kong 1,264 24,141 39 1 16 1 44
103 New Zealand 79 1,447 11 25 6 3 54
104 Australia 368 4,989 3 36 4 6 51
105 UnitedKingdom 7,988 81,223 6 18 10 11 55
106 Italy 5,646 87,765 17 8 7 10 58
107 (Kuwait 6 164 1 17 7 17 59
108 Belgium" 5,255 62,739 9 20 6 20 45
109 Netherlands 3,929 56,100 7 31 8 8 45
110 Austria 1,086 21,553 10 9 13 5 62
Ill tUnited Arab Emirates 0 366 27 22 5 2 44
112 France 6,125 99,708 6 18 9 23 45
113 Canada 6,262 64,535 1 8 6 41 44
114 Denmark 1,024 14,319 8 15 11 5 62
115 Germany, Fed. Rep. 16,278 227,674 5 14 10 21 49
116 Finland 748 13,767 4 8 8 5 75
117 Sweden 2,901 35,210 2 9 9 19 61
118 UnitedStates 16,484 161,794 2 12 13 20 53
119 Norway 774 7,394 2 22 7 9 60
120 Japan 5,760 164,334 1 3 19 30 46
121 Switzerland 2,724 38,624 6 21 10 2 61
Total reporting economies 90,6271 1,447,1611 9w 12w 12w 17w SOw
Oil exporters 1,3471 35,670t 8w 13w 20w lOw 49w
Nonreporting nonmembers 1,025 t 8,871 t 9w 27w Sw Sw 51w
Note: Includes only high-income OECD economies, a. All Australian data refer to 1987. b. Figures are for South Africa, Botswanu, and Lesotho. c. Excludes the
Canal Zone. d. Includeu Luxembourg.
211
Table 18. Balance of payments and reserves
Current account balance
(millions of dollars) Net Gross international reserves
After Before workers Net direct In months
official official remittances private investment of import
transfers transfers (millions of dollars) (millions of dollars) Millions of dollars coverage
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988 1988
Low-income economies 3,673 t 47,187 t 3.5 w
China and India 1,023 t 32,937 t 4.6 w
Other low-income 2,650 f 14,250 t 2.3 w
1 Mozambique _359a _733a . 0' . . .
2 Ethiopia -32 -389" -43 -510" . . . 4 . . 72 171 1.5
3 Chad 2 _l7a 33 -252" -6 0" 1 0" 2 66 1.7
4 Tanzania -36 -235 -37 -743 . . . . . . 65 78 0.6
5 Bangladesh _ll4a -289 _234a -1,112 0 737 0 829 2.9
6 Malawi -35 -53 -46 -134 -4 S 9 0 29 151 3.7
7 Somalia -6 -107" -18 -349" . . . . 5 -11 21 23 0.6
8 Zaire -64 -693 -141 -888 -98 0 42 11 189 372 1.4
9 Bhutan -68 . . -68 . . . . . . S
47
10 LaoPDR -92 -118 . . . . 6 16 0.9
11 Nepal -1" -245 -25' -251 . . 0 . . 0 94 283 4.5
12 Madagascar 10 _l27a -42 -261' -26 . . 10 0" 37 224 3.7
13 BurkinaFsso 9 66a -21 -310' 16 215' 0 0" 36 325 4.6
14 Mali -2 -104 -22 -350 -1 49 -1 1 1 44 0.7
15 Burendi 2" -62 -2" -163 . . 0" 1 15 76 2.9
16 Uganda 20 -163" 19 _289a 5 . . 4 1 57 49 0.8
17 Nigeria -368 -1,024 -412 -1,045 . . -34 205 836 223 933 1.3
18 Zambia 108 -174 107 -234 -48 -21 -297 0 515 139 1.2
19 Niger 0 -94 -32 -248 -3 -45 0 .. 19 237 4.7
20 Rwanda 7 -119 -12 -258 -4 -17 0 21 8 118 3.2
21 China* -81" -3,760 -81" -3,802 0" 129 . . 2,344 . . 23,751 5.0
22 India -380" -6,870" -590" -7,220" 80" 2,850" 0" 280" 1,023 9,186 3.8
23 Pakistan -667 -1,164 -705 -1,685 86 2,018 23 145 195 1,193 1.5
24 Kenya -49 -454 -86 -711 . . -3 14 7 220 296 1.3
25 logo 3 -50 -14 -122 -3 13 0 12 35 237 4.5
26 CentralAfricanRep. -12 -21" -24 -181" -4 29" 1 0" 1 113 3.9
27 Haiti II -53 4 -183 13 64 3 10 4 20 0.5
28 Benin -3 -105" -23 _l77a 0 87" 7 1" 16 9 0.2
29 Ghana -68 -59" -76 _232a -9 -2 68 5" 43 310 2.7
30 Lesotho 18" -73 -1" -130 . . . . 21 . . 56 1.2
31 SriLanka -59 -404 -71 -611 3 357 0 43 43 248 1.1
32 Guinea . . -209" . . -279" . . . . .
7a
33 Yemen, PDR -4 -383 -4 -436 52 253 -1 . . 59 97 1.3
34 Indonesia -310 -1,189 -376 -1,500 . . 99 83 542 160 6,322 3.3
35 Mauritania -5 -8 -13 -179 -6 -26 1 2 3 77 1.4
36 Sudan -42 775a 43 ._l,l44a . 300" -1 0" 22 107 0.6
37 Afghanistan . . 26 . . -243 . . . . . . 49 657 9.1
38 Myanmar -63 _208a -81 -307" . . . . . . 98 180 2.7
39 Kainpuchea, Dem. . . . . . . . . . . . . . . S
40 Liberia -16" -118 -27" -163 _18a -51 28" 39 . . 0 0.0
41 Sierra Leone -16 -74" -20 -86" . . 0 8 39 39 7 0.4
42 Vier Nam . . -1,086 -1,099 . . . . . . 243
Middle-income economies 16,396 t 120,433 t 2.9 w
Lower-middle-income 7,173 t 56,716 t 2.4 w
43 Bolivia 4 -306 2 -429 . . 1 -76 30 46 473 5.0
44 Philippines -48 -406 -138 -694 388 -29 986 255 2,169 2.2
45 YemenAtabRep. 34a -695" _52a -785" 39" 190" . . 0" . . 285 1.6
46 Senegal -16 -262 -66 -467 -16 78 5 -73 22 22 0.2
47 Zimbabwe -14" 9" _26a -56" . . . 5 4 59 341 2.2
48 Egypt, Arab Rep. -148 _l,868a -452 -2,848' 29 3,386" 973a 165 2,261 1.8
49 DominicanRep. -102 -128 -103 -218 25 328 72 106 32 261 1.4
50 Cbted'Ivoire -38 -1,278 -73 -1,335 -56 -480 31 . . 119 29 0.1
51 PapuaNewGuinea -89" -163 -239" -380 . . 42 . . 89 . . 419 2.6
52 Momcco -124 467 -161 164 27 1,289 20 85 142 836 1.5
53 Honduras -64 -314 -68 -431 . . 8 47 20 57 0.5
54 Guatemala -8 -402 -8 -506 . . 0 29 96 79 416 2.6
55 Congo, People's Rep. -45" 67 .53a 7 -3" -46 30" 43 9 8 0.1
56 El Salvador 9 14" 7 _242a . 4 55 64 354 3.0
57 Thailand -250 -1,671 -296 -1,859 . . 5 . 43 1,093 911 7,112 3.6
58 Botswana -30" 491 -35" 309 . . . . 6" 40 . . 2,258 17.7
59 Cameroon -30 _881a -47 -881" -11 3" 16 34" 81 163 0.7
60 Jamaica -153 74 -149 4 29 65 161 -16 139 147 0.8
61 Ecuador -113 -597 -122 -657 . . .. 89 80 76 568 2.1
62 Colombia -293 -355" -333 -355" 6 384" 39 186" 207 3,700 5.4
63 Paraguay -16 -150 -19 -173 . . 5 . 4 11 18 338 3.0
64 Tunisia -53 212 -88 93 20 539 16 59 60 976 2.5
65 Turkey -44 1,500 -57 1,139 273 1,755 58 352 440 3,912 2.5
66 Pent 202 -1,128 146 -1,285 .. -70 44 339 1,213 2.9
67 Jordan -20 285" -130 -281" . . 813" . . 0" 258 414 1.2
* DataforTaiwan,China,ain: 1" 10,177" 2" 10,174" 61" -3,161" 627 79,430 16.0
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified
212
Current account balance
(millions of dollars) Net Gross international reserves
After Before workers' Net direct In months
official official remittances private investment of import
transfers transfers (millions of dollars) (millions of dollars) Millions of dollars coverage
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988 1988
68 Chile -91 -168 -95 -282 -79 109 392 3,788 5.2
69 Syrian Arab Rep. -69 -76 -72 -604 210 0 57 342 1.6
70 Costa Rica -74 -143 -77 -356 0 26 76 16 677 3.9
71 Mexico -1,068 -2,905 -1,098 -3,068 264 323 2,594 756 6,327 2.1
72 Mauritius -45 -65 0 2 31 46 463 3.6
73 Poland -107 -107 0 -. -7 2,249 1.5
74 Malaysia 1,802 2 1,618 0 94 649 667 7,491 4.0
75 Panama -64 737 -79 625 - 33 -36 16 72 0.2
76 Brazil -837 4,448a -861 4,448n 0 407 2,68P 1,190 1,118 0.4
77 Angola 367 367 360
78 Lebanon 405 4,761
79 Nicaragua -40 15 49
Upper-middle-income 9,223 t 63,717 t 3.7 w
80 South Africa -1,215 1,292 -1,253 1,207 . - 318 4 1,057 2,204 1.1
81 Algeria -125 -2,040 -163 -2,040 178 279 45 -48 352 3,191 4.0
82 Hungary -61 -389' -61 -389 697 2,521 2.3
83 Uruguay -45 34 -55 13 - -2 186 1,602 10.3
84 Argentina -163 -t,6l5 -160 -1,615 - - 11 1,147 682 5,157 4.7
85 Yugoslavia - 37 2,487 -378 2,487 441 4,893 . - 0 143 3,074 2.2
86 Gabon -616 -15 -627 -8 -151 -1 121 15 71 0.4
87 Venezuela -104 -4,692 -98 -4,661 -87 -203 -23 89 1,047 7,793 5.4
88 Trinidad and Tobago -109 -184 -104 -184 0 83 26 43 148 1.0
89 Korea, Rep. of -623 14,161 -706 14,117 - 66 720 610 12,478 2.6
90 Portugal _l58a -601 - l58 -1,320 5Ø4 3,381 1
a
820 1,565 11,719 7.1
91 Greece -422 -958 -424 -2,894 333 1,675 50 907 318 5,012 3.8
92 Oman 851 844 . - -681 33 13 1,173 6.6
93 Libya 645 -2,259 758 -2,222 -134 -496 139 43 1,596 5,798 9.0
94 Iran, Islamic Rep. -507 -511 25 217
95 Iraq 105 104 24 472
96 Romania -23 -23
Low- and middle-income 20,069 t 167,620 3.1 w
Sub-Saharan Africa 2,028 r 7,861 1.9w
East Asia 2,885 t 60,151 3.6w
South Asia 1,453 r 11,987 3.1 w
Europe, M.East, & N.Africa 7,165 r 50,836 3.4w
Latin America & Caribbean 5,481 t 34,553 2.8 w
Severely indebted 6,042 t 36,607 t 2.5 w
High-income economies 75,667 t 842,605 t 3.3w
OECD members 72,938 t 787,004 3.3 w
fOther 2,729 t 55,601 t 3.7 w
97 tSaudi Arabia 71 -9,583 152 -6,283 -183 -4,935 20 -1,175 670 22,438 7.9
98 Spain 79 -3,730 79 -5,220 469 1,413 179 5,788 1,851 42,835 6.9
99 Ireland -198 625 -228 -1,034 32 91 698 5,234 2.8
100 tlsrael -562 -678 -766 -4,097 .10 183 452 4,433 2.6
101 tSingapore -572 1,660 -585 1,683 . - 93 1,066 1,012 17,073 4.2
102 tHong Kong 225 1,199 225 1,199 282
103 New Zealand -232 -780 -222 -704 3i2 137 119 258 2,845 2.7
104 Australia -777 -11,256 -682 -11,100 - 778 -460 1,709 16,853 3.6
105 United Kingdom 1,985 -26,590 2,393 -20,763 - .. -190 -13,078 2,918 51,899 2.0
106 Italy 800 -5,363 1,096 -2,614 446 1,229 498 1,337 5,547 62,067 4.3
107 tKuwait 853 4,713 853a 4,853 -1,179 -262 209 2,965 3.6
108 Belgium" 717 3,334 904 5,085 38 -30 140 1,365 2,963 23,321 2.1
109 Netherlands -588 5,282 -617 5,785 -49 -204 -15 2 3,362 34,102 3.2
110 Austria -75 -642 -73 -569 -7 284 104 294 1,806 16,043 3.7
111 tUnited Arab Emirates 90 2,700 I00 2,800 4,769 5.8
112 France -204 -3,522 18 875 -641 -1,950 248 -5,986 5,199 58,944 2.9
113 Canada 1,008 -8,258 960 -7,905 508 -3,306 4,733 22,422 1.8
114 Denmark -544 -1,819 -510 -1,686 . - 75 488 11,433 3.3
115 Germany, Fed. Rep. 852 48,499 1,899 60,320 -1,366 -4,188 -303 -8,722 13,879 97,576 3.6
116 Finland -239 -3,006 -232 -2,578 - - . -41 -1,752 455 7,171 2.8
117 Sweden -265 -2,567 -160 -1,424 -10 -104 -4,406 775 10,982 2.1
118 United States 2,330 -126,620 4,680 -113,740 -650 -820 -6,130 40,920 15,237 144,177 2.7
119 Norway -242 -3,671 -200 -2,858 -50 32 -23 813 13,753 4.3
120 Japan 1,980 79,590 2,160 82,610 -260 -34,710 4,876 106,668 4.4
121 Switzerland 72 8,326 114 8,311 -313 -1,549 -6,913 5,317 58,367 7.9
Total reporting economies 95,736 t 1,010,225 t 3.3 w
Oil exporters 7,082 t 81,171 t 3.4 w
Nonreporting nonmembers
a. World Bank estimate. b. Includes Luxembourg.
213
Table 19. Official development assistance from OECD & OPEC members
Amount
1965 1970 1975 1980 1984 1985 1986 1987 1988 1989'
OECD Millions of U.S. dollars
99 Ireland 0 0 8 30 35 39 62 51 57 47
103 New Zealand . . 14 66 72 55 54 75 87 104 87
104 Australia 119 212 552 667 777 749 752 627 1,101 1,017
105 United Kingdom 472 500 904 1,854 1,429 1,530 1,737 1,871 2,645 2,588
106 Italy 60 147 182 683 1,133 1,098 2,404 2,615 3,193
108 Belgium 102 120 378 595 446 440 547 687 597
109 Netherlands 70 196 608 1,630 1,268 1,136 1,740 2,094 2,231 2,089
110 Austria 10 11 79 178 181 248 198 201 301 279
112 France 752 971 2,093 4,162 3,788 3,995 5,105 6,525 6,865
113 Canada 96 337 880 1,075 1,625 1,631 1,695 1,885 2,347
114 Denmark 13 59 205 481 449 440 695 859 922 926
115 Germany, Fed. Rep. 456 599 1,689 3,567 2,782 2,942 3,832 4,391 4,731 4,953
116 Finland 2 7 48 110 178 211 313 433 608 705
117 Sweden 38 117 566 962 741 840 1,090 1,375 1,590 1,813
118 UnitedStates 4,023 3,153 4,161 7,138 8,711 9,403 9,564 9,115 10,141
119 Norway II 37 184 486 540 574 798 890 985 919
120 Japan 244 458 1,148 3,353 4,319 3,797 5,634 7,342 9,134
121 Switzerland 12 30 104 253 285 302 422 547 617 559
Total 6,480 6,968 13,855 27,296 28,742 29,429 36,663 41,595 48,167
OECD As a percentage of donor GNP
99 Ireland 0.00 0.00 0.09 0.16 0.22 0.24 0.28 0.19 0.20 0.17
103 New Zealand 0.23 0.52 0.33 0.25 0.25 0.30 0.26 0.27 0.22
104 Australia 0.53 0.59 0.65 0.48 0.45 0.48 0.47 0.34 0.46 0.37
105 United Kingdom 0.47 0.41 0.39 0.35 0.33 0.33 0.31 0.28 0.32 0.31
106 Italy 0.10 0.16 0.11 0.15 0.28 0.26 0.40 0.35 0.39
108 Belgium 0.60 0.46 0.59 0.50 0.58 0.55 0.48 0.48 0.40
109 Netherlands 0.36 0.61 0.75 0.97 1.02 0.91 1.01 0.98 0.98 0.93
110 Austria 0.11 0.07 0.21 0.23 0.28 0.38 0.21 0.17 0.24 0.22
112 France 0.76 0.66 0.62 0.63 0.77 0.78 0.70 0.74 0.72
113 Canada 0.19 0.41 0.54 0.43 0.50 0.49 0.48 0.47 0.50
114 Denmark 0.13 0.38 0.58 0.74 0.85 0.80 0.89 0.88 0.89 0.93
115 Germany, Fed. Rep. 0.40 0.32 0.40 0.44 0.45 0.47 0.43 0.39 0.39 0.41
116 Finland 0.02 0.06 0.18 0.22 0.35 0.40 0.45 0.49 0.59 0.63
117 Sweden 0.19 0.38 0.82 0.78 0.80 0.86 0.85 0.88 0.89 0.98
118 United States 0.58 0.32 0.27 0.27 0.24 0.24 0.23 0.20 0.21
119 Norway 0.16 0.32 0.66 0.87 1.03 1.01 1.17 1.09 1.10 1.02
120 Japan 0.27 0.23 0.23 0.32 0.34 0.29 0.29 0.31 0.32
121 Switzerland 0.09 0.15 0.19 0.24 0.30 0.31 0.30 0.31 0.32 0.30
OECD National currencies
99 Ireland(millionsofpounds) 0 0 4 15 32 37 46 34 37 35
103 New Zealand (millions of dollars) . . 13 55 74 95 109 143 146 158 145
104 Australia (millions of dollars) 106 189 402 591 873 966 1,121 895 1,404 1,283
105 UnitedKingdom(millionsofpounds) 169 208 409 798 1,070 1,180 1,194 1,142 1,485 1,578
106 Italy(billionsoflirc) 38 92 119 585 1,991 2,097 3,578 3,389 4,156
108 Belgium (millions of francs) 5,100 6,000 13,902 17,399 25,527 26,145 24,525 25,648 21,949
109 Netherlands (millions of guilders) 253 710 1,538 3,241 4,069 3,773 4,263 4,242 4,400 4,430
110 Austria (millions ofschillings) 260 286 1,376 2,303 3,622 5,132 3,023 2,541 3,717 3,691
112 France (millions of francs) 3,713 5,393 8,971 17,589 33,107 35,894 35,357 39,218 40,814
113 Canada (millions of dollars) 104 353 895 1,257 2,104 2,227 2,354 2,500 2,888
114 Denmark(millionsofkmner) 90 443 1,178 2,711 4,650 4,657 5,623 5,848 6,196 6,769
115 Germany, Fed. Rep. (millions
ofdeutschemarks) 1,824 2,192 4,155 6,484 7,917 8,661 8,323 8,004 8,292 9,318
116 Finland(millionsofmarkkaa) 6 29 177 414 1,070 1,308 1,587 1,902 2,550 3,025
117 Sweden(millionsofkronor) 197 605 2,350 4,069 6,129 7,226 7,765 8,718 9,742 11,688
118 United States (millions ofdollars) 4,023 3,153 4,161 7,138 8,711 9,403 9,564 9,115 10,141
119 Norway (millions of kmner) 79 264 962 2,400 4,407 4,946 5,901 5,998 6,412 6,345
120 Japan(billionsofyen) 88 165 341 760 1,026 749 950 1,062 1,169
121 Switzerland(millionsoffranes) 52 131 268 424 672 743 759 815 900 914
OECD Summary
ODA (billions of U.S. dollars, nominalprices) 6.48 6.97 13.86 27.30 28.74 29.43 36.66 41.59 48.17 .
ODAasapercentageofGNP 0.48 0.34 0.35 0.37 0.36 0.35 0.35 0.35 0.36
ODA (billions of U.S. dollars,
constant 1980 prices) 20.68 18.83 21.85 27.30 28.74 29.14 30.55 30.81 33.22 .
GNP (trillions of U.S. dollars, nominal prices) 1.35 2.04 3.96 7.39 8.03 8.49 10.39 12.05 13.48 .
GDPdeflator5 0.31 0.37 0.63 1.00 1.00 1.01 1.20 1.35 1.45 1.44
214
Anwunt
1976 1979 1980 1982 1983 1984 1985 1986 1987 1988
OPEC Millions of U.S. dollars
17 Nigeria 80 29 35 58 35 51 45 52 30 14
81 Algeria 11 281 81 129 37 52 54 114 39 13
87 Venezuela 109 110 135 125 142 90 32 85 24 49
93 Libya 98 145 376 44 144 24 57 68 63 129
94 Iran, Islamic Rep. 751 -20 -72 -193 10 52 -72 69 -10 39
95 Iraq 123 658 864 52 -10 -22 -32 -21 -37 -28
97 SaudiArabia 2,791 3,941 5,682 3,854 3,259 3,194 2,630 3,517 2,888 2,098
107 Kuwait 706 971 1,140 1,161 997 1,020 771 715 316 108
III UnitedArabEmirates 1,028 968 1,118 406 351 88 122 87 15 -17
Qatar 180 282 277 139 20 10 8 18 0 4
Total OAPEC 4,937 7,246 9,538 5,785 4,798 4,366 3,610 4,498 3,284 2,307
Total OPEC 5,877 7,365 9,636 5,775 4,985 4,559 3,615 4,704 3,328 2,409
OPEC As a percentage of donor GNP
17 Nigeria 0.19 0.04 0.04 0.08 0.04 0.06 0.06 0.11 0.13 0.03
81 Algeria 0.07 0.90 0.20 0.31 0.08 0.10 0.10 0.19 0.06 0.02
87 Venezuela 0.35 0.23 0.23 0.19 0.22 0.16 0.00 0.08 0.02 0.04
93 Libya 0.66 0.60 1.16 0.15 0.51 0.10 0.24 0.34 0.25 0.52
94 Iran, Islamic Rep. 1.16 -0.02 -0.08 -0.15 0.01 0.03 -0.04 0.04 -0.01 0.02
95 Iraq 0.76 1.97 2.36 0.13 -0.02 -0.05 -0.07 -0.05 -0.08 -0.05
97 SaudiArabia 5.95 5.16 4.87 2.50 2.69 3.20 2.98 4.67 3.88 2.70
107 Kuwait 4.82 3.52 3.52 4.34 3.83 3.95 3.17 2.91 1.23 0.41
Ill UnitedArabEmirates 8.95 5.08 4.21 1.39 1.26 0.32 0.45 0.41 0.07 -0.07
Qatar 7.35 6.07 4.16 2.22 0.40 0.18 0.15 0.47 0.00 0.08
Total OAPEC 4.23 3.31 3.22 1.81 1.70 1.60 1.39 1.80 1.10 0.86
Total OPEC 2.32 1.75 1.85 0.96 0.82 0.76 0.61 0.95 0.63 0.45
Net bilateral flows to low-income economies
1965 1970 1975 1980 1983 1984 1985 1986 1987 1988
OECD As a percentage of donor GNP
99 Ireland 0.03 0.03 0.05 0.06 0.07 -0.07
103 New Zealand 0.14 0.01 0.00 0.00 0.00 0.00 0.06 0.03
104 Australia 0.08 0.09 0.10 0.04 0.05 0.06 0.05 0.04 0.05 0.11
105 United Kingdom 0.23 0.15 0.11 0.11 0.10 0.09 0.09 0.09 0.09 0.10
106 Italy 0.04 0.06 0.01 0.01 0.05 0.09 0.12 0.16 0.16 0.16
108 Belgium 0.56 0.30 0.31 0.24 0.21 0.20 0.23 0.20 0.14 0.10
109 Netherlands 0.08 0.24 0.24 0.30 0.26 0.29 0.27 0.32 0.31 0.31
110 Austria 0.06 0.05 0.02 0.03 0.02 0.01 0.02 0.01 0.04 -0.04
112 France 0.12 0.09 0.10 0.08 0.09 0.14 0.14 0.13 0.14 0.14
113 Canada 0.10 0.22 0.24 0.11 0.13 0.15 0.15 0.12 0.14 0.14
114 Denmark 0.02 0.10 0.20 0.28 0.31 0.28 0.32 0.32 0.32 0.36
115 Germany, Fed. Rep. 0.14 0.10 0.12 0.08 0.13 0.11 0.14 0.12 0.11 0.11
116 Finland 0.06 0.08 0.12 0.13 0.17 0.18 0.18 0.23
117 Sweden 0.07 0.12 0.41 0.36 0.33 0.30 0.31 0.38 0.29 0.31
118 United States 0.26 0.14 0.08 0.03 0.03 0.03 0.04 0.03 0.03 0.04
119 Norway 0.04 0.12 0.25 0.31 0.39 0.34 0.40 0.47 0.38 0.42
120 Japan 0.13 0.11 0.08 0.08 0.09 0.07 0.09 0.10 0.07 0.07
121 Switzerland 0.02 0.05 0.10 0.08 0.10 0.12 0.12 0.12 0.10 0.10
Total 0.20 0.13 0.11 0.07 0.08 0.07 0.09 0.09 0.09 0.09
a. Praliminary estimates. b. See the technical notes.
215
Table 20. Official development assistance: receipts
Net disbursement of ODA from all sources
Per capita As a percentage
Millions of dollars
(doltars) of GNP
1982 1983 1984 1985 1986 1987 1988 1988 1988
Low-income economies 12,969 t 12,443 t 12,500t 13,9461 16,853 1 18,591 t 21,912 t 7.6w 2.4w
China and India 2,168 1 2,509 t 2,471 t 2,532 1 3,254 t 3,3001 4,087 1 2.1 w 0.6 w
Other low-income 10,801 r 9,9341 10,0291 11,4141 13,599 t 15,291 t 17,825 t 18.2w 6.6w
I Mozambique 208 211 259 300 422 651 886 59.3 70.6
2 Ethiopia 200 339 364 715 636 634 970 20.5 17.4
3 Chad 65 95 115 182 165 198 264 48.9 28.8
4 Tanzania 684 594 558 487 681 882 978 39.6 31.2
5 Bangladesh 1,341 1,049 1,200 1,152 1,455 1,635 1,592 14.6 8.2
6 Malawi 121 117 158 113 198 280 366 46.0 30.6
7 Somalia 462 343 350 353 511 580 433 73.4 42.9
8 Zaire 348 315 312 325 448 627 580 17.4 9.0
9 Bhutan 11 13 18 24 40 42 42 30.3 14.0
10 Lao PDR 38 30 34 37 48 58 77 19.6 14.4
11 Nepal 200 201 198 236 301 347 399 22.2 13.0
12 Madagascar 242 183 153 188 316 321 305 28.0 16.2
13 BurkinaFaso 213 184 189 198 284 281 298 34.9 16.0
14 Mali 210 215 321 380 372 366 427 53.5 22.0
15 Bumndi 127 140 141 142 187 202 187 36.5 17.1
16 Uganda 133 137 163 182 198 279 359 22.1 8.4
17 Nigeria 37 48 33 32 59 69 120 1.1 0.4
18 Zambia 317 217 239 328 464 430 478 63.3 12.0
19 Niger 257 175 161 304 307 353 371 51.1 15.5
20 Rwanda 151 150 165 181 211 245 253 38.0 11.0
21 China 524 669 798 940 1,134 1,462 1,990 1.8 0.5
22 India 1,644 1,840 1,673 1,592 2,120 1,839 2,098 2.6 0.8
23 Pakistan 916 735 749 801 970 879 1,408 13.3 3.7
24 Kenya 485 400 411 438 455 572 808 36.0 9.4
25 Togo 77 112 110 114 174 126 199 58.9 14.7
26 CentralAfrican Rep. 90 93 114 104 139 176 196 68.4 17.5
27 Haiti 128 134 135 153 175 218 147 23.5 5.9
28 Benin 81 86 77 95 138 138 162 36.3 9.0
29 Ghana 141 110 216 203 371 373 474 33.9 9.1
30 Lesotho 93 108 101 94 88 107 108 64.4 26.3
31 SriLanka 416 473 466 484 570 502 599 36.1 8.5
32 Guinea 90 68 123 119 175 213 262 48.5 10.3
33 Yemen,PDR 143 106 103 113 71 74 76 32.3 7.2
34 Indonesia 906 744 673 603 711 1,246 1,632 9.3 2.1
35 Mauritania 187 176 175 209 225 182 184 96.6 18.4
36 Sudan 740 962 622 1,128 945 898 918 38.6 7.8
37 Afghanistan 9 14 7 17 2 45 72
38 Myanmar 319 302 275 356 416 367 451 11.3
39 Kwnpuchea,Dem. 44 37 17 13 13 14 18
40 Liberia 109 118 133 90 97 78 65 26.8
41 Sierra Leone 82 66 61 66 87 68 102 26.0
42 VietNam 136 106 109 114 147 III 148 2.3
Middle-income economies 10,621 t 9,998 10,352 t 10,538 t 11,741 t 12,851 1 11,931 t 12.4w 0.7 w
Lower-middle-income 9,969 t 9,473 t 9,869 t 9,947 t 10,981 t 12,000 1 11,089 t 15.8 w 1.0w
43 Bolivia 148 174 172 202 322 318 392 56.7 9.1
44 Philippines 333 429 397 486 956 770 854 14.3 2.2
45 YemenArabRep. 412 328 326 283 257 348 223 26.3 3.8
46 Senegal 285 323 368 295 567 641 568 81.2 11.4
47 Zimbabwe 216 208 298 237 225 294 273 29.3 4.3
48 Egypt, Arab Rep. 1,441 1,463 1,794 1,791 1,716 1,773 1,537 30.6 4.3
49 Dominican Rep. 136 100 188 207 93 130 118 17.1 2.5
50 Côted'Ivoire 137 156 128 125 186 254 439 39.1 4.5
51 PapuaNewGuinea 311 333 322 259 263 322 379 101.9 10.8
52 Momcco 774 398 352 785 403 447 482 20.1 2.2
53 Honduras 158 190 286 272 283 258 321 66.4 7.3
54 Guatemala 64 76 65 83 135 241 235 27.0 2.9
55 Congo,People'sRep. 93 108 98 71 110 152 89 41.5 4.1
56 ElSalvador 218 290 261 345 341 426 420 83.4 7.7
57 Thailand 389 431 475 481 496 504 563 10.3 1.0
58 Botswana 101 104 102 96 102 156 151 127.7 7.8
59 Camemon 212 129 186 159 224 213 284 25.4 2.2
60 Jamaica 180 181 170 169 178 168 193 80.3 6.0
61 Ecuador 53 64 136 136 147 203 137 13.6 1.3
62 Colombia 97 86 88 62 63 78 61 1.9 0.2
63 Paraguay 85 51 50 50 66 81 76 18.7 1.3
64 Tunisia 210 205 178 163 222 274 316 40.5 3.2
65 Turkey 647 356 242 179 339 376 307 5.7 0.4
66 Pem 188 297 310 316 272 292 272 13.2 1.1
67 JoMan 798 787 687 538 564 579 425 108.8 9.3
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
216
Net disbursement of ODA from all sources
Per capita Ass percentage
Millions of dollars (dollars) of GNP
1982 1983 1984 1985 1986 1987 1988 1988 1988
68 Chile - -8 0 2 40 -5 21 44 3.4 0.2
69 SyrianArabRep. 962 813 641 610 728 684 191 16.4 1.3
70 CostaRica 80 252 218 280 196 228 187 69.9 4.0
71 Mexico 140 132 83 144 252 155 173 2.1 0.1
72 Mauritius 48 41 36 28 56 65 59 56.4 3.0
73 Poland . . . . . . . . . . . . . . .
74 Malaysia 135 177 327 229 192 363 104 6.1 0.3
75 Panama 41 47 72 69 52 40 22 9.3
76 Brazil 208 101 161 123 178 289 210 1.5 0.1
77 Angola 60 75 95 92 131 135 159 16.8
78 Lebanon 187 127 77 83 62 101 141
79 Nicaragua 121 120 114 102 150 141 213 58.8
Upper-middle-income 652 525 483 t 591 760 I 852 I 841 I 3.3 w 0.1 w
80 South Africa . . . . . . . . . . . . . . .
81 Algeria 136 95 122 173 165 214 171 7.2 0.3
82 Hungaly . . . . . . . . . . . . . . .
83 Umguay 4 3 4 5 27 18 41 13.4 0.5
84 Argentina 30 48 49 39 88 99 152 4.8 0.2
85 Yugoslavia -8 3 3 11 19 35 44 1.9 0.1
86 Gabon 62 64 76 61 79 82 106 98.3 3.2
87 Venezuela 13 10 14 11 16 19 18 0.9 0.0
88 TrinidadandTobago 6 5 5 7 19 34 8 6.8 0.2
89 Korea,Rep.of 34 8 -37 -9 -18 II 10 0.2 0.0
90 Portugal 49 43 97 101 139 64 102 9.9 0.2
91 Greece 12 13 13 11 19 35 35 3.5 0.1
92 Oman 133 71 67 78 84 16 1 0.4
93 Libya 12 6 5 5 II 6 6 1.3
94 Iran,IslamicRep. 3 48 13 16 27 71 82 1.7
95 Iraq 6 13 4 26 33 91 tO 0.6
96 Romania . . .
Low- and middle-income 23,589 1 22,442 22,852 24,484 28,594 1 31,4421 33,8421 8.8 w 1.3 w
Sub-Saharan Africa 7,496 I 7,305 7,595 1 8,616 10,572 I 11,902 13,416 I 28.9 w 8.8 w
East Asia 2,989t 3,121 3,243 1 3,2621 4,0801 5,071 t 5,985 1 3.9 w 0.7 w
South Asia 4,852 4,623 I 4,585 I 4,655 I 5,8881 5,6301 6,616 I 6.0w 1.9w
Europe, M.East, & N.Africa 5,9531 4,930 I 4,738 1 4,999 I 4,867 1 5,233 1 4,217 13.3 w 1.1 w
Latin America & Caribbean 2,295 2,4601 2,6891 2,9491 3,181 1 3,601 1 3,600 8.7 w 0.4 w
Severely indebted 2,7571 2,803 t 2,891 3,430 t 4,1451 4,3051 4,5901 10.3 w 0.5 w
High-income economies
OECD members .. .. .. .. .. ..
tOther 1,477 I 2,084 I 1,948 1 2,553 1 2,711 t 2,1241 1,993 1 62.5 w 0.5 w
97 tSaudi Arabia 57 44 36 29 31 22 19 1.5 0.0
98 Spain
99 Ireland
100 (Israel 857 1,345 1,256 1,978 1,937 1,251 1,241 279.3 2.8
101 (Singapore 20 15 41 24 29 23 22 8.2 0.1
102 tHong Kong 8 9 14 20 18 19 22 3.9 0.0
103 New Zealand
104 Australia
105 United Kingdom
106 Italy
107 tKuwait 6 5 4 4 5 3 6 3.0 0.0
108 Belgium
109 Netherlands
110 Austria
Ill tUnitedAmbEmirates 5 4 3 4 34 115 -12 -8.0 -0.1
112 France
113 Canada
114 Denmark
115 Germany, Fed. Rep.
116 Finland
117 Sweden
118 United States
119 Norway
120 Japan
121 Switzerland
Total reporting economies 25,067 1 24,525 I 24,8001 27,037 t 31,3041 33,5661 35,8361 9.2 w 1.2 w
Oil exporters 4,3761 4,084 1 4,192 1 4,0361 4,556 t 5,176 t 4,580 1 7.8 w 0.7 w
Nonreporting nonmembers 161 131 12 1 241 33 I 46 t 42t 1.3 w
217
Table 21. Total external debt
Long-term debt
(millions of dollars)
Public and Private Use of IMF credit Short-term debt Total external debt
publicly guaranteed nonguaranteed (millions of dollars) (millions of dollars) (millions of dollars)
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
Low-income economies
China and India
Other low-income
1 Mozambique 3,801 0 238 0 41 326 4,406
2 Ethiopia 169 2,790 0 0 0 55 133 2,978
3 Chad 33 300 0 0 3 17 29 346
4 Tanzania 250 4,091 15 9 0 69 560 4,729
5 Bangladesh 0 9,330 0 0 0 840 50 10,219
6 Malawi 122 1,190 0 3 0 106 51 1,349
7 Somalia 77 1,754 0 0 0 165 . 116 . 2,035
8 Zaire 311 7,013 0 0 0 786 . 675 . . 8,475
9 Bhutan . . 68 0 0 0 0 . 0 . . 68
10 LaoPDR 8 816 0 0 0 3 6 824
11 Nepal 3 1,088 0 0 0 53 . . 23 1,164
12 Madagascar 89 3,317 0 0 0 190 95 3,602
13 BurkinaFaso 21 805 0 0 0 3 59 866
14 Mali 238 1,928 0 0 9 74 . 65 . . 2,067
15 Burundi 7 749 0 0 8 33 . . 12 . . 794
16 Uganda 138 1,438 0 0 0 252 . . 235 1,925
17 Nigeria 452 28,630 115 337 0 0 1,752 30,718
18 Zambia 624 4,194 30 0 0 940 . 1,364 6,498
19 Niger 32 1,286 0 256 0 95 105 1,742
20 Rwanda 2 585 0 0 3 4 . . 44 632
21 China . . 32,196 0 0 0 1,013 8,806 42,015
22 India 7,838 49,695 100 1,473 0 2,573 3,772 57,513
23 Pakistan 3,064 13,944 5 84 45 557 2,425 17,010
24 Kenya 319 4,241 88 627 0 455 564 5,888
25 Togo 40 1,067 0 0 0 78 66 1,210
26 Central African Rep. 24 584 0 0 0 50 39 673
27 Haiti 40 683 0 0 3 47 92 823
28 Benin 41 904 0 0 0 4 147 . . 1,055
29 Ghana 488 2,238 10 32 46 762 . . 67 . 3,099
30 Lesotho 8 270 0 0 0 5 . . 6 . . 281
31 SriLanka 317 4,139 0 113 79 359 577 5,189
32 Guinea 312 2,312 0 0 3 61 190 2,563
33 Yemen,PDR 1 1,970 0 0 0 6 118 . 2,093
34 Indonesia 2,453 41,258 461 4,397 139 623 . . 6,322 52,600
35 Mauritania 27 1,823 0 0 0 71 . . 183 2,076
36 Sudan 298 8,044 0 374 31 905 2,530 11,853
37 Afghanistan . . . . . . . . . . . . .
38 Myanmar 106 4,217 0 0 17 8 96 . 4,321
39 Kampuchea,Dem. . . . . . . . . . . . . .
40 Liberia 158 1,101 0 0 4 309 . . 222 1,632
41 Sierra Leone 59 510 0 0 0 109 . . 108 727
42 VietNam . . . . . . . . . . . . .
Middle-income economies
Lower-middle-income
43 Bolivia 480 4,451 II 200 6 197 . 607 5,456
44 Philippines 625 23,475 919 992 69 1,093 3,888 . . 29,448
45 Yemen Arab Rep. . . 2,378 0 0 0 0 570 2,948
46 Senegal 100 2,985 31 34 0 318 . . 280 3,617
47 Zimbabwe 229 2,231 0 50 0 70 . . 308 . . 2,659
48 Egypt,ArabRep. 1,714 42,128 0 1,131 49 190 . . 6,522 . . 49,970
49 Dominican Rep. 212 3,216 141 118 7 218 . . 372 . 3,923
50 Côted'Ivoire 256 8,088 Il 3,700 0 509 1,828 14,125
51 PapuaNewGuinea 36 1,269 173 860 0 6 135 2,270
52 Morocco 712 18,567 15 200 28 956 200 19,923
53 Honduras 90 2,739 19 98 0 37 444 . 3,318
54 Guatemala 106 2,131 14 113 0 88 301 2,633
55 Congo, People's Rep. 124 4,098 0 0 0 15 650 4,763
56 ElSalvador 88 1,630 88 55 7 11 110 1,806
57 Thailand 324 13,375 402 3,530 0 662 2,964 20,530
58 Botswana 17 494 0 0 0 0 5 499
59 Cameroon 131 2,939 9 427 0 100 763 4,229
60 Jamaica 160 3,512 822 43 0 483 267 , . 4,304
61 Ecuador 193 9,353 49 25 14 405 1,082 10,864
62 Colombia 1,297 13,853 283 1,538 55 0 , , 1,609 . 17,001
63 Paraguay 112 2,091 0 28 0 0 375 , 2,493
64 Tunisia 541 5,886 . . 235 13 277 275 6,672
65 Turkey 1,844 31,054 42 535 74 299 7,704 39,592
66 Peru 856 12,475 1,799 1,423 10 801 3,880 18,579
67 Jordan 119 3,955 0 0 0 48 1,529 5,532
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
218
Long-term debt
(millions of dollars)
Public and Pri rate Use of IMF credit Short-term debt Total external debt
publicly guaranteed nonguaranteed (millions of dollars) (millions of dollars) (millions of dollars)
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
68 Chile 2,067 13,760 501 2,361 2 1,322 2,202 19,645
69 Syrian Arab Rep. 233 3,685 0 0 10 0 1,205 4,890
70 Costa Rica 134 3,531 112 317 0 71 611 4,530
71 Mexico 3,196 81,207 2,770 7,458 0 4,805 8,097 101,567
72 Mauritius 32 652 0 57 0 103 49 861
73 Poland 33,661 0 0 0 0 8,476 42,137
74 Malaysia 390 16,101 50 2,340 0 0 2,100 20,541
75 Panama 194 3,625 0 0 0 328 1,667 5,620
76 Brazil 3,421 89,841 1,706 11,514 0 3,333 9,903 114,592
77 Angola
78 Lebanon 64 229 0 0 0 0 270 499
79 Nicaragua 147 6,744 0 0 8 0 1,308 8,052
Upper-middle-income
80 South Africa
81 Algeria 945 23,229 6 0 0 0 1,621 24,850
82 Hungary 14,791 0 0 0 634 2,136 17,561
83 Uruguay 269 2,953 29 86 18 309 477 3,825
84 Argentina 1,880 48,166 3,291 1,378 0 3,678 5,714 58,936
85 Yugoslavia 1,199 13,949 854 5,392 0 1,310 1,033 21,684
86 Gabon 91 2,128 0 0 0 133 402 2,663
87 Venezuela 718 25,413 236 4,883 0 0 4,361 34,657
88 Trinidad and Tobago 101 1,718 0 0 0 115 163 1,995
89 Korea, Rep. of 1,816 21,349 175 6,027 0 0 9,780 37,156
90 Portugal 485 13,950 268 615 0 0 2,603 17,168
91 Greece 905 17,482 388 1,315 0 0 4,717 23,514
92 Oman 2,488 0 0 0 0 452 2,940
93 Libya
94 Iran, Islamic Rep.
95 Iraq
96 Romania 1,946 6 700 2,790
ow- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N.Africa
Latin America & Caribbean
everely indebted
High-income economies
OECD members
tOther
97 tSaudi Arabia
98 Spain
99 Ireland
100 flsrael
101 (Singapore
102 tHong Kong
103 New Zealand
104 Australia
105 United Kingdom
106 Italy
107 IKuwait
108 Belgium
109 Netherlands
110 Austria
Ill tUnited Arab Emirates
112 France
113 Canada
114 Denmark
115 Germany, Fed. Rep.
116 Finland
117 Sweden
118 United States
119 Norway
120 Japan
121 Switzerland
Total reporting economies
Oil exporters
Nonreporting nonmembers
219
Table 22. flow of public and private external capital
Disbursements Repayment ofprincipal Net flow
(millions of dollars) (millions of dollars) (millions of dollars)
Public and Public and Public and
publicly Private publicly Private publicly Private
guaranteed nonguaranteed guaranteed non guaranteed guaranteed nssnguaranteed
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
Low-income economies
China and India
Other low-income
I Mozambique 146 0 14 8 0 4 138 0 10
2 Ethiopia 28 465 0 0 15 160 0 0 13 305 0 0
3Chad 6 59 0 0 3 2 0 0 3 57 0 0
4 Tanzania 51 191 8 0 10 45 3
0
1 40
0
146
714
5
0
1 0
5 Bangladesh 0 890 0 0 0 177 0
6 Malawi 40 116 0 9 3 29 0 6 37 86 0 3
7 Somalia 4 47 0 0 1 1 0 0 4 46 0 0
8 Zaire 32 328 0 0 28 67 0 0 3 261 0 0
9 Bhutan 30 0 0 1 0 0 30 0 0
10 LaoPDR 6 119 0 0 1 7 0 0 4 112 0 0
11 Nepal
12 Madagascar 11
1 202
230
0
0
0
0
2
5
18
80
0
0
0
0
2 5
184
150
0
0
0
0
13 Burkina Faso 2 92 0 0 2 22 0 0 0 69 0 0
14 Mali 23 154 0 0 0 32 0 0 23 122 0 0
15 Bumndi 1 99 0 0 0 19 0 0 1 80 0 0
23 189
16 Uganda
17 Nigeria
27
56
212
693 25
0
83
0
0 4
38
23
483 30
0
90
0
0
18
316
209
16
50
10
7 0
0
18 Zambia 351 130 23 35 115 13
19 Niger 12 150 0 28 2 31 0 26 II 119 0 2
20 Rwanda 0 82 0 0 0 9 0 0 0 73 0 0
21 China 8,868 0 0 2,097 0 0 6,771 0 0
22 India 883 5,945 25 272 289 1,677 25 440 594 4,269 0 168
23 Pakistan 489 1,622 3 55 114 828 1 26 375 794 2 29
24 Kenya 35 291 41 196 17 216 12 65 17 75 30 131
25 Toga 5 73 0 0 2 23 0 0 3 50 0 0
26 Central African Rep. 2 84 1) 0 2 4 0 0 79 0 0
27 Haiti 4 46 0 0 3 15 0 0 1 31 0 0
28 Benin 2 51 0 0 1 10 0 0 1 41 0 0
29 Ghana 42 344 0 9 14 134 0 7 28 210 0 2
30 Lesotho 0 51 0 0 0 16 0 0 0 36 0 0
31
32
SriLanka
Guinea
66 402
308
.
0
.
0
0 30 200
112
. .
0
6
0
36
80
202
197
. .
0
60
90 11
33 Yemen,PDR 1 247 0 0 0 75 0 0 1 171 0 0
34 Indonesia 441 6,304 195 1,080 59 4,774 61 788 383 1,530 134 292
35 Mauritania 5 129 0 0 3 78 0 0 1 51 0 0
36 Sudan 53 373 0 0 22 44 0 0 30 329 0 0
37 Afghanistan . . . . . . . . . . . . . . . . . . . .
38 Myanmar 22 287 0 0 20 67 0 0 2 220 0 0
39 Kampuchea, Dem.
40 Liberia
. .
7 34
. .
0
. .
0
.
11
. . .
8
. .
0
. .
0 4. . . .
26
.
0 0
41 SierraLeone
42 VietNam
8 20 0 0 11 5 0 0 3 14 0 0
Middle-income economies
Lower-middle-income
43 Bolivia 55 328 3 0 17 132 2 0 38 196 1 0
44 Philippines 141 1,372 276 0 74 1,220 186 100 67 152 90 100
45 YemenArabRep. 0 0 137 0 0 124 0 0
46 Senegal
47 Zimbabwe
.
19
0
261
272
233
1
0 43
8 5
5
124
308
3
0 44
8
5
.
14
.
148
75
2 0 2 0
48 Egypt, Arab Rep. 397 1,467 0 180 310 635 0 147 87 832 0 33
49 DominicanRep. 38 108 22 0 7 91 20 10 31 17 2 10
50 Côted'Ivoire 78 440 4 850 29 224 2 414 49 216 2 436
51 PapuaNewGuinea 43 164 111 164 0 198 20 168 43 34
476
91 4
52 Morocco 168 1,008 8 8 37 533 3 8 131 5 0
53 Honduras 29 290 10 14 3 144 26 26 146 7 -12
54 Guatemala 37 262 6 0 20 252
3
2 3 17 10
223
4
0
3
55 Congo,Peoplc'sRep. 20 410 0 0 6 187 0 0 15 0
56 ElSalvador 8 190 24 0 6 93 16 15 2 96 8 15
57 Thailand 51 1,467 169 979 23 1,519 107 701 28 52 62 279
58 Botswana 6 51 0 0 0 39 0 0 6 11 0 0
59 Cameroon 29 553 II 115 5 149 2 228 24 405 9 112
60 Jamaica 15 204 165 0 6 200
266
164 7
26
9 3
357 4 I 7
5
61 Ecuador 41 623 7 0 16 11 5
62 Colombia 253 2,234 0 112 78 1,623 59 98 174 611 59 14
63 Paraguay 15 98 0 0 7 175 0 I 8 77 0
Tunisia 54 673 45 42 212 9
64
65 Turkey
66 Pens
89
329
148
5,565
412
885
240
1
. .
161
0
47
128
100 140
3,706
233
. .
3
10
138 201
48
1,859
272
2. .
7 10
23
67 Jordan 14 753 0 0 3 586 0 0 12 167 0 0
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
220
Disbursements Repayment ofprincipal Net flow
(millions of dollars) (millions of dollars) (millions of dollars)
Public and Public and Public and
publicly Private publicly Private publicly Prtvate
guaranteed nonguaranseed guaranteed nonguaranteed guaranteed nonguaranteed
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
68 Chile 408 891 247 669 166 389 41 209 242 502 206 461
69 SyrianArabRep. 60 435 0 0 31 260 0 0 29 176 0 0
70 CostaRica 30 94 30 30 21 132 20 16 9 -38 10 15
71 Mexico 772 4,706 603 1,140 475 3,087 542 3,155 297 1,619 61 -2,015
72 Mauritius 2 212 0 17 1 101 0 4 1 111 0 13
73 Poland 569 0 0 830 0 0 -262 0 0
74 Malaysia 45 1,687 12 615 47 3,023 9 885 -2 -1,336 3 -270
75 Panama 67 5 0 0 24 4 0 0 44 1 0 0
76 Brazil 892 5,534 900 170 256 2,980 200 657 636 2,553 700 -487
77 Angola - -
78 Lebanon 12 26 0 0 2 22 0 0 10 5 0 0
79 Nicaragua 44 579 0 0 16 82 0 0 28 497 0 0
Upper-middle-income
80 South Africa -
81 Algeria 313 5,922 0 0 35 4,534 0 0 279 1,388 0 0
82 Hungary 2,097 0 0 1,779 0 0 318 0 0
83 Uniguay 37 283 13 0 47 255 4 58 -10 28 9 -58
84 Argentina 482 1,676 424 160 344 1,190 428 150 139 486 -4 10
85 Yugoslavia 179 696 465 1,129 170 808 204 965 9 -112 261 164
86 Gabon 26 161 0 0 9 31 0 0 17 130 0 0
87 Venezuela 216 1,777 67 0 42 1,231 25 1,190 174 546 41 -1,190
88 TrinidadandTobago 8 183 0 0 10 60 0 0 -3 123 0 0
89 Korea,Rep.of 444 2,721 32 1,127 198 4,890 7 1,202 246 -2,168 25 -75
90 Portugal 18 2,879 20 75 63 3,220 22 99 -45 -341 -1 -24
91 Greece 163 2,271 144 187 62 2,032 37 147 101 239 107 40
92 Oman 414 0 0 - 348 0 0 66 0 0
93 Libya - - -
94 Iran, Islamic Rep.
95 Iraq
96 Romania - 94 0 0 3,359 0 0 - -3,266
Low- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N.Africa
Latin America & Caribbean
Severely indebted
High-income economies
OECD members
jOther
97 tSaudi Arabia
98 Spain
99 Ireland
100 jIsrael
101 jSingapore
102 (I-long Kong
103 New Zealand
104 Australia
105 United Kingdom
106 Italy
107 jKuwait
108 Belgium
109 Netherlands
110 Austria
Ill jUnited Arab Emirates
112 France
113 Canada
114 Denmark
115 Germany, Fed. Rep.
116 Finland
117 Sweden
118 United States
119 Norway
120 Japan
121 Switzerland
Total reporting economies
Oil exporters
Nonreporting nonmembers
a. Disbursements less repayments of principal may not equal net flow because of munding.
221
Table 23. Total external public and private debt and debt service ratios
Total long-term debt Total long-term debt
Total interest
outstanding and disbursed service as a percentage of
payments
As a percentage on long-term debt Exports of goods
Millions of dollars of GNP (millions of dollars) GNP and services
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
Low-income economies
China and India
Other low-income
I Mozambique . . 4,039 . . 399.7 . . 15 . . 2.7 . . 10.4
2 Ethiopia 169 2,790 9.5 50.6 6 78 1.2 4.3 11.4 37.4
3 Chad 33 300 9.9 33.2 0 4 0.9 0.7 4.2 2.7
4 Tanzania 265 4,100 20.7 140.1 8 41 1.6 3.0 6.3 17.8
5 Bangladesh 0 9,330 0.0 48.5 0 139 0.0 1.6 0.0 20.5
6 Malawi 122 1,193 43.2 85.9 4 29 2.3 4.6 7.8 19.0
7 Somalia 77 1,754 24.4 185.2 0 3 0.3 0.4 2.1 4.9
8 Zaire 311 7,013 9.1 118.0 9 98 1.1 2.8 4.4 6.9
9 Bhutan . . 68 27.9 . . 1 . . 0.5
10 LaoPDR 8 816 153.5 0 2 . . 1.8 143.5
11 Nepal 3 1,088 0.3 34.6 0 19 0.3 1.2 3.2 8.5
12 Madagascar 89 3,317 10.4 192.7 2 81 0.8 9.3 3.7 39.0
13 BurkinaFaso 21 805 6.6 43.4 0 14 0.7 2.0 7.1 11.9
14 Mali 238 1,928 71.4 100.8 0 15 0.2 2.5 1.4 14.2
15 Bunindi 7 749 3.1 69.8 0 16 0.3 3.3 2.3 25.1
16 Uganda 138 1,438 7.3 34.3 5 20 0.5 1.0 2.9 14.0
17 Nigeria 567 28,967 4.3 102.5 28 1,411 0.7 7.0 7.1 25.7
18 Zambia 654 4,194 37.5 116.7 32 62 4.6 4.9 8.0 14.2
19 Niger 32 1,542 5.0 66.0 1 74 0.4 5.6 4.0 32.6
20 Rwanda 2 585 0.9 25.5 0 8 0.2 0.7 1.5 9.6
21 China . . 32,196 . . 8.7 . . 1,593 . . 1.0 . . 6.9
22 India 7,938 51,168 13.9 19.3 193 2,554 0.9 1.8 23.7 24.9
23 Pakistan 3,069 14,027 30.6 37.6 78 436 1.9 3.5 23.9 24.1
24 Kenya 406 4,869 26.3 58.5 17 194 3.0 5.7 9.1 25.3
25 Togo 40 1,067 16.0 81.6 1 68 1.0 7.0 3.1 18.3
26 CentralAfricanRep. 24 584 13.5 53.3 1 7 1.7 1.1 5.1 5.9
27 Haiti 40 683 10.2 27.7 0 8 1.0 0.9 7.2 8.8
28 Benin 41 904 15.1 49.3 0 8 0.7 1.0 2.5 5.4
29 Ghana 498 2,270 22.9 44.6 12 64 1.2 4.0 5.5 20.6
30 Lesotho 8 270 7.7 36.5 0 7 0.5 3.0 4.5 5.2
31 SriLanka 317 4,253 16.1 61.6 12 125 2.1 4.8 11.0 17.6
32 Guinea 312 2,312 . . 94.7 4 31 . . 5.9 . . 21.9
33 Yemen, PDR 1 1,970 . . 199.4 0 31 . . 10.8 0.0 46.5
34 Indonesia 2,914 45,655 30.0 61.7 46 2,918 1.7 11.5 13.9 39.6
35 Mauritania 27 1,823 13.9 196.2 0 33 1.8 11.9 3.4 21.6
36 Sudan 298 8,418 14.8 74.6 12 19 1.7 0.6 10.6 9.5
37 Afghanistan . . . . . . . . .
38 Myanenar 106 4,217 3 39 . . .
39 Kanipuchea, Dens. . . . . . . . . . . . . .. .
40 Liberia 158 1,101 39.2 . . 6 6 4.3 8.0
41 SierraLeone 59 510 14.2 . . 3 3 3.1 10.8 5,9
42 VietNam . . . . . . . . . . . . . .
Middle-income economies
Lower-middle-income
43 Bolivia 491 4,651 49.3 114.9 7 95 2.6 5.6 12.6 32.9
44 Philippines 1,544 24,467 21.8 62.6 44 1,638 4.3 7.6 23.0 27.7
45 YemenArabRep. . . 2,378 . . 41.7 . . 56 . . 3.4 . . 16.0
46 Senegal 131 3,019 15.5 63.6 2 117 1.1 5.2 4.0 19.3
47 Zimbabwe 229 2,281 15.5 37.3 5 150 0.6 8.2 2.3 27.9
48 Egypt, Arab Rep. 1,714 43,259 22.5 126.7 56 729 4.8 4.4 38.0 16.6
49 DominicanRep. 353 3,334 23.9 77.3 13 151 2.7 5.8 15.3 14.4
50 Côted'Ivoire 267 11,788 19.5 135.1 12 447 3.1 12.4 7.5 31.9
51 PapuaNewGuinea 209 2,129 33.4 64.2 10 153 4.8 15.6 24.5 30.9
52 Morocco 727 18,767 18.6 89.8 25 814 1.7 6.5 9.2 25.1
53 Honduras 109 2,837 15.6 68.3 4 128 1.4 7.2 4.9 28.6
54 Guatemala 120 2,244 6.5 28.3 7 104 1.6 4.5 8.2 27.2
55 Congo, People's Rep. 124 4,098 46.5 205.0 3 75 3.4 13.1 11.5 28.7
56 ElSalvador 176 1,685 17.3 31.5 9 66 3.1 3.3 12.0 18.8
57 Thailand 726 16,905 10.2 29.7 33 1,184 2.3 6.0 14.0 15.7
58 Botswana 17 494 21.3 37.9 0 34 0.7 5.7 1.0 4.0
59 Cameroon 140 3,366 12.6 27.0 5 192 1.0 4.6 4.0 27.0
60 Jamaica 982 3,554 73.1 127.2 64 217 17.4 15.2 43.5 24.8
61 Ecuador 242 9,378 14.8 94.2 10 297 2.2 5.7 14.0 21.4
62 Colombia 1,580 15,392 22.5 42.1 59 1,213 2.8 8.0 19.3 42.3
63 Paraguay 112 2,119 19.2 36.4 4 114 1.8 5.0 11.8 24.6
64 Tunisia 541 6,121 38.6 64.2 18 380 4.7 11.5 19.7 25.5
65 Turkey 1,886 31,589 15.0 46.1 44 2,424 1.4 9.1 22.6 35.2
66 Peru 2,655 13,898 37.3 56.1 162 174 7.0 1.3 40.0 8.7
67 Jordan 119 3,955 22.9 94.0 2 239 0.9 19.6 3.6 31.9
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified
222
Total long-term debt Total long-term debt
Total interest
outstanding and disbursed service as a percentage of
payments
As a percentage on long-term debt Exports of goodo
Millions of dollars of GNP (millions of dollars) GNP and services
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
68 Chile 2,568 16,121 32.1 79.3 104 1,019 3.9 7.9 24.5 19.1
69 SyrianArabRep. 233 3,685 10.8 25.0 6 119 1.7 2.6 11.3 21.1
70 CostaRica 246 3,847 25.3 89.2 14 185 5.7 7.7 19.9 19.9
71 Mexico 5,966 88,665 16.2 52.4 283 7,590 3.5 8.2 44.3 43.5
72 Mauritius 32 709 14.3 37.1 2 43 1.4 7.7 3.2 10.4
73 Poland . . 33,661 . . 51.1 . . 829 . . 2.5 . . 10.0
74 Malaysia 440 18,441 10.8 56.3 25 1,498 2.0 16.5 4.5 22.3
75 Panama 194 3,625 19.5 81.2 7 4 3.1 0.2 7.7 0.2
76 Brazil 5,128 101,356 12.2 29.6 224 11,686 1.6 4.5 21.8 42.0
77 Angola . . . . . . . . .
78 Lebanon 64 229 4.2 . . 1 16 0.2 .
79 Nicaragua 147 6,744 19.5 . . 7 73 3.1 . . 10.6
Upper-middle-income
80 South Africa . . . . . . . . . . . . . . . . .
81 Algeria 945 23,229 19.8 46.6 10 1,809 0.9 12.7 4.0 77.0
82 Hungary . . 14,791 . . 54.9 . . 1,100 . . 10.7 0.0 23.3
83 Uruguay 298 3,039 12.5 39.8 17 257 2.9 7.5 23.6 30.3
84 Argentina 5,171 49,544 23.8 58.6 338 2,803 5.1 4.9 51.7 36.0
85 Yugoslavia 2,053 19,341 15.0 38.9 104 1,401 3.5 6.4 19.7 17.6
86 Gabon 91 2,128 28.8 65.6 3 57 3.8 2.7 5.7 6.2
87 Venezuela 954 30,296 7.5 49.0 53 2,675 0.9 8.2 4.2 39.7
88 TrinidadandTobago 101 1,718 13.3 43.2 6 88 2.1 3.7 4.6 9.2
89 Korea, Rep. of 1,991 27,376 22.3 16.2 76 2,081 3.1 4.8 20.4 11.5
90 Portugal 753 14,565 12.1 35.6 34 1,163 1.9 11.0 8.7 30.3
91 Greece 1,293 18,797 12.7 35.9 63 1,468 1.6 7.0 14.7 32.1
92 Oman 2,488 . . 34.7 . . 182 7.4
93 Libya . . .
94 Iran, Islamic Rep. . . . . .
95 Iraq . . . . . . . . .
96 Romania . 1,946 . . . . . . 420
Low- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N.Africa
Latin America & Caribbean
Severely indebted
High-income economies
OECD members
(Other
97 tSaudi Arabia
98 Spain
99 Ireland
100 (Israel
101 tSingapore
102 THong Kong
103 New Zealand
104 Australia
105 United Kingdom
106 Italy
107 (Kuwait
108 Belgium
109 Netherlands
110 Austria
Ill limited Arab Emirates
112 France
113 Canada
114 Denmark
115 Germany, Fed. Rep.
116 Finland
117 Sweden
118 United States
119 Norway
120 Japan
121 Switzerland
Total reporting economies
Oil exporters
Nonreporting nonmembers
Note: Public and private debt includes public, publicly guaranteed, and private nonguaranteed debt; data are shown only when they are available for all categories.
223
Table 24. External public debt and debt service ratios
External public debt outstanding and disbursed Debt service as a percentage of
Interest payments on
Millions of As a percentage external public debt Exports of
dollars of GNP (millions of dollars) GNP goods and services
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
Low-income economies . 248,548 t 27.6 w 9,662 t 2.4w 17.0 w
China and India . 81,891 t . . 12.8w . . 4,005 t .. 1.2w . 10.8w
Other low-income 10,424 t 166,656 t 15.2w 64.4w 257 t 5,658 I 1.1 w 5.3 w 7.1 w 25.3 w
3,801 376.1 . 12 . 2.0 . . 7.8
1 Mozambique . . . . . .
169 2,790 9.5 50.6 6 78 1.2 4.3 11.4 37.4
2 Ethiopia
33 300 9.9 33.2 0 4 0.9 0.7 4.2 2.7
3 Chad
4 Tanzania 250 4,091 19.5 139.8 7 40 1.3 2.9 5.3 17.1
0 9,330 0.0 48.5 0 139 0.0 1.6 0.0 20.5
5 Bangladesh
122 1,190 43.2 85.7 4 28 2.3 4.1 7.8 17.2
6 Malawi
Somalia 77 1,754 24.4 185.2 0 3 0.3 0.4 2.1 4.9
7
311 7,013 9.1 118.0 9 98 1.1 2.8 4.4 6.9
8 Zaire
9 Bhutan . . 68 27.9 . . I . . 0.5
8 816 153.5 0 2 . 1.8 143.5
10 Lao PDR .
Nepal 3 1,088 0.3 34.6 0 19 0.3 1.2 3.2 8.5
11
89 3,317 10.4 192.7 2 81 0.8 9.3 3.7 39.0
12 Madagascar
21 805 6.6 43.4 0 14 0.7 2.0 7.1 11.9
13 BurkinaFaso
Mali 238 1,928 71.4 100.8 0 15 0.2 2.5 1.4 14.2
14
7 749 3.1 69.8 0 16 0.3 3.3 2.3 25.1
15 Bunindi
138 1,438 7.3 34.3 5 20 0.5 1.0 2.9 14.0
16 Uganda
17 Nigeria 452 28,630 3.4 101.3 20 1,385 0.4 6.6 4.3 24.2
Zambia 624 4,194 35.8 116.7 29 62 3.7 4.9 6.4 14.2
18
19 Niger 32 1,286 5.0 55.1 1 54 0.4 3.6 4.0 21.1
2 585 0.9 25.5 0 8 0.2 0.7 1.5 9.6
20 Rwanda
China 32,196 . 8.7 . 1,593 . 1.0 . . 6.9
21 . . . . .
7,838 49,695 13.7 18.7 187 2,412 0.8 1.5 22.2 21.8
22 India
23 Pakislan 3,064 13,944 30.6 37.4 77 430 1.9 3.4 23.8 23.5
24 Kenya 319 4,241 20.6 51.0 13 149 2.0 4.4 6.0 19.4
40 1,067 16.0 81.6 68 1.0 7.0 3.1 18.3
25 Togo 1
CentrelAfricanRep. 24 584 13.5 53.3 7 1.7 1.1 5.1 5.9
26 1
40 683 10.2 27.7 0 8 1.0 0.9 7.2 8.8
27 Haiti
41 904 15.1 49.3 0 8 0.7 1.0 2.5 5.4
28 Benin 3.8 5.5 19.7
29 Ghana 488 2,238 22.5 43.9 12 61 1.2
8 270 7.7 36.5 0 7 0.5 3.0 4.5 5.2
30 Lesotho
31 SriLanka 317 4,139 16.1 59.9 12 123 2.1 4.7 11.0 17.2
32 Guinea 312 2,312 94.7 4 31 5.9 . 21.9
1,970 199.4 0 31 . 10.8 0.0 46.5
33 Yemen, PDR 1 . .
.
34 Indonesia 2,453 41,258 25.3 55.7 25 2,528 0.9 9.9 7.0 34.1
27 1,823 13.9 196.2 0 33 1.8 11.9 3.4 21.6
35 Mauritania
298 8,044 14.8 71.3 12 19 1.7 0.6 10.6 9.5
36 Sudan
37 Afghanistan . . . . . . . . . . .
38 Myanmar 106 4,217 3 39 . . .
39 Kampuchea, Dem. . . . . . . . . . . . . . . .
40 Liberia 158 1,101 39.2 . . 6 6 4.3 . 8.0
SierraLeone 59 510 14.2 3 3 3.1 . . 10.8 5.9
41 . .
42 VietNam . . . . . . . . .
Middle-income economies 28,7541 662,9721 11.5w 40.8w 1,236t 43,7001 1.7w 5.5w 11.1w 21.6w
Lower-middle-income 20,310 I 472,6361 12.1 w 44.7w 825 t 30,161 1 1.7w 5.2w 12.6w 23.0w
480 4,451 48.2 109.9 7 95 2.3 5.6 11.3 32.9
43 Bolivia
44 Philippines 625 23,475 8.8 60.1 26 1,515 1.4 7.0 7.5 25.6
2,378 41.7 56 3.4 . . 16.0
45 YemenArabRep. . . . . . . . .
2.9 18.4
46 Senegal 100 2,985 11.9 62.9 2 114 0.8 5.0
229 2,231 15.5 36.5 5 139 0.6 7.3 2.3 24.8
47 Zimbabwe
1,714 42,128 22.5 123.4 56 624 4.8 3.7 38.0 13.9
48 Egypt, Arab Rep.
49 Dominican Rep. 212 3,216 14.4 74.5 4 144 0.8 5.5 4.5 13.4
256 8,088 18.7 92.7 12 219 2.9 5.1 7.1 13.0
50 Côted'Ivoire 16.5
51 Papua New Guinea 36 1,269 5.8 38.3 1 80 0.2 8.4 1.1
712 18,567 18.2 88.8 24 809 1.6 6.4 8.7 24.8
52 Morocco
53 Honduras 90 2,739 12.9 65.9 3 121 0.8 6.4 2.8 25.5
106 2,131 5.7 26.9 6 98 1.4 4.4 7.4 26.5
54 Guatemala
Congo, People's Rep. 124 4,098 46.5 205.0 3 75 3.4 13.1 11.5 28.7
55
56 El Salvador 88 1,630 8.6 30.4 4 61 0.9 2.9 3.6 16.6
57 Thailand 324 13,375 4.6 23.5 16 926 0.6 4.3 3.3 11.3
58 Botswana 17 494 21.3 37.9 0 34 0.7 5.7 1.0 4.0
2,939 11.8 23.6 4 101 0.8 2.0 3.2 11.9
59 Cameroon 131
24.2
60 Jamaica 160 3,512 11.9 125.7 9 213 1.1 14.8 2.8
61 Ecuador 193 9,353 11.8 93.9 7 294 1.4 5.6 8.6 21.1
62 Colombia 1,297 13,853 18.5 37.9 44 1,135 1.7 7.5 12.0 39.8
112 2,091 19.2 35.9 4 114 1.8 4.9 11.8 24.5
63 Paraguay
Tunisia 541 5,886 38.6 61.7 18 369 4.7 10.9 19.7 24.2
64
65 Thrkey 1,844 31,054 14.7 45.3 42 2,364 1.4 8.9 21.9 34.1
856 12,475 12.0 50.3 43 164 2.0 1.2 11.6 8.1
66 Peru
Jordan 119 3,955 22.9 94.0 2 239 0.9 19.6 3.6 31.9
67
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified
224
External public debt outstanding and disbursed Debt service as a percentage of
Interest payments on
Millions of As a percentage external public debt Exports of
dollars of GNP (millions of dollars) GNP goods and services
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
68 Chile 2,067 13,760 25.8 67.7 78 868 3.1 6.2 19.2 14.9
69 Syrian Arab Rep. 233 3,685 10.8 25.0 6 119 1.7 2.6 11.3 21.1
70 Costa Rica 134 3,531 13.8 81.8 7 160 2.9 6.8 10.0 17.4
71 Mexico 3,196 81,207 8.7 48.0 216 6,554 1.9 5.7 23.6 30.3
72 Mauritius 32 652 14.3 34.1 2 41 1.4 7.4 3.2 10.1
73 Poland 33,661 51.1 829 2.5 10.0
74 Malaysia 390 16,101 9.5 49.1 22 1,220 1.7 13.0 3.8 17.5
75 Panama 194 3,625 19.5 81.2 7 4 3.1 0.2 7.7 0.2
76 Brazil 3,421 89,841 8.2 26.3 135 10,117 0.9 3.8 12.5 35.9
77 Angola
78 Lebanon 64 229 4.2 16 0.2
79 Nicaragua 147 6,744 19.5 7 73 3.1 10.
Upper-middle-income 8,444 I 190,336 10.3w 33.6w 411 I 13,5391 1.7w 6.0w 8.9w 19.7w
80 South Africa
81 Algeria 945 23,229 19.8 46.6 jo 1809 0.9 12.7 77.0
82 Hungaly 14,791 54.9 1,100 10.7 23.3
83 Umguay 269 2,953 11.3 38.7 257 2.7 6.7 21.7 27.3
84 Argentina 1,880 48,166 8.6 57.0 121 2,560 2.1 4.4 21.6 32.6
85 Yugoslavia 1,199 13,949 8.8 28.0 7 951 1.8 3.5 10.0 9.7
86 Gabon 91 2,128 28.8 65.6 57 3.8 2.7 5.7 6.2
87 Venezuela 718 25,413 5.6 41.1 40 2,043 0.6 5.3 2.9 25.5
88 Trinidad and Tobago 101 1,718 13.3 43.2 6 88 2.1 3.7 4.6 9.2
89 Korea, Rep. of 1,816 21,349 20.3 12.6 71 1,563 3.0 3.8 19.5 9.1
90 Portugal 485 13,950 7.8 34.1 29 1,110 1.5 10.6 6.8 29.3
91 Greece 905 17,482 8.9 33.4 41 1,355 1.0 6.5 9.4 29.8
92 Oman 2,488 34.7 182 0.0 7.4
93 Libya
94 Iran, Islamic Rep.
95 Iraq
96 Ro,nania 1,946 420
Low- and middle-income 47,015 t 911,520 12.5 w 36.1 w 1,680 53,3621 1.2w 4.4w 10.3w 20.5w
Sub-Saharan Africa 5,369 t 112,353 I 12.5w 78.2w 1691 3,128 t 1.1 w 4.3w 5.3w 16.5w
East Asia 5,667 t 150,401 15.0w 20.1 w 162 1 9,4561 0.4w 3.6w 5.9w 13.3w
South Asia 11,327 1 82,541 14.3 w 24.0w 279 1 3,1641 0.9w 1.8w 18.1w 21.5w
Europe, M.East, & N.Africa 8,7841 231,433 13.6w 50.5w 304t 12,3841 1.8w 7.1 w 10.2w 25.2w
Latin America & Caribbean 15,868 t 334,792 1 10.5 w 40.6w 7671 25,2291 1.6w 4.7w 13.1 w 28.1 w
Severely indebted 15,268t 406,298t 10.2 w 44.4w 745 r 27,968 t 1.6w 4.7 w 12.0w 26.2 w
High-income economies
OECD members
jOther
97 tSaudi Arabia
98 Spain
99 Ireland
100 (Israel
101 tSingapore
102 ¶Hong Kong
103 New Zealand
104 Australia
105 United Kingdom
106 Italy
107 jKuwait
108 Belgium
109 Netherlands
110 Austria
Ill jUnited Arab Emirates
112 France
113 Canada
114 Denmark
115 Germany, Fed. Rep.
116 Finland
117 Sweden
118 United States
119 Norway
120 Japan
121 Switzerland
Total reporting economies
Oil exporters
Nonreporting nonmembers
225
Table 25. Terms of external public borrowing
Public loans with
Average interest Average Average variable interest rates,
Commitments rate maturity grace period as a percentage
(millions of dollars) (percent) (years) (years) ofpublic debt
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
ow-income economies 35,7701 .. 5.3w 21w 6w 19.4w
China and India 18,070 t . . 6.7 w . . 16 w 5w 24.8 w
Other low-income 3,390 t 17,701 t 3.2 w 3.9 w 29w 26 w w 8w 0.2 w 16.7 ii
I Mozambique . . 294 . 2.2 . . 23 . . 6 . . 4.9
2 Ethiopia 21 487 4.4 3.0 32 24 7 7 0.1 6.7
3 Chad 10 176 5.7 0.9 8 41 1 10 0.0 0.1
4 Tanzania 284 220 1.2 0.8 39 41 11 10 1.6 2.5
5 Bangladesh 0 994 0.0 1.5 0 33 0 10 0.0 0.0
6 Malawi 14 123 3.9 0.9 29 41 6 10 0.0 3.8
7 Somalia 22 24 0.0 2.5 20 28 16 7 0.0 1.1
8 Zaire 258 350 6.5 4.1 13 28 4 8 0.0 5.4
9 Bhutan . . 28 . . 5.9 . . 17 . . 4 . . 0.0
10 Lao PDR 12 86 3.0 0.3 28 46 4 30 0.0 0.0
11 Nepal 17 217 2.8 0.9 27 39 6 10 0.0 0.7
12 Madagascar 23 352 2.3 2.4 39 31 9 9 0.0 10.5
13 BurkinaFaso 9 108 2.3 2.6 37 31 8 8 0.0 0.5
14 Mali 34 131 1.1 1.2 25 35 10 9 0.0 0.1
15 Bumndi 1 205 2.8 1.0 4 38 2 11 0.0 0.3
16 Uganda 12 252 3.8 2.4 28 36 7 9 0.0 0.6
17 Nigena 65 1,461 6.0 7.6 14 16 4 5 2.7 40.6
18 Zambia 557 64 4.2 3.8 27 20 9 7 0.0 13.4
19 Niger 19 159 1.2 1.4 40 31 8 9 0.0 7.7
20 Rwanda 9 64 0.8 1.7 50 35 II 8 0.0 0.0
21 China . . 10,086 . . 7.1 . . 13 . . 4 . . 39.1
22 India 954 7,984 2.5 6.3 34 20 8 6 0.0 15.6
23 Pakistan 951 2,020 2.8 5.2 32 23 12 6 0.0 8.8
24 Kenya 50 679 2.6 1.9 37 22 8 9 0.1 4.1
25 Togo 3 114 4.7 1.6 17 38 4 II 0.0 3.7
26 Central African Rep. 7 76 2.0 0.9 36 36 8 10 0.0 0.0
27 Haiti 5 0 4.8 0.0 10 0 1 0 0.0 1.1
28 Benin 7 162 1.8 2.3 32 32 7 8 0.0 3.9
29 Ghana 51 431 2.0 1.0 37 35 10 8 0.0 3.7
30 Lesotho 0 191 4.4 4.6 22 24 2 5 0.0 1.0
31 Sri Lanka 81 827 3.0 1.9 27 35 5 10 0.0 4.5
32 Guinea 68 271 2.9 1.4 13 30 5 8 0.0 10.5
33 Yemen, PDR 63 664 0.0 2.9 21 24 II 6 0.0 0.0
34 Indonesia 530 5,740 2.6 5.2 34 21 9 7 0.0 28.6
35 Mauritania 7 Ill 6.1 1.1 II 38 3 10 0.0 6.4
36 Sudan 95 472 1.8 2.0 17 35 9 8 0.0 0.8
37 Afghanistan . . . . . . . . . . . . . . . . .
38 Myanmar 48 0 4.1 0.0 16 0 5 0 0.0 0.8
39 Kampuchea, Dem. . . . . . . . . . . . . . . . . .
40 Liberia 12 0 6.7 0.0 19 0 5 0 0.0 9.3
41 SierraLeone 25 0 2.9 1.0 27 12 6 3 10.6 0.6
42 VietNam . . . . . . . . . . .
Middle-income economies 8,195 1 58,340 t 6.2 w 7.2 w 16 w 13 w 4w Sw 2,8w 52.6w
Lower-middle-income 5,823 t 36,009 6.1 w 6.9 w 17 w 16 w 5w 5w 3.2w 49.7w
43 Bolivia 24 309 1.9 5.4 48 27 4 7 0.0 22.7
44 Philippines 171 2,341 7.3 5.4 12 22 2 7 0.8 43.2
45 Yemen Arab Rep. . . 463 . . 4.8 . . 16 . . 4 . . 0.8
46 Senegal 7 320 3.8 1.8 24 31 7 9 0.0 2.5
47 Zimbabwe 448 . . 7.1 19 4 . . 22.9
48 Egypt, Arab Rep. 704 1,762 5.3 6.4 21 23 8 7 0.0 1.4
49 DominicanRep. 20 169 2.4 7.5 28 15 5 4 0.0 31.5
50 Côte d'Ivoire 71 659 5.8 5.7 19 16 5 5 9.0 52.2
51 Papua New Guinea 91 219 6.4 3.5 22 20 8 6 0.0 28.6
52 Monscco 187 1,156 4.6 6.9 20 18 3 4 0.0 38.8
53 Honduras 23 251 4.1 7.3 30 17 7 5 0.0 18.9
54 Guatemala 50 444 3.7 5.9 26 17 6 5 10.3 11.1
55 Congo,People'sRep. 31 486 2.8 8.3 17 10 6 2 0.0 40.2
56 ElSalvador 12 369 4.7 4.0 24 33 6 8 0.0 3.9
57 Thailand 106 1,999 6.8 5.3 19 18 4 6 0.0 34.4
58 Botswana 38 103 0.6 2.4 39 29 10 8 0.0 13.6
59 Camemon 42 922 4.7 4.5 29 14 8 4 0.0 9.4
60 Jamaica 24 174 6.0 6.7 16 18 3 4 0.0 26.1
61 Ecuador 78 515 6.2 6.2 20 18 4 5 0.0 66.9
62 Colombia 363 2,488 6.0 8.2 21 13 5 4 0.0 45.0
63 Paraguay 14 312 5.6 6.4 25 21 6 7 0.0 13.8
64 Tunisia 144 1,114 3.5 4.6 28 21 6 6 0.0 17.2
65 Turkey 484 4,568 3.6 7.0 19 12 5 5 0.9 34.1
66 Peru 125 350 7.4 7.6 14 21 4 18 0.0 32.7
67 Jordan 35 999 3.8 7.5 15 II 5 3 0.0 29.0
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified
226
Public loans with
Average interest Average Average variable interest rates,
Commitments rate maturity grace period as a percentage
(millions of dollars) (percent) (years) (years) ofpublic debt
1970 1988 1970 1988 1970 1988 1970 1988 1970 1988
68 Chile 361 649 6.8 6.8 12 13 4 3 0.0 76.7
69 SyrianArabRep. 14 275 4.4 7.3 9 17 2 4 0.0 1.1
70 Costa Rica 58 338 5.6 8.9 28 21 6 6 7.5 47.5
71 Mexico 858 2,579 8.0 8.2 12 15 3 3 5.7 78.6
72 Mauritius 14 176 0.0 7.7 24 15 2 3 6.0 19.2
73 Poland . . 1,042 . 6.4 . . 6 . . 3 . . 63.3
74 Malaysia 84 2,259 6.1 6.1 19 15 5 6 0.0 48.5
75 Panama 111 0 6.1 0.0 15 0 4 0 0.0 60.2
76 Brazil 1,439 5,003 6.8 9.5 14 11 3 4 11.8 70.0
77 Angola . . . . . . . . . . . . .
78 Lebanon 7 40 2.9 7.5 22 27 1 4 0.0 15.3
79 Nicaragua 23 563 7.1 5.1 18 19 4 3 0.0 20.6
Upper-middle-income 2,372 1 22,331 1 6.5w 7.6w 14w 9w 4w 4w 1.6w 59.6w
80 South Africa . . . . . . . . . . . . . . . . .
81 Algeria 378 8,457 5.6 7.6 12 6 3 2 2.8 35.8
82 Hungalya . 1,782 . . 7.0 . . 9 . . 7 . . 60.2
83 Umguay 71 293 8.0 9.5 12 10 3 1 0.7 71.6
84 Argentina 494 2,432 7.3 7.8 12 16 3 5 0.0 82.5
85 Yugoslavia 199 369 7.1 9.2 17 7 6 5 3.3 64.9
86 Gabon 33 152 5.1 7.7 11 16 2 6 0.0 7.6
87 Venezuela 188 2,129 7.6 8.1 8 10 2 3 2.6 86.8
88 TrinidadandTobago 3 190 7.6 7.5 10 5 1 5 0.0 37.4
89 Korea,Rep.of 691 1,071 5.8 7.6 19 20 6 4 1.2 30.5
90 Portugal 59 3,195 4.3 7.7 17 9 4 5 0.0 37.8
91 Greece 246 1,914 7.2 6.3 9 9 4 6 3.5 52.2
92 Oman 285 7.5 10 6 39.8
93 Libya
94 Iran, Islamic Rep. . . . . . . . . . . .
95 Iraq
96 Romania 0 0.0 .. '6 .. 21.9
Low- and middle-income 12,539 1 94,1101 5.1 w 6.5 w 21 w 16w 6w 5w 1.7w 43.5 w
Sub-Saharan Africa 1,900 t 10,329 t 3.6 w 4.0 w 26 w 24w 8w 7w 0.9w 19.2 w
East Asia 1,688 I 23,885 t 5.0w 6.2 w 23 w 17 w 6w 5w 0.5 w 35.8w
SouthAsia 2,052t 12,073t 2.7w 5.3w 32w 23w lOw 7w 0.0 w 11.2w
Europe, M.East, & N.Africa 2,5261 28,102 t 5.0w 7.0w 18 w 11 w 5w 4w 1.3w 36.3 w
Latin America & Caribbean 4,373 t 19,722 1 6.9 w 8.1 w 14w 14 w 4w 4w 4.0 w 68.1 w
Severely indebted 4,2081 23,198 1 7.0 w 7.6 w 14w 14 w 3w 5w 4.2w 66.3w
High-income economies
OECD members
j'Other
97 tSaudi Arabia
98 Spain
99 Ireland
100 (Israel
101 tSingapore
102 tHong Kong
103 New Zealand
104 Australia
105 United Kingdom
106 Italy
107 tKuwait
108 Belgium
109 Netherlands
110 Austria
111 tUnited Arab Emirates
112 France
113 Canada
114 Denmark
115 Germany, Fed. Rep.
116 Finland
117 Sweden
118 United States
119 Norway
120 Japan
121 Swilzerland
Total reporting economies
Oil exporters
Nonreporting nonmembers
a. Includes debt in convertible currencies only.
227
Table 26. Population growth and projections
Hypothetical Age structure of
size of population (percent)
Average annual growth ofpopulation stationary
(percent) Population (millions) 0-14 years 15-64 years
population
1965-80 1980-88 1988-2000 1988 2000 2025 (millions) 1988 2025 1988 2025
Low-income economies 2.3 w 2.0 w 1.9w 2,8841 3,6201 5,2001 35.7w 27.1 w 59.8 65.1 w
China and India 2.2w 1.6w 1.5 w 1,9041 2,283 t 2,917 31.8w 22.1 w 63.2 67.4w
Other low-income 2.6w 2.8 w 2.6 w 980 1 1,337 t 2,284 I 43.4w 33.4 w 53.4 62.2 w
I Mozambique 2.5 2.7 3.1 15 21 41 93 43.9 38.6 53.0 58.2
2 Ethiopia 2.7 2.9 3.3 47 70 156 471 47.2 43.7 49.5 53.8
3 Chad 2.0 2.4 2.7 5 7 14 30 41.6 37.7 54.9 58.3
4 Tanzania 3.3 3.5 3.4 25 37 74 158 48.8 37.9 49.0 59.2
5 Bangladesh 2.7 2.8 2.4 109 145 219 346 44.7 28.2 52.1 67.4
6 Malawi 2.9 3.4 3.5 8 12 26 79 46.4 43.4 51.0 54.0
7 Somalia 2.7 3.0 3.1 6 9 17 41 45.8 40.2 51.4 56.7
8 Zaire 2.8 3.1 3.0 33 47 87 173 46.2 35.4 51.4 61.0
9 Bhutan 1.6 2.1 2.4 1 2 3 5 40.0 33.0 56.7 62.7
10 La0PDR 1.9 2.6 2.9 4 6 10 19 43.7 35.0 53.5 60.9
11 Nepal 2.4 2.6 2.5 18 24 37 61 42.7 29.8 54.9 65.2
12 Madagascar 2.5 2.8 2.6 11 15 24 42 46.2 31.5 50.7 64.7
13 BurkinaFaso 2.1 2.6 2.9 9 12 23 52 45.1 39.0 51.9 58.3
14 Mali 2.1 2.4 3.0 8 II 24 63 46.6 41.8 50.5 55.7
15 Burundi 1.9 2.8 3.1 5 7 15 33 45.3 39.6 51.5 57.8
16 Uganda 2.9 3.2 3.5 16 24 51 128 48.4 41.5 49.2 56.2
17 Nigeria 2.5 3.3 3.1 110 159 302 617 48.0 36.4 49.4 60.4
18 Zambia 3.0 3.7 3.5 8 11 24 56 49.0 39.7 48.6 57.6
19 Niger 2.6 3.5 3.3 7 11 24 82 46.7 44.8 50.5 52.8
20 Rwanda 3.3 3.3 3.8 7 10 23 70 47.7 44.0 49.7 53.6
21 China 2.2 1.3 1.3 1,088 1,275 1,566 1,835 27.7 20.5 66.7 66.6
22 India 2.3 2.2 1.8 816 1,007 1,350 1,862 37.2 24.0 58.4 68.4
23 Pakistan 3.1 3.2 3.1 106 154 285 556 45.0 34.9 52.4 61.5
24 Kenya 3.6 3.8 3.4 22 34 62 113 50.9 31.9 46.1 64.7
25 Togo 3.0 3.5 3.3 3 5 9 19 47.7 36.5 49.3 60.4
26 CentralAfricanRep. 1.9 2.7 2.6 3 4 7 13 42.6 34.7 54.9 61.7
27 Haiti 2.0 1.8 1.9 6 8 II 17 39.3 27.1 56.8 67.3
28 Benin 2.7 3.2 2.9 4 6 11 21 47.2 34.4 50.0 62.2
29 Ghana 2.2 3.4 3.0 14 20 36 66 46.6 34.0 50.2 62.4
30 Lesotho 2.3 2.7 2.6 2 2 4 6 43.1 30.5 53.3 64.7
31 SriLanka 1.8 1.5 1.1 17 19 24 28 32.8 21.0 61.8 65.9
32 Guinea 1.5 2.4 2.6 5 7 14 34 42.1 40.5 54.3 56.2
33 Yemen, PDR 2.1 3.0 3.0 2 3 6 11 45.2 35.1 52.0 61.8
34 Indonesia 2.4 2.1 1.7 175 213 282 370 37.3 23.3 58.9 68.2
35 Mauritania 2.3 2.6 2.7 2 3 5 13 43.9 41.2 52.6 56.1
36 Sudan 2.8 3.1 2.7 24 33 57 107 44.8 34.7 52.1 61.6
37 Afghanistan 2.4 . . . . . . . . . . . . . . . . .
38 Myanmar 2.3 2.1 2.0 40 50 69 94 37.9 23.9 58.0 68.6
39 Kampuchea, Dem. 0.3 . . . . . . . . . . S .
40 Liberia 3.0 3.2 2.8 2 3 6 11 45.1 33.9 51.5 62.3
41 SierraLeone 2.0 2.4 2.6 4 5 10 25 43.0 40.2 54.0 56.4
42 VietNam . . 2.4 2.0 64 83 117 160 40.0 24.0 55.7 68.8
Middle-income economies 2.4 w 2.2 w 1.9w 1,068 I 1,342 t 1,923 1 36.6w 26.0 w 58.6 w 65.3 w
Lower-middle-income 2.5 w 2.3 w 2.0 w 742 I 940 1 1,3541 38.0 w 26.0 w 57.7 w 65.9w
43 Bolivia 2.5 2.7 2.7 7 10 16 27 43.9 31.5 52.9 64.2
44 Philippines 2.9 2.5 1.9 60 75 103 139 40.3 23.9 56.2 68.4
45 YemenArabRep. 2.3 3.4 3.6 8 13 29 86 48.4 43.7 48.5 54.3
46 Senegal 2.9 3.0 3.2 7 10 21 48 46.8 40.2 50.5 57.4
47 Zimbabwe 3.1 3.7 2.7 9 13 20 29 45.8 25.4 50.8 68.7
48 Egypt, Arab Rep. 2.1 2.6 2.3 50 66 97 140 40.4 24.9 55.4 67.1
49 DominicanRep. 2.7 2.4 1.8 7 9 11 15 38.5 23.2 57.9 68.0
50 Côted'Ivoire 4.1 4.0 3.8 11 18 38 94 49.0 40.7 49.0 56.5
51 PapuaNewGuinea 2.4 2.4 2.2 4 5 7 11 41.3 27.4 56.0 67.9
52 Morucco 2.5 2.7 2.4 24 32 47 69 41.2 26.1 54.5 68.0
53 Honduras 3.2 3.6 2.9 5 7 11 18 45.1 28.1 51.1 66.9
54 Guatemala 2.8 2.9 2.8 9 12 21 35 45.8 30.7 51.3 64.8
55 Congo, People's Rep. 2.8 3.5 3.3 2 3 7 17 44.9 40.3 51.0 56.7
56 ElSalvador 2.7 1.3 2.1 5 6 10 16 45.1 27.7 51.8 67.4
57 Thailand 2.9 1.9 1.3 54 64 83 103 34.2 21.6 61.4 68.2
58 Botswana 3.5 3.4 2.6 1 2 2 4 47.4 25.3 48.8 68.9
59 Cameroon 2.7 3.2 3.2 11 16 33 75 46.6 38.3 49.7 58.3
60 Jamaica 1.3 1.5 0.5 2 3 3 4 34.8 21.1 58.0 68.2
61 Ecuador 3.1 2.7 2.2 10 13 19 26 40.5 24.0 55.3 68.0
62 Colombia 2.5 2.1 1.6 32 38 50 63 35.9 22.3 59.8 68.2
63 Paraguay 2.8 3.2 2.7 4 6 9 13 40.6 26.9 55.7 66.3
64 Tunisia 2.1 2.5 2.2 8 10 14 19 38.7 23.7 57.3 68.3
65 Turkey 2.5 2.3 2.0 54 68 91 120 35.4 23.2 60.3 67.7
66 Pens 2.8 2.2 2.1 21 26 37 49 39.1 23.7 57.2 68.3
67 Jordan 2.5 3.7 3.6 4 6 12 22 47.2 34.4 50.0 61.9
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
228
Hypothetical Age structure of
size of population (percent)
Average annual growth ofpopulation stationary
(percent) Population (millions) 0-14 years 15-64 years
population
1965-81) 1980-88 1988-2000 1988 2000k 2025 (millions) 1988 2025 1988 2025
68 Chile - 1.7 1.7 1.3 13 15 19 23 30.8 21.3 63.1 65.7
69 SyrianArabRep. 3.4 3.6 3.6 12 18 36 74 48.3 36.0 49.1 60.3
70 Costa Rica 2.7 2.3 2.0 3 3 5 6 36.4 22.1 59.7 66.2
71 Mexico 3.1 2.2 1.9 84 105 142 184 38.6 22.8 57.7 68.2
72 Maurjtjus 1.6 1.0 0.8 1 1 1 2 29.5 19.3 66.5 67.5
73 Poland 0.8 0.8 0.5 38 40 45 50 25.2 19.7 64.9 62.0
74 Malaysia 2.5 2.6 2.2 17 22 30 40 37.2 23.4 58.9 67.2
75 Panama 2.6 2.2 1.6 2 3 4 5 35.9 21.9 59.4 67.2
76 Brazil 2.4 2.2 1.8 144 178 236 303 35.7 22.8 59.8 66.9
77 Angola 2.8 2.5 3.0 9 14 27 69 45.0 40.6 52.6 56.3
78 Lebanon 1.7 . . . . . . . . . . . . . . . . .
79 Nicaragua 3.1 3.4 3.0 4 5 9 14 46.1 28.4 51.0 66.4
Upper-middle-income 2.0 w 1.8 w 1.7 w 3261 402 t 5691 33.4w 26.0 w 60.5 w 64.0 w
80 South Africa 2.4 2.3 2.3 34 45 65 96 38.2 25.3 58.1 67.1
81 Algeria 3.1 3.1 2.9 24 33 52 78 44.4 25.7 52.2 68.4
82 Hungary 0.4 -0.1 -0.2 11 10 10 10 19.9 17.4 66.7 61.7
83 Uruguay 0.4 0.6 0.6 3 3 4 4 26.2 20.1 62.7 64.1
84 Argentina 1.6 1.4 1.1 32 36 44 53 30.1 21.5 61.0 65.0
85 Yugoslavia 0.9 0.7 0.6 24 25 28 30 23.5 18.7 67.6 62.4
86 Gabon 3.6 3.9 2.7 1 1 3 7 38.2 39.4 56.9 56.4
87 Venezuela 3.5 2.8 2.2 19 24 34 45 38.7 23.3 57.7 67.5
88 TnnidadandTobago 1.1 1.7 1.4 1 1 2 2 33.1 22.1 61.6 65.6
89 Korea, Rep. of 2.0 1.2 0.9 42 47 54 56 27.3 18.0 67.9 66.0
90 Portugal 0.4 0.7 0.4 10 11 11 11 21.8 16.4 65.6 63.5
91 Greece 0.7 0.5 0.2 10 10 10 10 19.9 16.0 66.3 60.9
92 Oman 3.7 4.7 3.9 1 2 5 II 45.5 38.1 52.0 57.3
93 Libya 4.3 4.3 3.6 4 6 14 35 46.0 39.5 51.2 56.7
94 Iran,IslamicRep. 3.1 3.0 3.1 49 70 129 247 43.7 34.0 53.5 61.3
95 Iraq 3.4 3.6 3.4 18 26 49 90 46.7 33.1 50.6 62.5
96 Ro,nania 1.1 0.4 0.5 23 25 27 31 23.8 20.0 66.2 63.6
Low- and middle-income 2.3 w 2.0w 1.9 w 3,952 4,961 7,123 36.0w 26.8 w 59.5 w 65.2 w
Sub-Saharan Africa 2.7 w 3.2 w 3.1 w 464t 672 1 1,310 46.9 w 38.0w 50.2 w 58.8 w
East Asia 2.3 w 1.5w 1.4w 1,538 1,824t 2,293 1 30.3 w 21.3 w 64.6 w 67.0 w
South Asia 2.4w 2.3 w 2.0w 1,107 1,401 I 1,987 1 38.7w 26.1 w 57.2 w 67.2 w
Europe, M.East, & N.Africa 2.0 w 2.1 w 2.1 w 396 t 505 t 769 1 36.5 w 28.7 w 57.9 w 63.4w
Latin America & Caribbean 2.5 w 2.2 w 1.8w 414 I 514 1 699 36.9w 23.5 w 58.6 w 67.1 w
Severely indebted 2.4w 2.1 w 1.8w 4961 614 t 844 t 36.5w 24.5 w 58.5 w 66.2w
High-income economies 0.9 w 0.7 w 0.6 w 783 1 840 t 9181 20.5 w 17.9w 66.8 w 60.6 w
OECD members 0.8 w 0.6w 0.5 w 751 1 796 1 847 1 19.9w 16.8 w 67.1 w 60.6 w
tOther 3.5 w 2.9w 2.5 w 32 1 43t 711 35.1 w 30.3 w 59.9w 60.2 w
97 tSaudi Arabia 4.7 4.2 . . 14 . . . . . . 45.2 . . 52.4
98 Spain 1.0 0.5 0.4 39 41 43 41 21.3 16.2 65.9 63.0
99 Ireland 1.2 0.5 0.3 4 4 4 5 27.7 19.8 61.2 64.9
100 1Israel 2.8 1.7 1.7 4 5 7 9 32.0 21.3 59.9 65.2
101 tSingapore 1.6 1.1 1.0 3 3 3 4 24.0 18.0 70.8 61.5
102 tHong Kong 2.0 1.5 0.9 6 6 7 6 21.9 16.0 69.2 61.3
103 New Zealand 1.3 0.8 0.7 3 4 4 4 23.6 18.3 67.1 62.8
104 Australia 1.8 1.4 1.4 17 20 23 24 22.6 18.1 66.6 63.0
105 United Kingdom 0.2 0.2 0.3 57 59 61 61 19.0 17.4 65.5 61.2
106 Italy 0.5 0.2 0.1 57 58 56 46 17.5 14.1 68.4 60.5
107 tKuwait 7.1 4.4 2.8 2 3 4 5 37.6 21.3 59.2 65.8
108 Belgium 0.3 0.0 0.0 10 10 10 9 18.4 16.0 67.2 60.1
109 Netherlands 0.9 0.5 0.5 15 16 16 14 18.0 15.6 68.9 59.7
110 Austria 0.3 0.0 0.1 8 8 8 7 18.0 15.4 67.1 60.4
111 tUnitedArabEmirates 16.5 4.8 2.3 1 2 3 3 31.0 22.1 67.0 61.2
112 France 0.7 0.4 0.4 56 59 63 63 20.5 17.3 66.0 60.6
113 Canada 1.3 0.9 0.9 26 29 32 32 21.3 16.9 68.4 60.7
114 Denmark 0.5 0.0 0.0 5 5 5 5 17.6 15.7 67.1 60.2
115 Germany, Fed. Rep. 0.3 -0.1 0.0 61 61 58 50 15.2 15.0 69.6 59.0
116 Finland 0.3 0.5 0.2 5 5 5 5 19.4 16.4 67.8 58.9
117 Sweden 0.5 0.2 0.4 8 9 9 9 17.6 17.6 64.7 59.3
118 United States 1.0 1.0 0.8 246 270 307 316 21.6 18.0 66.0 61.2
119 Norway 0.6 0.3 0.4 4 4 5 5 19.6 17.1 64.5 61.0
120 Japan 1.2 0.6 0.4 123 129 131 121 19.6 15.7 68.9 58.8
121 Switzerland 0.5 0.3 0.4 7 7 7 6 17.0 16.1 68.2 58.4
Total reporting economies 2.1 w 1.8 w 1.7w 4,735 1 5,8011 8,041 1 33.4w 25.7w 60.7w 64.6w
Oil exporters 2.7w 2.7w 2.4w 5921 790 I 1,261 1 41.4w 29.6w 55.1 w 64.2w
Nonreporting nonmembers 1.0w 0.9w 0.6w 3651 3921 4501 25.5w 20.1 w 64.1 w 63.4w
a. For the assumptions used in the pmjections, see the technical notes.
229
Table 27. Demography and fertifity
Crude birth Crude death Women of Married women of
rate per rate per childbearing age Assumed year childbearing age
thousand thousand as a percentage
of reaching net using contraception5
population population ofpopulation Total fertility rate (percent)
reproduction
1965 1988 1965 1988 1965 1988 1965 1988 206(? rote of] 1986
Low-income economies 42w 31w 16w lOw 46w 50w 6.3w 4.0w 3.3w
China and India 41w 26w 14w 9w 46w 53w 6.3w 3.1w 2.5w
Other low-income 46w 41w 21w 13w 46w 46w 6.4w 5.6w 4.7w
1 Mozambique 49 45 27 17 47 45 6.8 6.3 6.1 2045
2 Ethiopia 43 51 20 20 46 43 5.8 7.5 7.2 2055
3 Chad 45 44 28 19 47 46 6.0 5.9 6.0 2045
4 Tanzania 49 48 22 13 45 43 6.6 6.7 5.8 2040
5 Bangladesh 47 40 21 15 44 46 6.8 5.5 4.3 2025
6 Malawi 56 54 26 20 46 45 7.8 7.6 7.3 2055
7 Somalia 50 49 26 19 45 44 6.7 6.8 6.5 2050
8 Zaire 47 45 21 14 46 45 6.0 6.0 5.4 2040
9 Bhutan 42 39 23 17 48 48 6.0 5.5 5.3 2035
10 LaoPDR 45 47 23 17 47 46 6.1 6.6 5.4 2035
11 Nepal 46 42 24 15 50 47 6.0 5.8 4.6 2030 15
12 Madagascar 47 43 22 16 47 44 6.6 6.0 4.8 2030
13 BurkinaFaso 48 47 26 18 47 45 6.4 6.5 6.2 2045
14 Mali 50 50 27 19 46 45 6.5 7.0 6.9 2050 5
15 Burundi 47 47 24 16 48 45 6.4 6.8 6.0 2045 9
16 Uganda 49 51 19 17 44 43 7.0 7.3 6.5 2050 5
17 Nigeria 51 47 23 15 45 43 6.9 6.6 5.6 2040
18 Zambia 49 50 20 13 46 44 6.6 6.7 6.0 2045
19 Niger 48 51 29 20 43 44 6.8 7.1 7.3 2060
20 Rwanda 52 53 17 18 45 43 7.5 8.0 7.2 2055
21 China 38 21 10 7 44 55 6.4 2.4 2.1 2000 74
22 India 45 32 20 11 47 49 6.2 4.2 3.0 2015 35
23 Pakistan 48 46 21 13 43 46 7.0 6.6 5.4 2040 11
24 Kenya 52 47 20 11 40 41 8.0 6.9 5.3 2035 17
25 Togo 50 50 22 14 46 44 6.5 6.7 5.5 2040
26 CentralAfricanRep. 34 42 24 15 47 46 4.5 5.7 5.2 2035
27 Haiti 43 35 20 13 47 49 6.2 4.7 3.8 2025 5
28 Benin 49 47 24 15 44 44 6.8 6.4 5.2 2035
29 Ghana 47 45 18 13 45 44 6.8 6.3 5.1 2035 13
30 Lesotho 42 41 18 13 47 46 5.8 5.7 4.5 2030
31 SriLanka 33 22 8 6 47 53 4.9 2.5 2.1 1995 62
32 Guinea 46 48 29 22 47 46 5.9 6.5 6.5 2050
33 Yemen, PDR 50 48 26 15 45 46 7.0 6.6 5.4 2035
34 Indonesia 43 28 20 9 47 51 5.5 3.4 2.5 2005 45
35 Mauritania 47 48 26 19 46 45 6.5 6.5 6.5 2050
36 Sudan 47 44 24 16 46 45 6.7 6.4 5.4 2035
37 Afghanistan 53 . . 29 . . 49 . . 7.1 . . .
38 Myanmar 40 30 18 10 46 49 5.8 3.9 2.8 2010
39 Kainpuchea, Dem. 44 . . 20 . . 47 . . 6.2 . . . . .
40 Liberia 46 45 20 16 46 44 6.4 6.4 5.2 2035 6
41 SierraLeone 48 48 31 23 47 46 6.4 6.5 6.5 2050
42 VietNam 31 7 48 4.0 2.8 2010 58
Middle-income economies 38w 29w 13w 8w 45w 49w 5.6w 3.8w 3.1w
Lower-middle-income 41w 30w 13w 8w 44w 49w 6.1w 3.9w 3.1w
43 Bolivia 46 42 21 14 46 46 6.6 6.0 4.8 2030
44 Philippines 42 31 12 7 44 49 6.8 3.8 2.7 2010 44
45 Yemen Arab Rep. 49 54 27 20 47 43 7.0 8.0 7.2 2055
46 Senegal 47 45 23 16 46 44 6.4 6.5 6.2 2045 12
47 Zimbabwe 55 38 17 8 42 46 8.0 5.3 3.4 2015 43
48 Egypt, Arab Rep. 43 34 19 9 43 47 6.8 4.5 3.5 2020 38
49 Dominican Rep. 47 31 13 7 43 51 7.0 3.7 2.7 2010 50
50 Côted'Ivoire 52 50 22 14 44 42 7.4 7.3 6.4 2050
51 PapuaNewGuinea 43 36 20 12 47 47 6.2 5.2 4.0 2020
52 Morocco 49 35 18 9 45 48 7.1 4.7 3.5 2020 36
53 Honduras 51 39 17 8 44 45 7.4 5.5 4.2 2025 41
54 Guatemala 46 40 17 8 44 44 6.7 5.7 4.5 2030 23
55 Congo,People'sRep. 42 47 18 15 47 43 5.7 6.5 6.3 2050
56 El Salvador 46 36 13 8 44 45 6.7 4.8 3.8 2025 47
57 Thailand 41 22 10 7 44 53 6.3 2.5 2.1 1995 66
58 Botswana 53 38 19 7 45 44 6.9 5.1 3.1 2015 33
59 Cameroon 40 45 20 13 46 42 5.2 6.5 5.8 2045
60 Jamaica 38 23 9 6 42 50 5.7 2.6 2.1 1995
61 Ecuador 45 32 13 7 43 48 6.8 4.2 3.1 2015 44
62 Colombia 43 26 11 6 43 52 6.5 3.1 2.2 2000 65
63 Paraguay 41 35 8 6 41 49 6.6 4.5 3.7 2025 38
64 Tunisia 44 31 16 7 43 49 7.0 4.1 2.9 2010 50
65 Turkey 41 30 15 8 44 50 5.8 3.7 2.8 2010 77
66 Peru 45 31 16 9 44 49 6.7 4.0 2.9 2010 46
67 Jordan 53 42 21 6 45 43 8.0 6.4 5.2 2035
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
230
Crude birth Crude death Women of
Married women of
rate per rate per childbearing age Assumed year childbearing age
thousand thousand as a percentage
of reaching net using contraceptionb
population population ofpopulation Total fertility rate
reproduction (percent)
1965 1988 1965 1988 1965 1988 1965 1988 2000' rate ofl 1986
68 Chile 34 23 II 6 45 53 4.8 2.7 2.1 2000
69 SyrianArabRep. 48 45 16 7 41 43 7.7 6.7 5.5 2040
70 Costa Rica 45 27 8 4 42 52 6.3 3.2 2.4 2005 69
71 Mexico 45 28 11 6 43 50 6.7 3.5 2.4 2005 53
72 Mauritius 36 19 8 7 45 56 5.0 2.0 1.8 2030 78
73 Poland 17 16 7 10 47 48 2.5 2.2 2.1 1990
74 Malaysia 40 30 12 5 43 51 6.3 3.7 2.8 2010 51
75 Panama 40 26 9 5 44 51 5.7 3.1 2.2 2000 58
76 Brazil 39 28 11 8 45 51 5.6 3.4 2.4 2005 66
77 Angola 49 47 29 20 47 45 6.4 6.4 6.5 2050
78 Lebanon 40 . . 12 . . 42 . . 6.2 . . .
79 Nicaragua 49 41 16 7 43 45 7.2 5.4 4.2 2025
Upper-middle-income 31 w 26w 12 w 8w 47 w 49w 4.5 w 3.5 w 3.1 w
80 South Africa 40 35 16 10 46 49 6.1 4.4 3.5 2020
81 Algeria 50 37 18 8 44 45 7.4 5.4 3.8 2020
82 Hungaly 13 12 II 13 48 47 1.8 1.8 1.8 2030
83 Umguay 21 17 10 10 49 46 2.8 2.4 2.1 1995
84 Argentina 23 21 9 9 50 47 3.1 2.9 2.3 2005
85 Yugoslavia 21 15 9 9 50 49 2.7 2.0 2.0 2030
86 Gabon 31 42 22 16 48 47 4.1 5.5 6.0 2045
87 Venezuela 42 30 8 5 44 50 6.1 3.7 2.7 2010
88 TnnidadandTobago 33 26 8 6 46 53 4.3 2.9 2.3 2005 53
89 Korea,Rep.of 35 16 II 6 46 56 4.9 1.8 1.8 2030 70
90 Portugal 23 12 10 9 48 48 3.1 1.6 1.6 2030
91 Greece 18 11 8 10 51 47 2.3 1.6 1.6 2030
92 Oman 50 45 24 7 46 43 7.2 7.1 5.9 2045
93 Libya 49 44 17 9 45 44 7.4 6.8 5.9 2050
94 Iran, IslamicRep. 46 41 18 8 42 47 7.1 5.6 4.7 2040
95 Iraq 49 42 18 8 45 44 7.2 6.3 5.1 2035
96 Ro,nania 15 16 9 11 50 48 1.9 2.1 2.1 1985
Low- and middle-income 41 w 30 w 15w 10w 46w 50 w 6.1 w 3.9 w 3.3 w
Sub-Saharan Africa 48 w 47 w 22 w 16 w 45 w 44w 6.6 w 6.7 w 5.9w
East Asia 39 w 23 w 11w 7w 45 w 54 w 6.2 w 2.7 w 2.2 w
South Asia 45 w 34 w 20 w 12w 47 w 48 w 6,3 w 4.5 w 3.4w
Europe, M.East, & N.Africa 35 w 31w 15w lOw 46 w 47 w 5.1 w 4.3 w 3.7 w
Latin America & Caribbean 40 w 28 w 12w 7w 45 w 50 w 5.8w 3.6w 2.6 w
Severely indebted 37 w 28 w 12 w 8w 45 w 49w 5.4w 3.6 w 2.8 w
High-income economies 19w 14 w 10w 9w 47 w SOw 2.8w 1.8w 1.9w
OECDmembers 19w 13w lOw 9w 47w 50w 2.7w 1.7w 1.8w
tOther 35 w 29w 11 w 6w 45 w 48 w 5.4 w 4.5 w 4.0w
97 tSaudi Arabia 48 42 20 8 44 42 7.3 7.1 .
98 Spain 21 12 8 9 49 48 2.9 1.6 1.6 2030
99 Ireland 22 15 12 9 42 48 4.0 2.3 2.1 1990
100 1Israel 26 22 6 7 46 48 3.8 3.0 2.3 2005
101 tSingapore 31 18 6 5 45 59 4.7 1.9 1.9 2030
102 tHong Kong 27 14 6 5 45 55 4.7 1.6 1.6 2030 72
103 New Zealand 23 16 9 8 45 52 3.6 1.9 1.9 2030
104 Australia 20 15 9 7 47 52 3.0 1.9 1.9 2030
105 United Kingdom 18 14 12 11 45 48 2.9 1.8 1.8 2030
106 Italy 19 10 10 9 48 49 2.7 1.3 1.4 2030
107 tKuwait 48 26 7 3 45 50 7.4 3.7 2.6 2010
108 Belgium 17 12 12 11 44 48 2.6 1.6 1.6 2030
109 Netherlands 20 13 8 8 47 52 3.0 1.6 1.6 2030 ½
110 Austria 18 12 13 11 43 48 2.7 1.5 1.6 2030
111 tUnitedArabEmitates 41 23 14 4 47 47 6.8 4.7 3.7 2020
112 France 18 14 11 9 43 48 2.8 1.8 1.8 2030
113 Canada 21 14 8 7 47 53 3.1 1.7 1.7 2030 73
114 Denmark 18 12 10 12 47 50 2.6 1.5 1.6 2030
115 Germany, Fed. Rep. 18 11 12 11 45 48 2.5 1.5 1.5 2030 78
116 Finland 17 13 10 10 48 49 2.4 1.7 1.7 2030
117 Sweden 16 13 10 12 47 47 2.4 2.0 2.0 2030
118 United States 19 16 9 9 45 51 2.9 1.9 1.9 2030
119 Norway 18 14 10 11 45 48 2.9 1.8 1.8 2030
120 Japan 19 11 7 7 56 50 2.0 1.7 1.7 2030 I;4
121 Switzerland 19 12 10 9 48 50 2.6 1.6 1.7 2030
Total reporting economies 36w 28w 14w lOw 46w 50w 5.4w 3.6w 3.1w
Oil exporters 45 w 35 w 18 w 9w 45 w 48w 6.4w 4.7 w 3.8 w
Nonreporting nonmembers 20w 18w 8w 10 w 47 w 47 w 2.7w 2.3 w 2.1 w
a. For assumptions used in the projections, see the technical note to Table 26. b. Figures include women whose husbands practice contraception; see the technical
noteu.
231
Table 28. Health and nutrition
Births attended Babies with low Infant nwrtaltty
Population per
by health staff birth weight rate (per thousand Daily calorie supply
Physician Nursing person live births) (per capita)
(percent) (percent)
1965 1984 1965 1984 1985 1985 1965 1988 1965 1986
Low-income economies 9,760 w 5,580 w 6,010 w 2,200 w 124 w 72 w 1,993 w 2,384 w
China and India 2,930w 1,640w 4,420 w 1,710 w 114 w 59w 2,001 w 2,463 w
Other low-income 28,080 w 13,910w 10,170 w 3,250 w 149 w 98 w 1,976 w 2,226 w
1 Mozambique 18,000 37,960 5,370 5,760 28 15 179 139 1,979 1,595
2 Ethiopia 70,190 78,970 5,970 5,400 58 165 135 1,824 1,749
3 Chad 72,480 38,360 13,610 3,390 11 183 130 2,399 1,717
4 Tanzania 21,700 26,200 2,100 8,130 74 14 138 104 1,832 2,192
5 Bangladesh 8,100 6,730 8,980 31 144 118 1,971 1,927
6 Malawi 47,320 11,330 40,980 3,110 59 10 200 149 2,244 2,310
7 Somalia 36,840 16,080 3,950 1,530 2 165 130 2,167 2,138
8 Zaire 34,740 141 96 2,187 2,163
9 Bhutan 23,310 2,990 3 .. 171 127
10 Lao PDR 24,320 1,360 4,880 530 39 148 108 1,956 2,391
11 Nepal 46,180 32,710 87,650 4,680 10 171 126 1,901 2,052
12 Madagascar 10,620 9,780 3,650 62 10 201 119 2,462 2,440
13 BurkinaFaso 73,960 57,220 4,150 1,680 18 193 137 2,009 2,139
14 Mali 51,510 25,390 3,360 1,350 27 17 207 168 1,858 2,073
15 Bunindi 55,910 21,120 7,320 3,040 12 14 142 73 2,391 2,343
16 Uganda 11,110 21,900 3,130 2,060 . 10 121 101 2,360 2,344
17 Nigeria 29,530 7,990 6,160 1,020 . 25 177 103 2,185 2,146
18 Zambia 11,380 7,150 5,820 740 14 121 78
19 Niger 65,540 39,730 6,210 460 47 20 180 133 1,994 2,432
20 Rwanda 72,480 34,680 7,450 3,650 17 141 120 1,665 1,830
21 China 1,600 1,000 3,000 1,710 6 90 31 1,926 2,630
22 India 4,880 2,520 6,500 1,700 33 30 150 97 2,111 2,238
23 Pakistan 2,910 9,910 4,900 24 25 149 107 1,761 2,315
24 Kenya 13,280 9,970 1,930 950 13 112 70 2,289 2,060
25 Togo 23,240 8,700 4,990 1,240 20 156 92 2,378 2,207
26 CentralAfricanRep. 34,020 23,530 3,000 2,210 15 157 102 2,135 1,949
27 Haiti 14,000 7,180 12,890 2,290 20 17 178 116 2,000 1,902
28 Benin 32,390 15,940 2,540 1,750 34 10 166 115 2,009 2,184
29 Ghana 13,740 14,890 3,730 640 73 17 120 88 1,950 1,759
30 Lesotho 20,060 18,610 4,700 28 10 142 98 2,065 2,303
31 SriLanka 5,820 5,520 3,220 1,290 87 28 63 21 2,153 2,400
32 Guinea 47,050 46,420 4,110 5,160 18 191 143 1,923 1,776
33 Yemen, PDR 12,870 4,370 1,850 1,060 10 13 197 118 1,982 2,298
34 Indonesia 31,700 9,460 9,490 1,260 43 14 128 68 1,800 2,579
35 Mauritania 36,530 12,120 . . 1,200 23 10 178 125 2,064 2,322
36 Sudan 23,500 10,100 3,360 1,250 20 15 160 106 1,938 2,208
37 Afghanistan 15,770 . . 24,430 . . . . . . 206 . 2,294
38 Myanmar 11,860 3,740 11,370 900 97 16 125 68 1,917 2,609
39 Kampuchea,Dem. 22,410 . . 3,670 . . . . 134 . . 2,276
40 Liberia 12,560 9,340 2,330 1,370 89 . 176 130 2,154 2,381
41 SierraLeone 16,840 13,630 4,470 1,090 25 14 209 152 1,837 1,854
42 VietNam . . 1,000 . . 620 18 . . 44 . . 2,297
Middle-income economies 4,060 w 2,520 w 2,190w 980w 98 w 52 w 2,458 w 2,846w
Lower-middle-income 4,910 w 3,030 w 2,190 w 1,090w 107 w 57 w 2,378 w 2,733 w
43 Bolivia 3,300 1,540 3,990 2,480 36 15 160 108 1,869 2,143
44 Philippines . . 6,700 1,140 2,740 . . 18 72 44 1,924 2,372
45 YemenArabRep. 56,150 6,010 . 2,580 12 9 197 128 2,008 2,318
46 Senegal 19,490 13,060 2,440 2,030 . 10 126 78 2,479 2,350
47 Zimbabwe 8,010 6,700 990 1,000 69 15 103 49 2,105 2,132
48 Egypt, Arab Rep. 2,300 770 2,030 780 24 7 172 83 2,400 3,342
49 DominicanRep. 1,700 1,760 1,640 1,210 57 16 110 63 1,872 2,477
50 Côted'Ivoire 20,640 . . 2,000 . . 20 14 149 95 2,359 2,562
51 PapuaNewGuinea 12,640 6,070 620 880 34 25 143 61 1,905 2,205
52 Momcco 12,120 15,580 2,290 920 9 145 71 2,167 2,915
53 Honduras 5,370 1,510 1,530 670 50 20 128 68 1,963 2,068
54 Guatemala 3,690 2,180 8,250 850 19 10 112 57 2,027 2,307
55 Congo, People's Rep. 14,210 8,320 950 590 . . 12 129 117 2,259 2,619
56 El Salvador . . 2,830 1,300 930 35 15 120 57 1,859 2,160
57 Thailand 7,160 6,290 4,970 710 33 12 88 30 2,101 2,331
58 Botswana 27,450 6,900 17,710 700 52 8 112 41 2,019 2,201
59 Camemon 26,720 . . 5,830 . . . . 13 143 92 2,079 2,028
60 Jamaica 1,990 2,040 340 490 89 8 49 11 2,231 2,590
61 Ecuador 3,000 820 2,320 610 27 10 112 62 1,940 2,058
62 Colombia 2,500 1,240 890 660 51 15 86 39 2,174 2,542
63 Paraguay 1,850 1,460 1,550 1,000 22 6 73 41 2,627 2,853
64 Tunisia 8,000 2,150 370 60 7 145 48 2,201 2,994
65 Turkey 2,900 1,380 . . 1,030 78 7 165 75 2,659 3,229
66 Pens 1,650 1,040 900 . . 55 9 130 86 2,325 2,246
67 Jordan 4,690 1,140 1,800 1,300 75 7 114 43 2,314 2,991
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
232
Infant mortality
Population per Births attended Babies with low
rate (per thousand Daily calone supply
by health staff birth weight
Physician Nursing person live births) (per capita)
(percent) (percent)
1965 1984 1965 1984 1985 1985 1965 1988 1965 1986
68 Chile 2,120 1,230 600 370 97 7 101 20 2,592 2,579
69 SyrianArabRep. 5,400 1,260 1,440 37 9 114 46 2,195 3,260
70 Costa Rica 2,010 960 630 450 93 9 72 18 2,366 2,803
71 Mexico 2,080 1,240 980 880 15 82 46 2,644 3,132
72 Mauritius 3,930 1,900 2,030 580 90 9 65 22 2,271 2,748
73 Poland 800 490 410 190 8 42 16 3,229 3,336
74 Malaysia 6,200 1,930 1,320 1,010 82 9 55 23 2,247 2,730
75 Panama 2,130 980 1,600 390 83 8 56 22 2,255 2,446
76 Brazil 2,500 1,080 3,100 1,210 73 8 104 61 2,402 2,656
77 Angola 13,150 17,790 3,820 1,020 15 17 192 135 1,897 1,880
78 Lebanon 1,010 2,030 . 56 . 2,489
79 Nicaragua 2,560 1,500 1,390 530 15 121 60 2,398 2,495
Upper-middle-income 2,380w 1,220w 2,190w 680w 82w 42w 2,629w 3,117w
80 South Africa 2,050 . . 490 . . 12 124 70 2,623 2,924
81 Algeria 8,590 2,340 11,770 330 . . 9 154 72 1,681 2,715
82 Hungaiy 630 310 240 170 99 10 39 16 3,171 3,569
83 Uruguay 880 520 590 . . 8 47 23 2,811 2,648
84 Argentina 600 370 610 980 6 58 31 3,210 3,210
85 Yugoslavia 1,200 550 850 260 . . 7 72 25 3,289 3,563
86 Gabon .
. 2,790 760 270 92 16 153 101 1,881 2,521
87 Venezuela 1,210 700 560 . . 82 9 65 35 2,321 2,494
88 TnnidadandTobago 3,810 950 560 260 90 . . 42 16 2,497 3,082
89 Korea, Rep. of 2,680 1,160 2,970 580 65 9 62 24 2,256 2,907
90 Portugal 1,240 410 1,160 . . . . 8 65 14 2,517 3,151
91 Greece 710 350 600 450 . 6 34 12 3,049 3,688
92 Oman 23,790 1,700 6,420 760 60 14 194 38 .
93 Libya 3,860 690 850 350 76 5 138 80 1,925 3,601
94 Iran,IslamicRep. 3,890 2,690 4,270 1,050 . . 9 152 64 2,204 3,313
95 Iraq 5,000 1,740 2,910 1,660 50 9 119 68 2,150 2,932
96 Roenania 760 570 400 280 99 6 44 24 2,978 3,373
Low- and middle-income 8,270w 4,790 w 5,020 w 1,900w 117w 67 w 2,116w 2,507w
Sub-Saharan Africa 33,200w 23,850w 5,420w 2,460w 160w 108w 2,092w 2,096w
East Asia 5,600w 2,390w 4,050w 1,570 w 92 w 37 w 1,937 w 2,594w
South Asia 6,220 w 3,570w 8,380 w 2,710 w 147 w 99w 2,060w 2,228 w
Europe, M.East, & N.Africa 4,760w 2,430 w 3,440w 1,160w 115 w 62 w 2,610w 3,177 w
Latin America & Caribbean 2,370w 1,230w 2,090w 1,020w 95 w 53 w 2,457w 2,700 w
Severely indebted 3,010w 2,560w 1,620w 1,180w 88 w 51 w 2,537w 2,768 w
High-income economies 940 w 470w 470w 140w 25 w 9w 3,083 w 3,376 w
OECD members 870 w 450 w 420w 130w 24w 8w 3,100w 3,390 w
tOther 4,430w 800 w 2,510w 260w 69w 36w 2,323 w 3,001 w
97 fSaudi Arabia 9,400 690 6,060 320 78 6 148 69 1,853 3,004
98 Spain 800 320 1,220 260 96 . 38 9 2,822 3,359
99 Ireland 950 680 170 140 . . 4 25 7 3,546 3,632
100 (Israel 400 350 300 110 99 7 27 11 2,784 3,061
101 tSingapore 1,900 1,310 600 100 7 26 7 2,297 2,840
102 tHong Kong 2,520 1,070 1,250 240 . . 4 27 7 2,504 2,859
103 New Zealand 820 580 570 80 99 5 20 11 3,237 3,463
104 Australia 720 440 150 110 99 6 19 9 3,118 3,326
105 United Kingdom 870 . . 200 120 98 7 20 9 3,353 3,256
106 Italy 1,850 230 790 . . 7 36 10 3,091 3,523
107 tKuwait 790 640 270 200 99 7 64 15 2,945 3,021
108 Belgium 700 330 590 110 100 5 24 9 .
109 Netherlands 860 450 270 170 . . 4 14 8 3,108 3,326
110 Austria 720 390 350 180 . . 6 28 8 3,231 3,428
111 tUnitedArabEmirates 1,020 390 96 108 25 2,705 3,733
112 France 830 320 380 110 . . 5 22 8 3,217 3,336
113 Canada 770 510 190 120 99 6 24 7 3,212 3,462
114 Denmark 740 400 190 60 6 19 8 3,395 3,633
115 Gennany,Fed.Rep. 640 380 500 230 . 5 24 8 3,102 3,528
116 Finland 1,300 440 180 60 . . 4 17 6 3,111 3,122
117 Sweden 910 390 310 100 100 4 13 6 2,888 3,064
118 United States 670 470 310 70 100 7 25 10 3,224 3,645
119 Norway 790 450 340 60 100 4 17 8 3,032 3,223
120 Japan 970 660 410 180 100 5 18 5 2,687 2,864
121 Switzerland 710 700 270 130 5 18 7 3,412 3,437
Total reporting economies 6,630w 4,070w 4,000w 1,600w 97 w 57 w 2,321 w 2,653 w
Oil exporters 17,920w 5,130 w 5,740w 1,010w 134w 71 w 2,128 w 2,737 w
Nonreporting nonmembers 540 w 380w 300w 170w 30w 24w 3,151 w 3,395w
233
Table 29. Education
Percentage of age group enrolled in education
Primary net Primary
Prtmary Secondary
Tertiary enrollment pupil-teacher
Total Female Total Female (total) (percent) ratio
196.5 1987 1965 1987 1965 1987 1965 1987 1965 1987 1975 1987 1965 1987
Low-income economies 73w 104w 95 w 20w 37w 29w 2w
China and India 83w 117w 106w 25w 41w 33w 2w
Other low-income 49w 76w 37w 68w 9w 26w 5w 16w 1w 3w
Mozambique
1 37 68 26 59 3 5 2 4 0 0 45
2 Ethiopia 11 37 6 28 2 15 1 12 0 1 27 41
3 Chad 34 51 13 29 1 6 0 2 0 38 83
4 Tanzania 32 66 25 66 2 4 1 3 0 0 50 52 33
5 Bangladesh 49 59 31 49 13 18 3 II 1 5 53 45 48
6 Malawi 44 66 32 59 2 4 1 3 0 1 49
7 Somalia 10 15 4 10 2 9 1 6 0 3 16 26
8 Zaire 70 76 45 64 5 22 2 13 0 2 37
9 Bhutan 7 24 1 17 0 4 1 0
10 LaoPDR 40 1]] 30 100 2 23 1 19 0 2 37
11 Nepal 20 82 4 5 26 2 1 5 ..
12 Madagascar 65 59 8 21 5 19 1 4 71
13 BurkinaFaso 12 32 8 24 1 6 1 4 0 1 27 47 65
14 Mali 24 23 16 17 4 6 2 4 0 1 18 46 38
15 Bunindi 26 67 15 50 1 4 1 3 0 1 40
16 Uganda 67 70 50 63 4 13 2 9 0 1
17 Nigeria 32 77 24 5 3 0 33
18 Zambia 53 97 46 92 7 17 3 13 1 51
19 Niger 11 29 7 20 1 6 0 3 1 42
20 Rwanda 53 67 43 66 2 6 1 5 0 0 64 67
21 China 89 132 124 24 43 37 1 2 98
22 India 74 98 57 81 27 39 13 27 5 42
23 Pakistan 40 52 20 35 12 19 5 11 2 5 42 41
24 Kenya 54 96 40 93 4 23 2 19 0 1 88 34
25 logo 55 101 32 78 5 24 2 12 0 3 73 50
26 CentralAfrican Rep. 56 66 28 51 2 12 1 6 1 49 54 63
27 Haiti 50 95 44 89 5 17 3 16 0
28 Benin 34 63 21 43 3 16 2 9 0 3 50 41 33
29 Ghana 69 71 57 63 13 40 7 32 1 2 32 24
30 Lesotho 94 113 114 125 4 4 0 57
31 SriLanka 93 104 86 102 35 66 35 69 2 4
32 Guinea 31 30 19 18 5 9 2 4 0 1 23
33 Yemen,PDR 23 10 11 5
34 Indonesia 72 118 65 115 12 46 7 1 72
35 Mauritania 13 52 6 42 1 16 0 9 3
36 Sudan 29 49 21 4] 4 20 2 17 1 2 48
37 Afghanistan 16 5 . . 2 1 . . 0 . 53
38 Myanmar 71 65 . . 15 . . 11 1 .
39 KampucheaDem. 77 . . 56 9 4 . . 1 . 48
40 Liberia 41 35 23 5 - . 3 . . 1 3 .
41 Sierra Leone 29 . 21 . . 5 . 3 . . 0 . . 32
42 VietNam 102 99 42 40 . .
Middle-income economies 92w 104w 86w 101w 26w 54w 22w 54w 6w 17w 36w 29w
Lower-middle-income 92w 104w 86w 101w 23w 49w 20w SOw 6w 16w 37w 29w
43 Bolivia 73 91 60 85 18 37 15 35 5 17 73 83 28 27
44 Philippines 113 106 111 106 41 68 40 69 19 38 95 . . 31
45 Yemen Arab Rep. 9 91 1 40 0 26 . . 6 2 22 . . 56
46 Senegal 40 60 29 49 7 15 3 10 1 3 50 43 54
47 Zimbabwe 110 136 92 132 6 45 5 36 0 4 . . . . .
48 Egypt,ArabRep. 75 90 60 79 26 69 15 58 7 20 39
49 Dominican Rep. 87 133 87 135 12 47 12 56 2 19 53
50 Côted'Ivoire 60 70 41 58 6 19 2 12 0 . . 47
51 PapuaNewGuinea 44 70 35 64 4 12 2 9 . . 2 . . . . 19 31
52 Morocco 57 71 35 56 11 37 5 30 1 10 47 57 39 26
53 Honduras 80 106 79 108 10 32 9 . . 1 9 . . . . .
54 Guatemala 50 77 45 70 8 21 7 . . 2 9 53 33 35
55 Congo,People'sRep. 114 . . 94 . . 10 . . 5 . . 1 7 . 60
56 ElSalvador 82 79 79 81 17 29 17 30 2 18 71 34 45
57 Thailand 78 95 74 14 28 II 2 20 35
58 Botswana 65 114 71 117 3 32 3 33 . . 3 58 40
59 Cameroon 94 109 75 100 5 26 2 20 0 3 69 47
60 Jamaica 109 105 106 106 51 65 50 67 3 4 90
61 Ecuador 91 117 88 116 17 56 16 57 3 30 78 37
62 Colombia 84 114 86 115 17 56 16 56 3 14 . . . . 36
63 Paraguay 102 102 96 99 13 30 13 30 4 . . 83 . . 30
64 Thnisia 91 116 65 107 16 40 9 34 2 6 95 56 31
65 Thrkey 101 117 83 113 16 46 9 34 4 10 84 46 31
66 Peni 99 122 90 120 25 65 21 61 8 25 36
67 Jordan 95 83 . . 38 23 2 . . 38
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
234
Percentage of age group enrolled in education
Primary net Primary
Primary Secondary
Tertiary enrollment pupil.teacher
Total Female Total Female (total) (percent) ratio
1965 1987 1965 1987 1965 1987 1965 1987 1965 1987 1975 1987 1965 1987
68 Chile 124 103 122 104 34 70 36 71 6 18 94 52
69 Syrian Arab Rep. 78 110 52 104 28 59 13 48 8 18 87 99 36 26
70 Costa Rica 106 98 lOS 97 24 41 25 43 6 25 92 85 27 31
71 Mexico 92 118 90 116 17 53 13 53 4 16 100 47 32
72 Mauritius 101 106 97 107 26 51 18 50 3 82 94 34 23
73 Poland 104 101 102 101 69 80 69 82 18 18 96 99 28 16
74 Malaysia 90 102 84 102 28 59 22 59 2 7 . . . . 22
75 Panama 102 106 99 104 34 59 36 63 7 28 87 . . 30
76 Brazil 108 103 108 . . 16 39 16 45 2 II 71 84 28
77 Angola 39 26 5 . . 4 . . 0 . . . .
78 Lebanon lflF 93 . . 26 . . 20 . 14 . . .
79 Nicaragua 69 99 69 104 14 43 13 58 2 8 65 76 34 32
Upper-middle-income 92 w 104 w 86 w 101 w 32 w 67 w 27 w 65 w 7w 20 w 34 w 30 w
80 South Africa 90 . . 88 . . 15 . . 14 . . 4 . . . . . . .
81 Algeria 68 96 53 81 7 55 5 46 I 9 77 88 43 29
82 Hungary 101 97 100 97 . . 70 . . 70 13 15 95 23 14
83 Umguay 106 110 106 109 44 73 46 . 8 42 .
84 Argentina 101 110 102 110 28 74 31 78 14 39 96 . . 20
85 Yugoslavia 106 95 103 94 65 80 59 79 13 19 31
86 Gabon 134 . . 122 . . II . . 5 . . . . 5 . . . . 39
87 Venezuela 94 107 94 107 27 54 28 59 7 27 81 . . 34
88 Trinidadand Tobago 93 100 90 100 36 82 34 85 2 4 87 88 34 24
89 Korea, Rep. of 101 101 99 101 35 88 25 86 6 36 99 99 62 37
90 Portugal 84 124 83 121 42 56 34 . . 5 12 91 32
91 Greece 110 104 109 104 49 90 41 89 10 26 97 . . 36
92 Oman . . 97 . . 92 . . 38 . . 29 . . 2 32 80 .
93 Libya 78 . . 44 . . 14 . . 4 . . 1 10 . . 31
94 lran,!slamicRep. 63 114 40 105 18 48 11 39 2 5 . . 94 32 29
95 Iraq 74 98 45 91 28 49 14 38 4 13 79 86 22 25
96 Roinania 101 97 100 98 39 79 32 80 10 10 23
Low- and middle-income 78 w 104 w 62 w 96 w 22 w 41 w 14 w 35 w 3w 8w
Sub-Saharan Africa 41 w 68 w 30 w 57 w 4w 17w 2w 12w Ow 1w 42 w
East Asia 88 w 125 w 120 w 23 w 45 w 40 w 1w 5w
South Asia 68 w 89 w w 73 w 24 w 35w 12 w 24 w 4w 42 w
Europe, M.East, & N.Africa 83 w 97 w 71 w 90w 32 w 57 w 26 w 50 w 7w i2 w 36 w
Latin America & Caribbean 98 w 108 w 96 w 110w 19 w 49 w 19 w 53w 4w 17w 34 w
Severely indebted 99w 104w 97w 103w 27w 53w 26w 55w 8w 19w 33w 27w
High-income economies 104w 102 w 105 w 103 w 62 w 93 w 59w 96w 21w 39w 25w
OECD members 104w 103 w 106w 103 w 63 w 94 w 61w 97w 21w 39w 25w
tOther 75 w 87 w 68 w 85 w 27 w 62 w 25w 59w 7w 17w 26w
97 (Saudi Arabia 24 71 II 65 4 44 1 35 1 13 42 22
98 Spain 115 113 114 113 38 102 29 107 6 30 100 . 34
99 Ireland 108 100 108 100 51 98 50 103 12 24 91
100 lisrael 95 95 95 97 48 83 51 87 20 34 . . 19
101 tSingapore 105 . . 100 45 . . 41 10 100 . . 29
102 tHong Kong 103 106 99 105 29 74 25 76 5 . . 92 . . 29
103 New Zealand 106 107 104 106 75 85 74 86 15 36 100 . . 22
104 Australia 99 106 99 105 62 98 61 99 16 29 98 98 28 17
105 United Kingdom 92 106 92 106 66 83 66 85 12 22 97
106 Italy 112 95 110 97 47 75 41 74 II 24 97 22
107 (Kuwait 116 94 103 92 52 82 43 79 . 17 68 79 23 12
108 Belgium 109 100 108 100 75 99 72 100 15 33 . . 83 21 10
109 Netherlands 104 115 104 116 61 104 57 103 17 31 92 31
110 Austria 106 101 105 101 52 80 52 81 9 29 89 . . 20 11
Ill fUnitedArabEmirates 99 100 60 . . 66 0 9 . . 88 . . 25
112 France 134 113 133 113 56 92 59 96 18 31 98 100 30
113 Canada 105 105 104 104 56 104 55 104 26 58 97 26
114 Denmark 98 99 99 99 83 107 67 107 14 30 II
115 Germany, Fed. Rep. . . 103 . . 101 . . 94 . . 92 9 30 . . . . 17
116 Finland 92 101 89 101 76 106 80 114 11 38 . 23
117 Sweden 95 100 96 . 62 91 60 92 13 31 100 100 20
118 United States 100 100 . . 100 . . 98 . . 99 40 60 72 . . 25
119 Norway 97 95 98 95 64 95 62 97 11 35 100 . 21
120 Japan 100 102 100 102 82 96 81 97 13 28 99 . 29
121 Switzerland 87 . . 87 37 . 35 8 24
Total reporting economies 83w 104w 70w 97w 28w 50w 23w 46w 7w 15w 35w
Oil exporters 68w 104w 59w 105w 15w 52w lOw 49w 2w 15w 37w
Nonreporting nonmembers 104 w 105 w 103 w 101 w 67 w 93 w 73 w 72 w 27 w 23 w
235
Table 30. Income distribution and ICP estimates of GDP
ICP estimates of
GDPper capita Percentage share of household income, by percentile group ofhouseholds5
(United States = 100) Lowest Second Third Fourth Highest Highest
1985 Year 20 percent quintile quintile quintile 20 percent 10 percent
Low-income economies
China and India
Other low-income
1Mozambique
2Ethiopia 1.6
3 Chad
4 Tanzania 2.6
5 Bangladeshc 5.0 1981-82 9.3 13.1 16.8 21.8 39.0 24.9
6 Malawi 3.6
7 Somalia
8 Zaire
9 Bhutan
10 Lao PDR
11 Nepal
12 Madagascar 3.9
13 Burkina Faso
14 Mali 2.4
15 Burundi
16 Uganda
17 Nigeria 7.2
18 Zambia 4.7
19 Niger
20 Rwanda 3.8
21 China
22 Indiac 4.5 1983 8.1 12.3 16.3 22.0 41.4 26.7
23 Pakistan'1 8.1 1984-85 7.8 11.2 15.0 20.6 45.6 31.3
24 Kenya 5.3
25 Togo
26 Central African Rep.
27 Haiti
28 Benin 6.5
29 Ghana's 1987 6.5 10.9 15.7 22.3 44.6 29.1
30 Lesotho
31 SriLanka 11.2 1985-86 4.8 8.5 12.1 18.4 56.1 43.0
32 Guinea
33 Yemen, PDR
34 Indonesia 1987 8.8 12.4 16.0 21.5 41.3 26.5
35 Mauritania
36 Sudan
37 Afghanistan
38 Myanmar
39 Kampuchea,Dem.
40 Liberia
41 Sierra Leone 3.0
42 VietNam
Middle-income economies
Lower-middle-income
43 Bolivia . . . . . . . . . . .
44 Philippines" 10.8 1985 5.5 9.7 14.8 22.0 48.0 32.1
45 Yemen Arab Rep. . . . . .
46 Senegal 7.0 . . . . .
47 Zimbabwe 9.9 .
48 Egypt, Arab Rep. 15.8 .
49 Dominican Rep. . . . . . . . . .
50 CÔtedIvoirec 10.2 1986 5.0 8.0 13.1 21.3 52.7 36.3
51 Papua New Guinea . . . . . . . . . . .
52 Momccod 13.1 1984-85 9.8 13.0 16.4 21.4 39.4 25.4
53 Honduras . . . . . . . . .
54 Guatemala . . 1979-81 5.5 8.6 12.2 18.7 55.0 40.8
55 Congo, People's Rep. 16.4
56 El Salvador . . . . .
57 Thailand 16.0
58 Botswana 16.1 1985-86 2.5 6.5 11.8 20.2 59.0 42.8
59 Cameroon 14.0 . . . . . . .
60 Jumaicac . 1988 5.4 9.9 14.4 21.2 49.2 33.4
61 Ecuador . . . . . . .
62 Colombiac . 1988 4.0 8.7 13.5 20.8 53.0 37.1
63 Paraguay .
64 Tunisia 19.8
65 Turkey 2t.8 . . . . . . . . . .
66 Peruc 1985 4.4 8.5 13.7 21.5 51.9 35.8
67 Jordan
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
236
sir esrunates oj
GDPper cita Percentage share of household income, by percentile group of householdsb
(United States = 190) Lowest Second Third Fourth Highest Highest
1985 Year 20 percent quintile quinlile quintile 20 percent 10 percent
68 Chile
69 Syrian Arab Rep. . . . . . . . . .
70 Costa Ricae . . 1986 3.3 8.3 13.2 20.7 54.5 38.8
71 Mexico
72 Mauritius 24.8 .
73 Polande 24.5 1987 9.7 14.2 18.0 22.9 35.2 21.0
74 Malaysia 1987 4.6 9.3 13.9 21.2 51.2 34.8
75 Panama
76 Brazil 1983 2.4 5.7 10.7 18.6 62.6 46.2
77 Angola
78 Lebanon
79 Nicaragua
Upper-middle-income
80 South Africa
81 Algeria
82 Hungalyc 31.2 1983 10.9 15.3 18.7 22.8 32.4 18.7
83 Uruguay
84 Argentina
85 Yugoslaviae 29.2 1987 6.1 11.0 16.5 23.7 42.8 26.6
86 Gabon
87 Venezuela 1987 4.7 9.2 14.0 21.5 50.6 34.2
88 Trinidad and Tobago
89 Korea, Rep. of 24.1
90 Portugal 33.8
91 Greece 35.5
92 Oman
93 Libya
94 Iran, Islamic Rep. 27.9
95 Iraq
96 Romania
Low- and middle-income
Sub-Saharan Africa
East Asia
South Asia
Europe, M.East, & N,Africa
Latin America & Caribbean
Severely indebted
High-income economies
OECD members
tOther
97 tSaudi Arabia . . . . . . . . .
98 Spain 46.0 1980-81 6.9 12.5 17.3 23.2 40.0 24.5
99 Ireland 40.9 . . . . . . .
100 tlsrael 1979 6.0 12.1 17.8 24.5 39.6 23.5
101 tSingapore .
. 1982-83 5.1 9.9 14.6 21.4 48.9 33.5
102 tHongKong 61.7 1980 5.4 10.8 15.2 21.6 47.0 31.3
103 New Zealand 60.9 1981-82 5.1 10.8 16.2 23.2 44.7 28.7
104 Australia 71.1 1985 4.4 11.1 17.5 24.8 42.2 25.8
105 UnitedKingdom 66.1 1979 5.8 11.5 18.2 25.0 39.5 23.3
106 Italy 65.6 1986 6.8 12.0 16.7 23.5 41.0 25.3
107 fKuwait . . . . . . . . . . .
108 Belgium 64.7 1978-79 7.9 13.7 18.6 23.8 36.0 21.5
109 Netherlands 68.2 1983 6.9 13.2 17.9 23.7 38.3 23.0
110 Austria 66.1 . . . . . .
Ill tUnited Arab Emirates . S
112 France 69.3 1979 6.3 12.1 17.2 23.5 40.8 25.5
113 Canada 92.5 1987 5.7 11.8 17.7 24.6 40.2 24.1
114 Denmark 74.2 1981 5.4 12.0 18.4 25.6 38.6 22.3
115 Germany,Fed.Rep. 73.8 1984 6.8 12.7 17.8 24.1 38.7 23.4
116 Finland 69.5 1981 6.3 12.1 18.4 25.5 37.6 21.7
117 Sweden 76.9 1981 8.0 13.2 17.4 24.5 36.9 20.8
118 UnitedStates 100.0 1985 4.7 11.0 17.4 25.0 41.9 25.0
119 Norway 84.4 1979 6.2 12.8 18.9 25.3 36.7 21.2
120 Japan 71.5 1979 8.7 13.2 17.5 23.1 37.5 22.4
121 Switzeriand . . 1982 5.2 11.7 16.4 22.1 44.6 29.8
Total reporting economies
Oil exporters
Nonreporting nonmembers
a. ICP refers to the United Nations' International Comparison Program; data are preliminary Phase V results. b. These estimates should be treated with caution; see the
technical notes for details of different distribution measures, as in the following footnotes. c. Per capita expenditure. d. Household expenditure. e. Per capita
income.
237
Table 31. Urbanization
Urban population Number of
Percentage of urban population cities of
As a percentage Average annual
of total growth rate In cities of over over 500,000
population (percent) In largest city 500 000 persons persons
1965 1988 1965 -80 1980-88 1960 1980 1960 1980 1960 1980
Low-income economies 17w 35w 3.5 w 11 w 13 w 30w 43 w 59 t 165
China and India 18w 40w 3.0w .. 6w 6w 36w 43w 49t 114t
Other low-income 14w 25 w 4.9w 5.2w 24w 28w 17w 43 w lot 51
1 Mozambique 5 24 10.2 11.0 75 83 0 83 0 1
2 Ethiopia 8 13 4.9 5.2 30 37 0 37 0 1
3 Chad 9 31 7.8 7.4 . . 39 0 0 0 0
4 Tanzania 5 30 11.7 11.6 34 50 0 50 0 1
5 Bangladesh 6 13 6.4 5.6 20 30 20 51 1 3
6 Malawi 5 14 7.5 7.9 19 0 0 0 0
7 Somalia 20 37 5.5 5.6 . . 34 0 0 0 0
8 Zaire 26 39 4.6 4.6 14 28 14 38 1 2
9 Bhutan 3 5 3.9 4.9 . . . . 0 0 0 0
10 La0PDR 8 18 5.3 6.1 69 48 0 0 0 0
11 Nepal 4 9 6.4 7.4 41 27 0 0 0 0
12 Madagascar 12 24 5.5 5.9 44 36 0 36 0 1
13 BurkinaFaso 5 9 4.1 5.4 . . 41 0 0 0 0
14 Mali 13 19 4.4 3.5 32 24 0 0 0 0
15 Burundi 2 7 6.7 9.5 0 0 0 0
16 Uganda 7 10 4.7 5.1 38 52 0 52 0 1
17 Nigeria 17 34 5.7 6.3 13 17 22 58 2 9
18 Zambia 23 54 7.2 6.7 35 0 35 0 1
19 Niger 7 18 7.2 8.0 . . 31 0 0 0 0
20 Rwanda 3 7 7.5 8.2 . . 0 0 0 0
21 China 18 50 2.3 . . 6 6 42 45 38 78
22 India 19 27 3.9 4.0 7 6 26 39 11 36
23 Pakistan 24 31 4.3 4.5 20 21 33 51 2 7
24 Kenya 9 22 8.1 8.2 40 57 0 57 0 1
25 Togo 11 25 6.6 7.0 60 0 0 0 0
26 CentralAfricanRep. 27 45 4.3 4.9 40 36 0 0 0 0
27 Haiti 18 29 4.2 4.0 42 56 0 56 0 1
28 Benin 11 40 9.0 7.8 . . 63 0 63 0 1
29 Ghana 26 33 3.2 4.2 25 35 0 48 0 2
30 Lesotho 6 19 7.5 7.2 0 0 0 0
31 SriLanka 20 21 2.3 1.3 28 16 0 16 0 1
32 Guinea 12 24 4.9 5.7 37 80 0 80 0 1
33 Yemen, PDR 30 42 3.5 4.7 61 49 0 0 0 0
34 Indonesia 16 27 4.8 4.8 20 23 34 50 3 9
35 Mauritania 10 40 9.5 7.8 39 0 0 0 0
36 Sudan 13 21 5.6 4.1 30 31 0 31 0 I
37 Afghanistan 9 . . 6.0 . . 33 . . 0 . . 0
38 Myan,nar 21 24 3.2 2.3 23 23 23 23 1 2
39 Kampuchea,Dem. 11 . . -0.5 . . . . . . . . .
40 Liberia 22 43 6.2 5.8 . . 0 0 0 0
41 SierraLeone 15 26 4.4 5.0 37 47 0 0 0 0
42 VietNam .
. 3.9 . . 21 50 4
Middle-income economies 42 w 58 w 3.9 w 3.3 w 29 w 31 w 34w 47 w Sit 113
Lower-middle-income 40w 56 w 4.0w 3.5 w 29 w 32 w 32 w 47 w 31 t 70
43 Bolivia 40 50 3.1 4.3 47 44 0 44 0 1
44 Philippines 32 41 4.1 3.7 27 30 27 34 1 2
45 Yemen Arab Rep. 5 23 10.2 8.8 . . 25 0 0 0 0
46 Senegal 33 38 3.3 4.0 53 65 0 65 0 1
47 Zimbabwe 14 27 6.0 6.2 40 50 0 50 0 1
48 Egypt, Arab Rep. 41 48 2.8 3.5 38 39 53 53 2 3
49 Dominican Rep. 35 59 5.2 4.3 50 54 0 54 0 1
50 Côted'Ivoire 23 45 7.5 6.6 27 34 0 34 0 I
51 Papua New Guinea 5 15 8.2 4.5 . . 25 0 0 0 0
52 Morocco 32 47 4.3 4.4 16 26 16 50 1 4
53 Honduras 26 42 5.5 5.6 31 33 0 0 0 0
54 Guatemala 34 33 2.7 2.9 41 36 41 36 1 1
55 Congo,People's Rep. 34 41 4.5 4.8 77 56 0 0 0 0
56 El Salvador 39 44 3.2 1.9 26 22 0 0 0 0
57 Thailand 13 21 5.1 4.7 65 69 65 69 I
58 Botswana 4 22 12.6 8.4 . . . . . . . . .
59 Camemon 16 47 8.1 7.2 26 21 0 21 0 1
60 Jamaica 38 51 2.8 2.6 77 66 0 66 0 1
61 Ecuador 37 55 4.7 4.7 31 29 0 51 0 2
62 Colombia 54 69 3.7 3.0 17 26 28 51 3 4
63 Paraguay 36 46 3.8 4.5 44 44 0 44 0 1
64 Tunisia 40 54 4.0 2.9 40 30 40 30 1 1
65 Turkey 34 47 4.2 3.4 18 24 32 42 3 4
66 Pens 52 69 4.3 3.1 38 39 38 44 1 2
67 Jordan 46 67 4.3 5.1 31 37 0 37 0 I
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
238
Urban populal ion Number of
Percentage of urban population
As a percentage Average annual cities of
of total growth rate In cities of over over 500000
population (percent) In largest city 500,000 persons persons
1965 1988 1965-80 1980-88 19(1) 1980 1960 1980 1960 1980
68 Chile 72 85 2.6 2.3 38 44 38 44 I I
69 SyrianArabRep. 40 51 4.6 4.5 35 33 35 55 1 2
70 Costa Rica 38 45 4.0 1.9 67 64 0 64 0 1
71 Mexico 55 71 4.4 3.1 28 32 36 48 3 7
72 Mauritius 37 42 2.5 0.8
73 Poland 50 61 1.9 1.4 17 15 41 47 5 8
74 Malaysia 26 41 4.5 4.9 19 27 0 27 0 1
75 Panama 44 54 3.5 3.0 61 66 0 66 0 1
76 Brazil 50 75 4.5 3.6 14 15 35 52 6 14
77 Angola 13 27 6.4 5.8 44 64 0 64 0 1
78 Lebanon 50 4.5 64 64
79 Nicaragua 43 59 4.7 41 47 0 0
Upper-middle-income 45 w 62 w 3.6w 2.8 w 31 w 30 w 38 w 48 w 20 1 431
80 South Africa 47 58 3.3 3.4 16 13 44 53 7
81 Algeria 38 44 3.7 3.9 27 12 27 12
82 Hungary 43 60 2.0 1.2 45 37 45 37 I I
83 Uniguay 81 85 0.7 0.8 56 52 56 52 I I
84 Argentina 76 86 2.2 1.8 46 45 54 60 3 5
85 Yugoslavia 31 49 3.0 2.5 11 10 11 23 1 3
86 Gabon 21 44 7.3 6.7 . . . . . . . . . .
87 Venezuela 70 83 4.8 2.6 26 26 26 44 1 4
88 TrinidadandTobago 30 67 5.6 3.9 . . . . 0 0 0 0
89 Korea,Rep.of 32 69 5.8 3.7 35 41 61 77 3 7
90 Portugal 24 32 1.7 1.9 47 44 47 44 1
91 Greece 48 62 2.0 1.3 51 57 51 70 1 2
92 Oman 4 10 7.6 8.7 . . . . . . . . .
93 Libya 26 68 9.8 6.7 57 64 0 64 0 1
94 Iran,IslaenicRep. 37 54 5.1 4.1 26 28 26 47 1 6
95 Iraq 51 73 5.3 4.8 35 55 35 70 1 3
96 Romania 38 49 3.0 0.3 22 17 22 17 1
Low- and middle-income 24w 41 w 3.7 w 6.9 w 16 w 18 w 31 w 44w 110 1 278
Sub-Saharan Africa 14w 28 w 5.8 w 6.2 w 28 w 36 w 6w 41 w 31 28
East Asia 19w 46w 3.0w . . 11w 13w 41w 47w 461 1021
SouthAsia 18w 26w 4.0w 4.0w 11w 11w 25w 40w 15t 491
Europe, M.East, & N.Africa 37 w SOw 3.5 w 3.1 w 28 w 28 w 31 w 40w 22 1 431
Latin America & Caribbean 53 w 71 w 4.0w 3.2 w 27 w 29 w 32 w 49w 201 49 I
Severely indebted 50w 66 w 3.8 w 3.1 w 26 w 28w 33 w 47w 24t 561
High-income economies 71 w 78 w 1.4w 0.8w 19w 19w 47 w 55 w 107 1 157 1
OECD members 72 w 77 w 1.3 w 0.7 w 18w 18 w 47 w 55 w 1041 152 1
(Other 68 w 83 w 4.6 w 3.5 w 58 w 49 w 51 w 54 w 3t S
97 tSaudi Arabia 39 76 8.5 5.8 15 18 0 33 1) 2
98 Spain 61 77 2.2 1.3 13 17 37 44 5 6
99 Ireland 49 58 2.1 1.1 51 48 51 48 1 1
100 tlsrael 81 91 3.5 2.1 46 35 46 35 1 1
101 tSingapore 100 100 1.6 1.1 100 100 100 100 I
102 tHong Kong 89 93 2.1 1.7 100 100 100 100 1
103 New Zealand 79 84 1.6 0.9 25 30 0 30 0 1
104 Australia 83 86 2.0 1.4 26 24 62 68 4 5
105 UnitedKingdom 87 92 0.5 0.4 24 20 61 55 15 17
106 Italy 62 68 1.0 0.5 13 17 46 52 7 9
107 tKuwait 78 95 8.2 5.1 75 30 0 0 0 0
108 Belgium 93 97 0.4 0.2 17 14 28 24 2 2
109 Netherlands 86 88 1.2 0.5 9 9 27 24 3 3
110 Austria 51 57 0.8 0.6 51 39 51 39 1 1
Ill tUnited Arab Emirates 41 78 23.7 4.2 . . .
112 France 67 74 1.3 0.5 25 23 34 34 4 6
113 Canada 73 76 1.5 1.0 14 18 31 62 2 9
114 Denmark 77 86 1.1 0.3 40 32 40 32 1
115 Germany,Fed.Rep. 79 86 0.7 0.1 20 I 48 45 11 11
116 Finland 44 60 2.5 0.4 28 27 0 27 0 1
117 Sweden 77 84 0.9 0.3 15 15 15 35 1 3
118 UnitedStates 72 74 1.2 1.0 13 12 61 77 40 65
119 Norway 48 74 3.0 0.9 50 32 50 32 1
120 Japan 67 77 2.1 0.7 18 22 35 42 5 9
121 Switzerland 53 61 1.0 1.3 19 22 19 22 1
Total reporting economies 34 w 47 w 2.7w 4.9w 17w 18w 35w 46w 2171 4351
Oil exporters 30 w 46 w 4.8w 4.3w 24w 26w 31w 49w 16t Sot
Nonreporting nonmembers 53 w 67 w 2.1 w 1.6 w 9w 7w 20 w 30 w 31t 58
239
Table 32. Women in development
Health and welfare Education
Life expectancy Maternal Persistence to grade 4
Risk of dying
at birth (years) mortality as a percentage of cohort Feniales per 100 males
byage5 (peri00000
Female Male Female Male Primary Secondary
Female Male live births)
1988 1988 1965 1988 1965 1988 1980 1970 1984 1970 /984 1970 1987 1970 1987
Low-income economies 89 w 97 w 50 w 60 w 48 w 60 w
China and India 67 w 74 w 52 w 63 w 50 w 64 w .
Other low-income 132 w 142 w 45 w 56w 43 w 53 w 64w 75 w 72 w 73 w 61 w 67 w 44 w 60 w
1 Mozambique 190 210 39 50 36 47 479t 78 . . 54
2 Ethiopia 183 203 43 49 42 46 2,000k' 57 45 56 50 46 61 32 63
3 Chad 198 221 38 47 35 45 700 34 40 9 18
4 Tanzania 135 153 45 55 41 52 370b 82 88 88 89 65 99 38 66
5 Bangladesh 175 160 44 51 45 51 600 . . 30 47 66 . . 45
6 Malawi 234 248 40 48 38 47 250 55 64 60 65 59 80 36 60
7 Somalia 200 223 40 49 37 46 1,100 46 59 51 65 33 . . 27
8 Zaire 139 157 45 54 42 51 800b 56 . 65 . 58 . . 26
9 Bhutan 186 178 40 47 41 49 . . 26 . 29 5 . . 3
10 LaoPDR 158 178 42 51 39 48 59 78 36 73
11 Nepal 187 173 40 51 41 52 .. .. 18 .. 16
12 Madagascar 160 178 45 51 42 50 300 65 . . 63 . . 86 94 70 85
13 BurkinaFaso 188 208 40 49 37 46 600 71 84 68 82 57 59 33 46
14 Mali 209 237 39 49 37 46 . . 52 68 89 75 55 59 29 42
15 Burundi 98 113 45 51 42 47 47 84 45 84 49 . . 17
16 Uganda 147 167 47 50 44 47 300 65 . . 31
17 Nigeria 151 170 43 53 40 49 1,500 64 . . 66 . . 59 . . 49
18 Zambia 108 124 46 55 43 52 110 93 97 99 . . 80 . 49
19 Niger 204 226 38 47 35 43 420b 75 76 74 88 53 . . 35 42
20 Rwanda 184 205 51 50 47 47 210 63 82 65 81 79 97 44 35
21 China 30 40 57 66 53 69 44 . . 76 . . 77 . . 83 . . 69
22 India 118 120 44 58 46 58 500 42 . . 45 . . 60 . . 39
23 Pakistan 139 128 45 55 47 55 600 56 . . 60 . . 36 49 25 39
24 Kenya 95 110 50 61 46 57 510b 84 75 84 73 71 93 42 70
25 Togo 132 150 44 55 40 52 476b 85 77 88 70 45 63 26 32
26 CentralAfricanRep. 150 169 41 52 40 49 600 67 67 67 74 49 62 20 37
27 Haiti 137 158 47 57 44 53 340 .
28 Benin 152 170 43 52 41 49 1,680" 59 64 67 63 45 51 44
29 Ghana 126 144 49 56 46 52 1,070" 77 . 82 . 75 80 35 66
30 Lesotho 125 142 50 58 47 54 . 87 86 70 75 150 125 III 153
31 SriLanka 19 27 64 73 63 68 90 94 97 73 99 89 93 101 106
32 Guinea 210 233 36 43 34 42 . . 62 . . 67 46 45 26 31
33 Yemen, PDR 157 175 41 53 39 50 . . 68 . . 56 . . 25 . . 25
34 Indonesia 75 90 45 62 43 59 800 67 78 89 99 84 . . 59 79
35 Mauritania 192 214 39 48 36 45 119 . . 91 . . 96 39 70 13 44
36 Sudan 158 177 41 51 39 49 607" . . 81 . . 80 61 . . 40
37 Afghanistan . . . . 35 . . 35 . . 64 . . 71 . . 16 . . 16
38 Myanmar 79 94 49 62 46 59 140 39 . . 58 . . 89 . . 65
39 Kampuchea,Dem. .. . . 46 . . 43 . . . . .. 70 43
40 Liberia 173 192 44 52 41 48 173 . . . . 49 . . 30
41 Sierra Leone 235 260 34 43 31 41 450 67 . . 40
42 VietNam 46 59 . . 68 . . 64 110
Middle-income economies 59 w 69 w 59w 68 w 55 w 63 w 77 w 84 w 76 w 89 w 85 w 88 w 89 w 101 w
Lower-middle-income 64w 75w 57w67w 54w62w 70w82w 71w86w 85w88w 85w94w
43 Bolivia 140 158 47 56 42 51 480 69 87 64
44 Philippines 45 58 57 66 54 62 80 . . 82 . . 76 . . 97 .
45 YemenArabRep. 193 215 41 48 39 47 . . 44 71 31 76 10 29 3 12
46 Senegal 109 125 42 50 40 46 530" . . 88 . . 92 63 69 39 51
47 Zimbabwe 55 67 50 65 46 61 150b 74 87 80 87 79 . . 63
48 Egypt, Arab Rep. 98 107 50 64 48 61 500 85 . . 93 . . 61 75 48 68
49 Dominican Rep. 73 80 57 68 54 64 56 . . 52 . . 70 99 100 .
50 Côted'Ivoire 137 156 44 54 40 51 . . 77 82 83 83 57 . . 27 44
51 PapuaNewGuinea 69 84 44 55 44 53 1,000 76 . . 84 . . 57 79 37 57
52 Morocco 84 100 51 63 48 59 327b 77 . . 79 51 63 40 66
53 Honduras 71 85 51 66 48 62 82 38 63 35 59 99 . . 79
54 Guatemala 63 77 50 65 48 60 110 33 62 73 73 79 . . 65
55 Congo,People'sRep. 166 180 47 56 41 50 . . 86 82 89 89 78 95 43 76
56 El Salvador 63 77 56 67 53 58 74 56 . . 56 . . 92 102 77 92
57 Thailand 28 38 58 68 54 63 270 71 . . 69 . . 88 . . 69
58 Botswana 41 52 49 69 46 65 300 97 95 90 95 113 107 88 107
59 Cameroon 116 133 47 58 44 55 303 59 85 58 86 74 85 36 64
60 Jamaica 10 14 67 75 64 71 100 . 100 97 103
61 Ecuador 74 81 57 68 55 64 220 69 . 70 . 93 96 76 91
62 Colombia 39 49 61 71 57 66 130 57 75 51 67 101 . . 73
63 Paraguay 44 56 67 69 63 65 469 70 75 71 76 89 92 91 98
64 Tunisia 50 63 52 67 51 65 l,000c 90 . 94 64 81 38 75
65 Turkey 89 97 55 66 52 63 207 76 97 81 98 73 89 37 60
66 Peru 105 114 52 64 49 60 310 85 . . 74
67 Jordan 49 60 52 68 49 65 . 90 99 92 . 78 96 53 95
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
240
Health and welfare Education
Life expectancy Maternal Persistence to grade 4
Risk of dying
at birth (years) mortality as a percentage of cohort Females per 100 males
byage5
(perlOO,000
Female Male Female Male Primary Secondary
Female Male live births)
1988 1988 1965 1988 1965 1988 1980 1970 1984 1970 1984 1970 1987 1970 1987
68 Chile 20 25 63 75 57 68 55 86 96 83 97 98 95 130 108
69 Syrian Arab Rep. 55 67 54 67 51 64 280 92 96 95 97 57 87 36 70
70 CostaRica 18 22 66 77 63 73 26 93 92 91 90 96 94 111 105
71 Mexico 51 57 61 73 58 66 92 . . 72 . . 95 92 94 . . 88
72 Maurilius 21 29 63 70 59 63 99 97 99 97 99 94 98 66 92
73 Poland 17 22 72 76 66 68 12 99 . . 97 . . 93 95 251 263
74 Malaysia 22 30 60 72 56 68 59 . . 100 . . 99 88 95 69 98
75 Panama 21 29 65 74 62 70 90 97 90 97 89 92 92 99 105
76 Brazil 62 75 59 68 55 63 150 56 54 . . 99 . . 99
77 Angola 207 229 37 47 34 43 . . 0 . 55 . . 77
78 Lebanon . . . . 64 . . 60 . . . . 83 . . 67
79 Nicaragua 66 80 52 65 49 62 65 48 64 45 58 101 107 89 168
Upper-middle-income 46w 55w 62w 70w 59w 65w 89w 89w 86w 97w 85w 89w 95w112w
80 South Africa 81 97 54 64 49 58 550C 98 . . 95
81 Algeria 83 91 51 65 49 63 129 90 . . 95 . . 60 79 40 73
82 Hungary 15 22 72 74 67 67 28 90 97 99 97 93 95 202 190
83 Uniguay 22 28 72 76 65 69 56 . . 99 . . 99 91 95 129
84 Argentina 30 40 69 74 63 68 85 92 . . 69 . . 98 97 156 172
85 Yugoslavia 25 30 68 75 64 69 27 91 . . 99 . . 91 94 86 94
86 Gabon 147 166 44 54 41 51 124b 73 80 78 78 91 98 43 81
87 Venezuela 36 45 65 73 61 67 65 84 82 61 89 99 . . 102
88 TrinidadandTobago 15 20 67 74 63 69 81 78 99 74 96 97 98 113 100
89 Korea,Rep.of 22 31 58 73 55 66 34 96 100 96 99 92 94 65 87
90 Portugal 13 17 68 78 62 71 15 92 . . 92 . . 95 97 98 114
91 Greece 13 15 72 79 69 74 12 97 98 96 99 92 94 98 101
92 Oman 35 46 45 66 43 63 82 96 82 99 16 85 . . 65
93 Libya 84 100 51 63 48 59 . 92 . . 95 . . 59 . . 21
94 Iran, IslansicRep. 74 89 52 64 52 62 75 79 74 99 55 80 49 68
95 Iraq 81 89 53 65 51 63 . . 84 90 90 92 41 82 41 62
96 Romania 24 31 70 73 66 68 180 90 . . 89 . . 97 95 151 233
Low- and middle-income 81 w 90w 52 w 62 w 50 w 61 w 61w 78w 64w 79w 69 w 81w 59w 75w
Sub-Saharan Africa 154 w 172 w 43 w 52 w 41 w 49 w 66w 73w 69w 74w 60 w 77 w 40w 59w
East Asia 37 w 47 w 56 w 66 w 52 w 67 w 78w 81w 84 w 71w
South Asia 124 w 123 w 45 w 57 w 46 w 57 w 45w iR 57 w 40w
Europe, M.East, & N.Africa 74 w 84w 58 w 66 w 55 w 62 w 84w 88w 87w 94w 70 w w 86 w 104 w
Latin America & Caribbean 57 w 67 w 60 w 70 w 56 w 64w 64w 75w 59w 86w 96 w 96 w 100w 110w
Severely indebted 56w 67 w 61 w 69 w 57 w 63 w 73 w 79 w 69 w 87 w 92 w 92 w 122 w 131 w
High-income economies 10 w 12 w 74 w 79 w 67 w 73 w 95 w 97 w 94 w 96 w 95 w 95 w 94 w 100 w
OECD members 9w 11 w 74 w 79 w 68 w 73 w 96 w 97 w 94 w 96 w 96 w 95 w 95 w 100 w
tOther 37 w 46 w 64 w 73 w 59 w 68 w 95 w 94 w 94 w 94 w 74 w 94 w 69 w 105 w
97 tSaudi Arabia 72 86 50 65 47 62 52 93 93 91 93 46 80 16 66
98 Spain 9 11 74 80 69 74 10 76 97 76 96 99 93 84 101
99 Ireland 8 10 73 77 69 72 7 96 95 124 101
100 tlsrael 11 15 74 77 71 74 5 96 98 96 98 92 98 131 121
101 tSingapore 7 10 68 77 64 71 11 99 . 99 . 88 . . 103
102 tHongKong 7 10 71 80 64 75 4 94 . . 92 . . 90 92 74 104
103 NewZealand II 15 74 78 68 72 . . . . 98 . . 98 94 95 94 98
104 Australia 10 13 74 80 68 73 11 . . 97 . . 94 94 95 91 98
105 United Kingdom 9 12 74 78 68 73 7 . . 95 95 94 96
106 Italy 9 12 73 80 68 74 13 . . . . 94 . . 86
107 lKuwait 14 20 65 76 61 71 18 96 92 98 93 73 95 74 67
108 Belgium 9 13 74 78 68 72 10 87 . . 85 . . 94 96 87 103
109 Netherlands 8 10 76 80 71 74 5 99 . . 96 . . 96 98 91 111
110 Austria 9 13 73 79 66 72 11 95 99 92 100 95 94 95 93
Ill tUnited Arab Emirates 23 32 59 73 56 69 97 95 93 92 61 94 . . 99
112 France 8 10 75 81 68 72 13 97 96 90 99 95 94 107 108
113 Canada 8 10 75 81 69 73 2 95 97 92 93 95 93 95 95
114 Denmark 9 11 75 78 70 72 4 98 100 96 100 97 96 102 105
115 Gerrnany,Fed.Rep. 8 11 73 78 67 72 11 97 97 96 96 96 96 93 97
116 Finland 7 9 73 78 66 72 5 . . 99 . . 98 90 95 112 112
117 Sweden 6 8 76 80 72 74 4 98 .. 96 . . 96 95 92 104
118 United States 10 13 74 79 67 72 9 . . 96 . . 94 95 94 98
119 Norway 9 11 76 80 71 74 . . 99 99 98 99 105 95 97 103
120 Japan 5 7 73 81 68 76 15 100 100 100 100 96 95 101 99
121 Switzerland 7 9 75 81 69 74 5 94 99 93 99 98 97 93 99
Total reporting economies 69w 77w 57w65w 54w63w 67w 83w 69w83w 76w 84w 67w 79w
Oil exporters 87w 99w 50w63w 48w 59w 73w 80w 82w97w 73w 85w 54w 77w
Nonreporting nonmembers 25 w 31 w 72 w 75 w 65 w 66 w 94 w 98 w 99 w 98 w
a. See the technical notes. b. Data refer to maternal mortality in hospitals and other medical institutions only. c. Community data frum rural areas only.
241
echnical notes
This thirteenth edition of the World Development 1980. This is accomplished by rescaling, which
Indicators provides economic and social indicators moves the year in which current and constant
for selected periods or years in a form suitable for price versions of the same time series have the
comparing economies and groups of economies. same value, without altering the trend of either.
The main criterion of country classification is Components of gross domestic product (GDP) are
gross national product (GNP) per capita. With the individually rescaled and summed to provide GDP
inclusion of a new World Bank member, Angola, and its subaggregates. In this process, a rescaling
the main tables now include country data on 121 deviation may occur between constant price GDP
economies. As in last year's edition, since only by industrial origin and GDP by expenditure. Such
sparse data are available for nine nonreporting rescaling deviations are absorbed under the head-
nonmember economies, these countries are not in- ing private consumption, etc., because GDP by in-
cluded in the main tables. Summary measures for dustrial origin is usually the more reliable esti-
them are shown in the main tables where avail- mate.
able, and selected country data are presented in This approach takes into account the effects of
Box A.2. Box A.1, Basic indicators for economies changes in intersectoral relative prices between the
with populations of less than 1 million, covers an- original and the new base period. Because private
other fifty-five economies. As in last year's edition, consumption is calculated as a residual, the na-
the table on labor force is not included because of tional accounting identities are maintained. It
the continuing lack of new data. This table will be does, however, involve incorporating in private
reinserted when the 1990 round of census results consumption whatever statistical discrepancies
has been tabulated and collected by the Interna- arise for expenditure in the rebasing process. The
tional Labour Office (ILO). Other changes are out- value added in the services sector also includes a
lined in the introduction. statistical discrepancy as reported by the original
Considerable effort has been made to standard- source.
ize the data; nevertheless, statistical methods, cov- The summary measures are calculated by simple
erage, practices, and definitions differ widely. In addition when a variable is expressed in reason-
addition, the statistical systems in many develop- ably comparable units of account. Economic indi-
ing economies are still weak, and this affects the cators that do not seem naturally additive are usu-
availability and reliability of the data. Moreover, ally combined by a price-weighting scheme. The
intercountry and intertemporal comparisons al- summary measures for social indicators are
ways involve complex technical problems, which weighted by population. It should be emphasized,
cannot be fully and unequivocally resolved. The however, that use of a single base year raises prob-
data are drawn from sources thought to be most lems over a period encompassing profound struc-
authoritative, but many of them are subject to con- tural changes and significant changes in relative
siderable margins of error. Readers are urged to prices, such as occurred from 1965 to 1988.
take these limitations into account in interpreting The Bank's statistical publications will soon shift
the indicators, particularly when making compari- to a 1987 base year. With some exceptions, use of
sons across economies. 1987 rather than 1980 values as country weights
To facilitate international comparisons, national will not greatly alter the group indexes and growth
accounts constant price data series with base years rates reported here. Most exceptions relate to oil
other than 1980 have been partially rebased to exporters and reflect declining shares of group
242
Box A.1 Basic indicators for economies with populations of less than I million
GNP per capita'
Average Life
Area annual Average annual expectancy
growth rate rate of inflation' at birth Adult illiteracy
Population (thousands
(thousands) of square Dollars (percent) (percent) (years) Female Total
mid-1988 kilometers) 1988 1965-88 1965-80 1980-88 1988 1985 1985
1 Guinea-Bissau 940 36 190 -1.9 . . 49.0 40 83 69
2 Gambia, The 822 11 200 1.1 8.1 13.9 44 85 75
3 Equatorial Guinea 336 28 410 . . . . 46 63
4 Maldives 202 b 410 2.3 . . 7.1 60 .
5 Guyana 799 215 420 -4.4 8.1 15.9 63 5 4
6 Comoros 442 2 440 0.6 5.8 56
7 SaoTomeandPrincipe 119 1 490 -0.1 . . 18.1 65
8 Solomon Islands 303 29 630 . . 7.7 13.1 64
9 Western Samoa 159 3 640 . . 10.5 66
10 Kiribati 67 1 650 5.7 55
11 Cape Verde 360 4 680 . . . . 8.9 65 61 53
12 Swaziland 735 17 810 2.2 9.1 11.4 56 34 32
13 Tonga 97 1 830 . . . . . . 66
14 Vanuatu 147 12 840 . . . . 4.3 64
15 St.VincentandtheGrenadines 112 b 1,200 2.0 10.9 4.3 70
16 Belize 180 23 1,500 2.4 7.1 2.2 67 .
17 Fiji 732 18 1,520 1.9 10.4 5.7 71 19 15
18 St. Lucia 145 1 1,540 2.7 9.3 3.9 71 . I
19 Dominica 82 1 1,680 0.6 12.8 5.8 74
20 Grenada 94 b 1,720 . . 11.2 7.4 69
21 Suriname 427 163 2,460 1.1 . . 5.8 67 10 10
22 St. Kitts and Nevis 42 b 2,630 3.6 9.7 6.2 69
23 Antigua and Barbuda 78 b 3,690 0.6 .. .. 73
24 Seychelles 68 b 3,800 3.2 12.2 3.9 70 .
25 Malta 348 b 5,190 7.4 3.5 1.9 73 18 16
26 Barbados 254 b 6,010 2.3 11.3 6.1 75
27 Cyprus 687 9 6,260 . . . . 6.4 76 .
28 Bahrain 473 1 6,340 . . -1.3 68 36 27
29 Qatar 411 11 9,930 . . . . . . 70
30 Bahamas, The 244 14 10,700 1.0 6.4 6.2 68
31 Iceland 249 103 20,190 3.3 26.8 38.0 78
32 Luxembourg 375 3 22,400 4.1 6.7 4.2 75
33 American Samoa 37 b c
34Aruba 60 b c
35 Bermuda 57 b c 8.1 10.7
36 Brunei 241 6 c -4.4 75
37 Channel Islands 137 . . c 76
38 Djibouti . . 23 d . . 47
39 Faeroe Islands 47 1 c
40 Fed. States of Micronesia 99 c
41 French Guiana 88 90 e . . 7.4 73
42 French Polynesia 186 4 c 72
43 Gibraltar 30 b e
44 Greenland 55 342 c .
45 Guadeloupe 338 2 e 8.7 74
46 Guam 130 1 c . . 73
47 Isle of Man 66 c
48 Marshall Islands 41 . . c .
49 Macan 442 b e .. . . 71
50 Martinique 335 1 e . . 9.2 75
51 Netherlands Antilles 183 1 c . . 66
52 New Caledonia 158 19 c . . 68
53 Puerto Rico 3,321 9 e 4.5 75
54 Reunion 576 3 e . . . . . . 71
55 Virgin Islands (U.S.) 106 b c 1.9 6.0 4.5 70
Note: Economies in italics are those for which 1988 GNP per capita cannot be calculated; figures in italics are for years other than those specified.
I a. See the technical note to Table 1. b. Less than 500 square kilometers. c. GNP per capita estimated to be in the high-income
range. d. GNP per capita estimated to be in the lower-middle-income range. e. CNP per capita estimated to be in the upper-middle-income
range. f. Population is more than I million.
243
Box A.2 Selected indicators for nonreporting nonmember economies
Democratic
People's German
Republic of Democratic
USSR Korea Republic Czechoslovakia Cuba
Indicator 1965 1988 1965 1988 1965 1988 1965 1988 1965 1988
Population (millions) 232 286 12 21 17 17 14 16 8 10
Urban population (percentage of total) 52 67 45 66 73 77 51 67 58 74
Life expectancy at birth (years) 69 70 57 70 70 73 69 71 67 76
Crude birth rate (per thousand) 18 18 44 20 17 13 16 14 34 18
Crude death rate (per thousand) 7 11 12 5 14 13 10 12 8 7
Population per physician 480 270 . . 420 870 440 540 280 1,150 530
Total fertility rate 2.5 2.4 6.5 2.5 2.5 1.8 2.4 2 4.4 1.9
Infant mortality (per 1,000 live births) 28 25 63 27 25 8 26 13 38 12
Low birth weight (percent) 6 6 8
Risk of dying by age 5 (female) 24 26 . . 8 . . 13 . 13
Risk of dying by age 5 (male) 33 35 . 11 . . 17 16
Daily calorie supply (per capita) 3,205 3,399 2,329 3,232 3,204 3,814 3,383 3,448 2,374 3,124
Food production per capita index
(1979-81 = 100) 85 111 72 111 73 118 73 119 81 103
Education, primary (female) 103 . . 111 105 97 96 119 100
Education, primary (total) 103 106 . . 109 106 99 96 121 104
Area (thousands of square kilometers) . 22,402 121 108 128 111
Population projected to year 2000 (millions) . 307 25 16 . . 16 12
Note: For data comparability and coverage, see the technical notes. Figures in italics are for years other than those specified.
GNP, trade, and so on from 1980 to 1987. This is tals. This is explained more fully in the introduc-
most notable for Sub-Saharan Africa, with the dra- tion to the World Tables.
matic decline in Nigeria's weight. In contrast, All growth rates shown are calculated from con-
changing the base year for country series them- stant price series and, unless otherwise noted,
selves, as described above, is likely to alter trends have been computed using the least-squares
significantly. Differences of half a percentage point method. The least-squares growth rate, r, is esti-
a year in growth rates could be quite common; mated by fitting a least-squares linear regression
larger changes may occur for economies that have trend line to the logarithmic annual values of the
undergone significant structural change, such as variable in the relevant period. More specifically,
oil exporters. the regression equation takes the form log X = a
The World Development Indicators, unlike the + bt + et, where this is equivalent to the logarith-
World Tables, provide data for (usually) two refer- mic transformation of the compound growth rate
ence points, rather than annual time series. For equation, X = X0 (1 + r)t. In these equations, X is
summary measures that cover many years, the cal- the variable, t is time, and a = log X0 and b = log (1
culation is based on the same country composition + r) are the parameters to be estimated; e is the
over time and across topics. The World Develop- error term. If b* is the least-squares estimate of b,
ment Indicators do so by permitting group mea- then the average annual percentage growth rate, r,
sures to be compiled only if the country data avail- is obtained as [antilog (b*)] - 1 and multiplied by
able for a given year account for at least two-thirds 100 to express it as a percentage.
of the full group, as defined by the 1980 bench-
marks. So long as that criterion is met, uncurrent Table 1. Basic indicators
reporters (and those not providing ample history) For basic indicators for economies with popula-
are, for years with missing data, assumed to be- tions of less than 1 million, see Box A.1. For se-
have like the sample of the group that does pro- lected indicators for nonreporting nonmember
vide estimates. Readers should keep in mind that economies, see Box A.2.
the purpose is to maintain an appropriate relation- Population numbers for mid-1988 are World Bank
ship across topics, despite myriad problems with estimates. These are normally projections from the
country data, and that nothing meaningful can be most recent population censuses or surveys,
deduced about behavior at the country level by which, in some cases, are very dated. Note that
working back from group indicators. In addition, refugees not permanently settled in the country of
the weighting process may result in discrepancies asylum are generally considered to be part of the
between summed subgroup figures and overall to- population of their country of origin.
244
Bulgaria Albania Mongolia Namibia
1965 1988 1965 1988 1965 1988 1965 1988 Indicator
8 9 2 3 1 2 1 1 Population (millions)
46 69 32 35 42 51 28 55 Urban population (percentage of total)
69 72 66 72 50 62 45 57 Life expectancy at birth (years)
15 13 35 25 42 36 46 44 Crude birth rate (per thousand)
8 12 9 6 16 9 22 12 Crude death rate (per thousand)
600 280 2,100 . . 710 100 Population per physician
2.1 1.9 5.4 3.1 5.9 4.9 6.1 6.0 Total fertility rate
31 14 87 27 113 66 145 104 Infant mortality (per 1,000 live births)
Low birth weight (percent)
. 1i Risk of dying by age 5 (female)
. 18 . . 34 .. 91 137 Risk of dying by age 5 (male)
3,452 3,642 2,389 2,713 2,597 2,847 1,904 1,824 Daily calorie supply (per capita)
Food production per capita index
78 102 85 95 138 99 114 86 (1979-81 = 100)
102 103 87 99 97 103 Education, primary (female)
103 104 92 100 98 102 Education, primary (total)
111 29 . . 1,565 824 Area (thousands of square kilometers)
8 4 .. 3 2 Population projected to year 2000 (mfflions)
j
The data on area are from the Food and Agricul- and capital) less similar payments made to nonres-
ture Organization. idents who contributed to the domestic economy.
GNP per ca pita figures in U.S. dollars are calcu- In estimating GNP per capita, the Bank recog-
lated according to the World Bank Atlas method, nizes that perfect cross-country comparability of
which is described below. GNP per capita estimates cannot be achieved. Be-
GNP per capita does not, by itself, constitute or yond the classic, strictly intractable index number
measure welfare or success in development. It problem, two obstacles stand in the way of ade-
does not distinguish between the aims and ulti- quate comparability. One concerns the GNP and
mate uses of a given product, nor does it say population estimates themselves. There are differ-
whether it merely offsets some natural or other ences in national accounting and demographic re-
obstacle, or harms or contributes to welfare. For porting systems and in the coverage and reliability
example, GNP is higher in colder countries, where of underlying statistical information among vari-
people spend money on heating and warm ous countries. The other relates to the use of offi-
clothes, than in balmy climates, where people are cial exchange rates for converting GNP data, ex-
comfortable wearing light clothes in the open air. pressed in different national currencies, to a
More generally, GNP abstracts from environ- common denominationconventionally the U.S.
mental issues, particularly natural resource use. dollarto compare them across countries.
The Bank has joined with others to see how na- Recognizing that these shortcomings affect the
tional accounts might provide insights into these comparability of the GNP per capita estimates, the
issues. The possibility of developing "satellite" ac- World Bank has introduced several improvements
counts is being considered; such accounts could in the estimation procedures. Through its regular
delve into practical and conceptual difficulties, review of member countries' national accounts,
such as assigning a meaningful economic value to the Bank systematically evaluates the GNP esti-
resources that markets do not yet perceive as mates, focusing on the coverage and concepts em-
"scarce" and allocating costs that are essentially ployed and, where appropriate, making adjust-
global within a framework that is inherently na- ments to improve comparability. As part of the
tional. review, Bank staff estimates of GNP (and some-
GNP measures the total domestic and foreign times of population) may be developed for the
value added claimed by residents. It comprises most recent period.
GDP (defined in the note for Table 2) plus net fac- The Bank also systematically assesses the appro-
tor income from abroad, which is the income resi- priateness of official exchange rates as conversion
dents receive from abroad for factor services (labor factors. An alternative conversion factor is used
245
(and reported in the World Tables) when the official study is Phase V, parts of which have already been
exchange rate is judged to diverge by an exception- published by the European Communities (EC)
ally large margin from the rate effectively applied covering Europe and Africaand the Organisation
to foreign transactions. This applies to only a small for Economic Co-operation and Development
number of countries. For all other countries the (OECD).
Bank calculates GNP per capita using the Atlas The ICP has covered more than seventy coun-
method. tries in five phases, at five-year intervals. The Bank
The Atlas conversion factor for any year is the is currently reviewing the data and methodology
average of the exchange rate for that year and the underlying the latest estimates and will include an
exchange rates for the two preceding years, after updated comparison of ICP and Atlas numbers in a
adjusting them for differences in relative inflation future edition of the Atlas or another statistical
between the country and the United States. This publication.
three-year average smooths fluctuations in prices The ICP figures reported in Table 30 are prelimi-
and exchange rates for each country. The resulting nary and may be revised. The United Nations and
GNP in U.S. dollars is divided by the midyear its regional economic commissions, as well as
population for the latest of the three years to de- other international agencies, such as the EC, the
rive GNP per capita. OECD, and the World Bank, are working to im-
Some sixty low- and middle-income economies prove the methodology and to extend annual pur-
have suffered declining real GNP per capita in con- chasing power comparisons to all countries. How-
stant prices during the 1980s. In addition, signifi- ever, exchange rates remain the only generally
cant currency and terms of trade fluctuations have available means of converting GNP from national
affected relative income levels. For this reason the currencies to U.S. dollars.
levels and ranking of GNP per capita estimates, The average annual rate of inflation is measured by
calculated by the Atlas method, have sometimes the growth rate of the GDP implicit deflator for
changed in ways not necessarily related to the rela- each of the periods shown. The GDP deflator is
tive domestic growth performance of the econo- first calculated by dividing, for each year of the
mies. period, the value of GDP at current values by the
value of GDP at constant values, both in national
The following formulas describe the procedures currency. The least-squares method is then used to
for computing the conversion factor for year t: calculate the growth rate of the GDP deflator for
the period. This measure of inflation, like any
i Ipt p \ (p, Jp other, has limitations. For some purposes, how-
(e_2,,) =- [et2 + e,_i IiI + e,]
3 iI _2) _1) ever, it is used as an indicator of inflation because it
is the most broadly based measure, showing an-
and for calculating GNP per capita in U.S. dollars nual price movements for all goods and services
for year t: produced in an economy.
Life expectancy at birth indicates the number of
(Yl) = (Y, / N,
years a newborn infant would live if prevailing
where patterns of mortality at the time of its birth were to
stay the same throughout its life. Data are from the
= current GNP (local currency) for year U.N. Population Division, supplemented by
P, = GNP deflator for year World Bank estimates.
= annual average exchange rate (local currency/U. S. Adult illiteracy is defined here as the proportion
dollar) for year
N, = midyear population for year
of the over-15 population who cannot, with under-
P = U.S. GNP deflator for year t. standing, read and write a short, simple statement
on their everyday life. This is only one of three
Because of problems associated with the avail- widely accepted definitions, and its application is
ability of comparable data and the determination subject to qualifiers in a number of countries.
of conversion factors, information on GNP per The summary measures for GNP per capita, life
capita is not shown for nonreporting nonmarket expectancy, and adult illiteracy in this table are
economies. weighted by population. Those for average annual
The use of official exchange rates to convert na- rates of inflation are weighted by the 1980 share of
tional currency figures to U.S. dollars does not re- country GDP valued in current U.S. dollars.
flect the relative domestic purchasing powers of
currencies. The United Nations International Com- Tables 2 and 3. Growth and structure of
parison Program (ICP) has developed measures of production
real GDP on an internationally comparable scale, Most of the definitions used are those of the LI. N.
using purchasing power parities (PPPs) instead of System of National Accounts (SNA), Series F, No. 2,
exchange rates as conversion factors. Table 30 Revision 3. Estimates are obtained from national
shows the most recent ICP estimates. Information sources, sometimes reaching the World Bank
on the ICP has been published in four studies and through other international agencies but more of-
in a number of other reports. The most recent ten collected during World Bank staff missions.
246
World Bank staff review the quality of national U.S. dollar values for each economy are calculated
accounts data and in some instances, through mis- for each of the years of the periods covered; the
sion work or technical assistance, help adjust na- values are aggregated across countries for each
tional series. Because of the sometimes limited ca- year; and the least-squares procedure is used to
pabilities of statistical offices and basic data compute the growth rates. The average sectoral
problems, strict international comparability cannot percentage shares in Table 3 are computed from
be achieved, especially in economic activities that group aggregates of sectoral GDP in current U.S.
are difficult to measure such as parallel market dollars.
transactions, the informal sector, or subsistence ag- Table 4. Agriculture and food
riculture.
GDP measures the total output of goods and ser- The basic data for value added in agriculture are from
vices for final use produced by residents and non- the World Bank's national accounts series at cur-
residents, regardless of the allocation to domestic rent prices in national currencies. Value added in
and foreign claims It is calculated without making current prices in national currencies is converted to
deductions for depreciation of "manmade" assets U.s. dollars by applying the single-year conver-
or depletion and degradation of natural resources. sion procedure, as described in the technical note
Although SNA envisages estimates of GDP by in- for Tables 2 and 3.
dustrial origin to be at producer prices, many The figures for the remainder of this table are
countries still report such details at factor cost. In- from the Food and Agriculture Organization
ternational comparability of the estimates is af- (FAO). Cereal imports are measured in grain equiva-
fected by the use of differing country practices in lents and defined as comprising all cereals in the
valuation systems for reporting value added by Standard International Trade Classification (SITC), Re-
production sectors. As a partial solution, GDP esti- vision 2, Groups 041-046. Food aid in cereals covers
mates are shown at purchaser values if the compo- wheat and flour, bulgur, rice, coarse grains, and
nents are on this basis, and such instances are foot- the cereal component of blended foods. The fig-
noted. However, for a few countries in Tables 2 ures are not directly comparable because of report-
and 3, GDP at purchaser values has been replaced ing and timing differences. Cereal imports are
by GDP at factor cost. Note that in editions before based on calendar-year data reported by recipient
1986, GDP at producer prices and GDP at purchaser countries, and food aid in cereals is based on data
values were referred to for convenience as GDP at for crop years reported by donors and interna-
factor cost and GDP at market prices, respectively. tional organizations, including the International
The figures for GDP are dollar values converted Wheat Council and the World Food Programme.
from domestic currencies using single-year official Furthermore, food aid information from donors
exchange rates. For a few countries where the offi- may not correspond to actual receipts by beneficia-
cial exchange rate does not reflect the rate effec- ries during a given period because of delays in
tively applied to actual foreign exchange transac- transportation and recording, or because aid is
tions, an alternative conversion factor is used (and sometimes not reported to the FAO or other rele-
reported in the World Tables). Note that this table vant international organizations. Food aid imports
does not use the three-year averaging technique may also not show up in customs records. The
applied to GNP per capita in Table 1. earliest available food aid data are for 1974. The
Agriculture covers forestry, hunting, and fishing time reference for food aid is the crop year, July to
as well as agriculture. In developing countries with June.
high levels of subsistence farming, much of agri- Fertilizer consumption measures the plant nutri-
cultural production is either not exchanged or not ents used in relation to arable land. Fertilizer prod-
exchanged for money. This increases the difficulty ucts cover nitrogenous, potash, and phosphate
of measuring the contribution of agriculture to fertilizers (which include ground rock phosphate).
GDP and reduces the reliability and comparability Arable land is defined as land under temporary
of such numbers. Industry comprises value added crops (double-cropped areas are counted once),
in mining; manufacturing (also reported as a sub- temporary meadows for mowing or pastures, land
group); construction; and electricity, water, and under market or kitchen gardens, and land tempo-
gas. Value added in all other branches of economic rarily fallow or lying idle, as well as land under
activity, including imputed bank service charges, permanent crops. The time reference for fertilizer
import duties, and any statistical discrepancies consumption is the crop year, July to June.
noted by national compilers, are categorized as ser- The average index of food production per capita
vices, etc. shows the average annual quantity of food pro-
Partially rebased 1980 series in domestic curren- duced per capita in 1986-88 in relation to the aver-
cies, as explained at the beginning of the technical age produced annually in 1979-81. The estimates
notes, are used to compute the growth rates in are derived by dividing the quantity of food pro-
Table 2. The sectoral shares of GDP in Table 3 are duction by the total population. For this index food
based on current price series. is defined as comprising nuts, pulses, fruits, cere-
In calculating the summary measures for each als, vegetables, sugar cane, sugar beet, starchy
indicator in Table 2, partially rebased constant 1980 roots, edible oils, livestock, and livestock prod-
247
ucts. Quantities of food production are measured Industrial Classification of All Economic Activities
net of animal feed, seeds for use in agriculture, (ISIC), Revision 2. Food, beverages, and tobacco com-
and food lost in processing and distribution. prise ISIC Division 31; textiles and clothing, Division
The summary measures for fertilizer consump- 32; machinery and transport equipment, Major
tion are weighted by total arable land area; the Groups 382-84; and chemicals, Major Groups 351
summary measures for food production are and 352. Other comprises wood and related prod-
weighted by population. ucts (Division 33), paper and related products (Di-
vision 34), petroleum and related products (Major
Table 5. Commercial energy
Groups 353-56), basic metals and mineral products
The data on energy are primarily from U.N. (Divisions 36 and 37), fabricated metal products
sources. They refer to commercial forms of pri- and professional goods (Major Groups 381 and
mary energypetroleum and natural gas liquids, 385), and other industries (Major Group 390).
natural gas, solid fuels (coal, lignite, and so on), When data for textiles, machinery, or chemicals are
and primary electricity (nuclear, geothermal, and shown as not available, they are also included in
hydroelectric power)all converted into oil equiv- other.
alents. Figures on liquid fuel consumption include Summary measures given for value added in
petroleum derivatives that have been consumed in manufacturing are totals calculated by the aggrega-
nonenergy uses. For converting primary electricity tion method noted at the beginning of the techni-
into oil equivalents, a notional thermal efficiency of cal notes.
34 percent has been assumed. The use of firewood, Table 7. Manufacturing earnings and output
dried animal excrement, and other traditional fu-
els, although substantial in some developing coun- Four indicators are showntwo relate to real earn-
tries, is not taken into account because reliable and ings per employee, one to labor's share in total
comprehensive data are not available. value added generated, and one to labor produc-
Energy imports refer to the dollar value of energy tivity in the manufacturing sector. The indicators
importsSection 3 in the SITC, Revision 1and are based on data from the United Nations Indus-
are expressed as a percentage of earnings from trial Development Organization (UNIDO), al-
merchandise exports. though the deflators are from other sources, as ex-
Because data on energy imports do not permit a plained below.
distinction between petroleum imports for fuel Earnings per employee are in constant prices and
and those for use in the petrochemicals industry, are derived by deflating nominal earnings per em-
these percentages may overestimate the depen- ployee by the country's consumer price index
dence on imported energy. (CPI). The CPI is from the International Monetary
The summary measures of energy production Fund's International Financial Statistics. Total earn-
and consumption are computed by aggregating ings as a percentage of value added are derived by
the respective volumes for each of the years cov- dividing total earnings of employees by value
ered by the periods and then applying the least- added in current prices, to show labor's share in
squares growth rate procedure. For energy con- income generated in the manufacturing sector.
sumption per capita, population weights are used Gross output per employee is in constant prices and is
to compute summary measures for the specified presented as an index of overall labor productivity
years. in manufacturing with 1980 as the base year. To
The summary measures of energy imports as a derive this indicator, UNIDO data on gross output
percentage of merchandise exports are computed per employee in current prices are adjusted using
from group aggregates for energy imports and the implicit deflators for value added in manufac-
merchandise exports in current dollars. turing or in industry, taken from the World Bank's
national accounts data files.
Table 6. Structure of manufacturing To improve cross-country comparability, UNIDO
has, where possible, standardized the coverage of
The basic data for value added in manufacturing are establishments to those with five or more employ-
from the World Bank's national accounts series at ees.
current prices in national currencies. Value added The concepts and definitions are in accordance
in current prices in national currencies is converted with the International Recommendations for Industrial
to U.S. dollars by applying the single-year conver- Statistics, published by the United Nations. Earn-
sion procedure, as described in the technical note ings (wages and salaries) cover all remuneration to
for Tables 2 and 3. employees paid by the employer during the year.
The data for distribution of manufacturing value The payments include (a) all regular and overtime
added among industries are provided by the United cash payments and bonuses and cost of living al-
Nations Industrial Development Organization, lowances; (b) wages and salaries paid during vaca-
and distribution calculations are from national cur- tion and sick leave; (c) taxes and social insurance
rencies in current prices. contributions and the like, payable by the employ-
The classification of manufacturing industries is ees and deducted by the employer; and (d) pay-
in accordance with the U.N. International Standard ments in kind.
248
The value of gross output is estimated on the ports of goods and nonfactor services and imports
basis of either production or shipments. On the of goods and nonf actor services.
production basis it consists of (a) the value of all Partially rebased 1980series inconstant domestic
products of the establishment, (b) the value of in- currency units are used to compute the indicators
dustrial services rendered to others, (c) the value in Table 8. Table 9 uses national accounts series in
of goods shipped in the same condition as re- current domestic currency units.
ceived, (d) the value of electricity sold, and (e) the The summary measures are calculated by the
net change in the value of work-in-progress be- method explained in the note to Tables 2 and 3.
tween the beginning and the end of the reference Table 10. Structure of consumption
period. In the case of estimates compiled on a ship-
ment basis, the net change between the beginning Percentage shares of selected items in total house-
and the end of the reference period in the value of hold consumption expenditure are computed from
stocks of finished goods is also included. "Value details of GDP (expenditure at national market
added" is defined as the current value of gross prices) defined in the U.N. System of National Ac-
output less the current cost of (a) materials, fuels, counts (SNA), mostly as collected for International
and other supplies consumed, (b) contract and Comparison Program (ICP) Phases IV (1980) and V
commission work done by others, (c) repair and (1985). For countries not covered by the ICP, less
maintenance work done by others, and (d) goods detailed national accounts estimates are included,
shipped in the same condition as received. where available, in order to present a general idea
The term "employees" in this table combines of the broad structure of consumption. The data
two categories defined by the U.N., regular em- cover eighty-four countries (including Bank staff
ployees and persons engaged. Together these estimates for China) and refer to the most recent
groups comprise regular employees, working pro- estimates, generally for 1980 and 1985. Where they
prietors, active business partners, and unpaid f am- refer to other years the figures are shown in italics.
ily workers; they exclude homeworkers. The data Consumption here refers to private (nongovern-
refer to the average number of employees working ment) consumption as defined in the SNA and in
during the year. the notes to Tables 2 and 3, 4, and 9, except that
education and medical care comprise government
Tables 8 and 9. Growth of consumption and as well as private outlays. This ICP concept of "en-
investment; structure of demand hanced consumption" reflects who uses rather
GDP is defined in the note for Tables 2 and 3, but than who pays for consumption goods, and it im-
for these two tables it is in purchaser values. proves international comparability because it is
General government consumption includes all cur- less sensitive to differing national practices regard-
rent expenditure for purchases of goods and ser- ing the financing of health and education services.
vices by all levels of government. Capital expendi- Cereals and tubers, a major subitem of food, com-
ture on national defense and security is regarded prise the main staple products: rice, flour, bread,
as consumption expenditure. all other cereals and cereal preparations, potatoes,
Private consumption, etc., is the market value of all yams, and other tubers. For high-income OECD
goods and services, including durable products members, however, this subitem does not include
(such as cars, washing machines, and home com- tubers. Gross rents, fiiel and power consist of actual
puters) purchased or received as income in kind by and imputed rents and repair and maintenance
households and nonprofit institutions. It excludes charges, as well as the subitem fuel and power (for
purchases of dwellings but includes imputed rent heating, lighting, cooking, air conditioning, and so
for owner-occupied dwellings (see the note to Ta- forth). Note that this item excludes energy used for
ble 10 for details). In practice, it includes any statis- transport (rarely reported to be more than 1 per-
tical discrepancy in the use of resources. At con- cent of total consumption in low- and middle-
stant prices, it also includes the rescaling deviation income economies). As mentioned, medical care
from partial rebasing, which is explained in the and education include government as well as pri-
beginning of the technical notes. vate consumption expenditure. Transport and com-
Gross domestic investment consists of outlays on munication also includes the purchase of motor cars,
additions to the fixed assets of the economy plus which are reported as a subitem. Other consump-
net changes in the level of inventories. tion, the residual group, includes beverages and
Gross domestic savings are calculated by deducting tobacco, nondurable household goods and house-
total consumption from GDP. hold services, recreational services, and services
Exports of goods and nonfactor services represent the (including meals) supplied by hotels and restau-
value of all goods and nonfactor services provided rants; carry-out food is recorded here. It also in-
to the rest of the world; they include merchandise, cludes the separately reported subitem other con-
freight, insurance, travel, and other nonfactor ser- sumer durables, comprising household appliances,
vices. The value of factor services, such as invest- furniture, floor coverings, recreational equipment,
ment income, interest, and labor income, is ex- and watches and jewelry.
cluded. Current transfers are also excluded. Estimating the structure of consumption is one
The resource balance is the difference between ex- of the weakest aspects of national accounting in
249
low- and middle-income economies. The structure government. Caution should therefore be exer-
is estimated through household expenditure sur- cised in using the data for cross-country compari-
veys and similar survey techniques. It therefore sons. Central government expenditure comprises
shares any bias inherent in the sample frame. the expenditure by all government offices, depart-
Since, conceptually, expenditure is not identical to ments, establishments, and other bodies that are
consumption, other apparent discrepancies occur agencies or instruments of the central authority of
and data for some countries should be treated with a country. It includes both current and capital (de-
caution. For example, some countries limit surveys velopment) expenditure.
to urban areas or, even more narrowly, to capital Defense comprises all expenditure, whether by
cities. This tends to produce lower than average defense or other departments, on the maintenance
shares for food and high shares for transport and of military forces, including the purchase of mili-
communication, gross rents, fuel and power, and tary supplies and equipment, construction, re-
other consumption. Controlled food prices and in- cruiting, and training. Also in this category are
complete national accounting for subsistence activ- closely related items such as military aid programs.
ities also contribute to low food shares. Education comprises expenditure on the provi-
sion, management, inspection, and support of pre-
Table 11. Central government expenditure primary, primary, and secondary schools; of uni-
The data on central government finance in Tables versities and colleges; and of vocational, technical,
11 and 12 are from the IMF Government Finance and other training institutions. Also included is
Statistics Yearbook (1989) and IMF data files. The ac- expenditure on the general administration and reg-
counts of each country are reported using the sys- ulation of the education system; on research into
tem of common definitions and classifications its objectives, organization, administration, and
found in the IMF Manual on Government Finance methods; and on such subsidiary services as trans-
Statistics (1986). port, school meals, and school medical and dental
For complete and authoritative explanations of services. Note that Table 10 provides an alternative
concepts, definitions, and data sources, see these measure of expenditure on education, private as
IMF sources. The commentary that follows is in- well as public, relative to household consumption.
tended mainly to place these data in the context of Health covers public expenditure on hospitals,
the broad range of indicators reported in this edi- maternity and dental centers, and clinics with a
tion. major medical component; on national health and
The shares of total expenditure and current revenue medical insurance schemes; and on family plan-
by category are calculated from series in national ning and preventive care. Note that Table 10 pro-
currencies. Because of differences in coverage of vides a more comprehensive measure of expendi-
available data, the individual components of cen- ture on medical care, private as well as public,
tral government expenditure and current revenue relative to household consumption.
shown in these tables may not be strictly compara- Housing, amenities; social security and welfare cover
ble across all economies. expenditure on housing, such as income-related
Moreover, inadequate statistical coverage of schemes; on provision and support of housing and
state, provincial, and local governments dictates slum clearance activities; on community develop-
the use of central government data; this may seri- ment; and on sanitary services. These categories
ously understate or distort the statistical portrayal also cover compensation for loss of income to the
of the allocation of resources for various purposes, sick and temporarily disabled; payments to the el-
especially in countries where lower levels of gov- derly, the permanently disabled, and the unem-
ernment have considerable autonomy and are re- ployed; family, maternity, and child allowances;
sponsible for many economic and social services. and the cost of welfare services, such as care of the
In addition, "central government" can mean ei- aged, the disabled, and children. Many expendi-
ther of two accounting concepts: consolidated or tures relevant to environmental defense, such as
budgetary. For most countries, central government pollution abatement, water supply, sanitary af-
finance data have been consolidated into one over- fairs, and refuse collection, are included indistin-
all account, but for others only the budgetary cen- guishably in this category.
tral government accounts are available. Since all Economic services comprise expenditure associ-
central government units are not included in the ated with the regulation, support, and more effi-
budgetary accounts, the overall picture of central cient operation of business; economic develop-
government activities is incomplete. Countries re- ment; redress of regional imbalances; and creation
porting budgetary data are footnoted. of employment opportunities. Research, trade pro-
It must be emphasized that for these and other motion, geological surveys, and inspection and
reasons the data presented, especially those for ed- regulation of particular industry groups are among
ucation and health, are not comparable across the activities included.
countries. In many economies private health and Other covers items not included elsewhere; for a
education services are substantial; in others public few economies it also includes amounts that could
services represent the major component of total not be allocated to other components (or adjust-
expenditure but may be financed by lower levels of ments from accrual to cash accounts).
250
Total expenditure is more narrowly defined than property. Proceeds of grants and borrowing, funds
the measure of general government consumption arising from the repayment of previous lending by
given in Table 9, because it excludes consumption governments, incurrence of liabilities, and pro-
expenditure by state and local governments. At ceeds from the sale of capital assets are not in-
the same time, central government expenditure is cluded.
more broadly defined because it indudes govern- Summary measures for the components of cur-
ment's gross domestic investment and transfer rent revenue are computed from group totals for
payments. revenue components and total current revenue in
Overall surplus/deficit is defined as current and current dollars; those for current revenue as a per-
capital revenue and grants received, less total ex- centage of GNP are computed from group totals
penditure and lending minus repayments. for total current revenue and GNP in current dol-
Summary measures for the components of cen- lars. Since 1988 data are not available for more
tral government expenditure are computed from than half the countries, 1987 data are used as
group totals for expenditure components and cen- weights for the summary measures for this table.
tral government expenditure in current dollars.
Those for total expenditure as a percentage of GNP Table 13. Money and interest rates
and for overall surplus/deficit as a percentage of The data on monetary holdings are based on the
GNP are computed from group totals for the above IMF's International Financial Statistics (IFS). Mone-
total expenditures and overall surplus/deficit in tary holdings, broadly defined, comprise the mone-
current dollars, and GNP in current dollars, re- tary and quasi-monetary liabilities of a country's
spectively. Since 1988 data are not available for financial institutions to residents other than the
more than half the countries, 1987 data are used as central government. For most countries, monetary
weights for the summary measures in this table. holdings are the sum of money (IFS line 34) and
Table 12. Central government current revenue quasi-money (IFS line 35). Money comprises the
Information on data sources and comparability is economy's means of payment: currency outside
given in the note to Table 11. Current revenue by banks and demand deposits. Quasi-money com-
source is expressed as a percentage of total current prises time and savings deposits and similar bank
revenue, which is the sum of tax revenue and non- accounts that the issuer will readily exchange for
tax revenue and is calculated from national curren- money. Where nonmonetary financial institutions
cies. are important issuers of quasi-monetary liabilities,
Tax revenue comprises compulsory, unrequited, these are also included in the measure of monetary
nonrepayable receipts for public purposes. It in- holdings.
cludes interest collected on tax arrears and penal- The growth rates for monetary holdings are cal-
ties collected on nonpayment or late payment of culated from year-end figures, while the average of
taxes and is shown net of refunds and other correc- the year-end figures for the specified year and the
tive transactions. Taxes on income, profit, and capital previous year is used for the ratio of monetary
gain are taxes levied on the actual or presumptive holdings to GDP.
net income of individuals, on the profits of enter- The nominal interest rates of banks, also from IFS,
prises, and on capital gains, whether realized on represent the rates paid by commercial or similar
land sales, securities, or other assets. Social security banks to holders of their quasi-monetary liabilities
contributions include employers' and employees' (deposit rate) and charged by the banks on loans to
social security contributions as well as those of prime customers (lending rate). The data are, how-
self-employed and unemployed persons. Domestic ever, of limited international comparability partly
taxes on goods and services include general sales and because coverage and definitions vary, and partly
turnover or value added taxes, selective excises on because countries differ in the scope available to
goods, selective taxes on services, taxes on the use banks for adjusting interest rates to reflect market
of goods or property, and profits of fiscal monopo- conditions.
lies. Taxes on international trade and transactions in- Since interest rates (and growth rates for mone-
clude import duties, export duties, profits of ex- tary holdings) are expressed in nominal terms,
port or import monopolies, exchange profits, and much of the variation among countries stems from
exchange taxes. Other taxes include employers' differences in inflation. For easy reference, the Ta-
payroll or labor taxes, taxes on property, and taxes ble 1 indicator of recent inflation is repeated in this
not allocable to other categories. They may include table.
negative values that are adjustments, for instance, Table 14. Growth of merchandise trade
for taxes collected on behalf of state and local gov-
ernments and not allocable to individual tax cate- This year, a significant change has been made to
gories. the data source of Table 14. For low- and middle-
Nontax revenue comprises receipts that are not a income countries, the main data source for current
compulsory nonrepayable payment for public pur- trade values has changed from the U.N. Commod-
poses, such as fines, administrative fees, or entre- ity Trade data file to the World Bank data file. The
preneurial income from government ownership of latter includes data collected from World Bank
251
member countries before more detailed data are data system and the U.N. Yearbook of International
reported to the U.N. Trade Statistics, supplemented by other secondary
The statistics on merchandise trade for high- sources and World Bank estimates, as explained in
income countries continue to be primarily from the the note to Table 14.
U.N. trade data system, which accords with the Merchandise exports and imports are also defined
U.N. Yearbook of International Trade Statisticsthat in the note to Table 14.
is, the data are based on countries' customs re- The categorization of exports and imports fol-
turns. In some cases, data from secondary sources lows the Standard International Trade Classification
permit coverage adjustments for significant com- (SITC), Series M, No. 34, Revision 1. Estimates
ponents of a country's foreign trade not subject to from secondary sources also usually follow this
regular customs reports. Such cases are identified definition. For some countries, data for certain
in the country notes to the World Tables. Values in commodity categories are unavailable and the full
these tables are in current U.S. dollars. breakdown cannot be shown.
Merchandise exports and imports, with some ex- In Table 15, food commodities are those in SITC
ceptions, cover international movements of goods Sections 0, 1, and 4 and Division 22 (food and live
across customs borders. Exports are valued f.o.b. animals, beverages, oils and fats, and oilseeds and
(free on board) and imports c.i.f. (cost, insurance, nuts), less Division 12 (tobacco). Fuels are the com-
and freight), unless otherwise specified in the fore- modities in SITC Section 3 (mineral fuels, lubri-
going sources. These values are in current dollars; cants and related materials). Other primary commod-
note that they do not include trade in services. ities comprise SITC Section 2 (crude materials,
The growth rates of merchandise exports and excluding fuels), less Division 22 (oilseeds and
imports are in constant terms and are calculated nuts), plus Divisions 12 (tobacco) and 68 (nonfer-
from quantum indexes of exports and imports. rous metals). Machinery and transport equipment are
Quantum indexes are obtained from the export or the commodities in SITC Section 7. Other manufac-
import value index as deflated by the correspond- tures, calculated residually from the total value of
ing price index. To calculate these quantum in- manufactured imports, represent SITC Sections 5
dexes, the World Bank uses its own price indexes, through 9, less Section 7 and Division 68.
which are based on international prices for pri- In Table 16, fuels, minerals, and metals are the com-
mary commodities and unit value indexes for man- modities in SITC Section 3 (mineral fuels and lubri-
ufactures. These price indexes are country-specific cants and related materials), Divisions 27 and 28
and disaggregated by broad commodity groups. (minerals and crude fertilizers, and metalliferous
This ensures consistency between data for a group ores), and Division 68 (nonferrous metals). Other
of countries and those for individual countries. primary commodities comprise SITC Sections 0, 1, 2,
Such consistency will increase as the World Bank and 4 (food and live animals, beverages and to-
continues to improve its trade price indexes for an bacco, inedible crude materials, oils, fats, and
increasing number of countries. These growth waxes), less Divisions 27 and 28. Machinery and
rates can differ from those derived from national transport equipment are the commodities in SITC
practices because national price indexes may use Section 7. Other manufactures represent SITC Sec-
different base years and weighting procedures tions 5 through 9, less Section 7 and Division 68.
from those used by the World Bank. Textiles and clothing, representing SITC Divisions 65
The terms of trade, or the net barter terms of and 84 (textiles, yarns, fabrics, and clothing), are
trade, measure the relative movement of export shown as a subgroup of other manufactures.
prices against that of import prices. Calculated as The summary measures in Table 15 are weighted
the ratio of a country's index of average export by total merchandise imports of individual coun-
prices to its average import price index, this indica- tries in current dollars, those in Table 16 by total
tor shows changes over a base year in the level of merchandise exports of individual countries in cur-
export prices as a percentage of import prices. The rent dollars. (See the note to Table 14.)
terms of trade index numbers are shown for 1985 Table 17. OECD imports of manufactured goods:
and 1988, where 1980 = 100. The price indexes are origin and composition
from the source cited above for the growth rates of
exports and imports. The data are from the U.N., reported by high-
The summary measures for the growth rates are income OECD economies, which are the OECD
calculated by aggregating the 1980 constant U.S. members excluding Greece, Portugal, and Turkey.
dollar price series for each year and then applying The table reports the value of imports of manufac-
the least-squares growth rate procedure for the pe- tures of high-income OECD countries by the econ-
riods shown. Note again that these values do not omy of origin, and the composition of such im-
include trade in services. ports by major manufactured product groups.
The table replaces one in past editions on the
Tables 15 and 16. Structure of merchandise origin and destination of manufactured exports,
imports and exports which was based on exports reported by individ-
The shares in these tables are derived from trade ual economies. Since there was a lag of several
values in current dollars reported in the U.N. trade years in reporting by many developing economies,
252
estimates based on various sources were used to workers' remittances as a factor income receipt
fill the gaps. Until these estimates can be im- (and hence a component of GNP). The World Bank
proved, this table, based on up-to-date and consis- adheres to international guidelines in defining
tent but less comprehensive data, is included in- GNP and, therefore, may differ from national prac-
stead. Manufactured imports of the predominant tices.
markets from individual economies are the best Net direct private investment is the net amount in-
available proxy of the magnitude and composition vested or reinvested by nonresidents in enter-
of the manufactured exports of these economies to prises in which they or other nonresidents exercise
all destinations taken together. significant managerial control, including equity
Manufactured goods are the commodities in Stan- capital, reinvested earnings, and other capital. The
dard International Trade Classification (SITC), Revi- net figures are obtained by subtracting the value of
sion 1, Sections 5 through 9 (chemical and related direct investment abroad by residents of the re-
products, basic manufactures, manufactured arti- porting country.
cles, machinery and transport equipment, and Gross international reserves comprise holdings of
other manufactured articles and goods not else- monetary gold, special drawing rights (SDR5), the
where classified), excluding Division 68 (nonfer- reserve position of members in the IMF, and hold-
rous metals). This definition is somewhat broader ings of foreign exchange under the control of mon-
than the one used to define exporters of manufac- etary authorities. The data on holdings of interna-
tures. tional reserves are from IMF data files. The gold
The major manufactured product groups re- component of these reserves is valued throughout
ported are defined as follows: textiles and clothing at year-end (December 31) London prices: that is,
(SITC Sections 65 and 84), chemicals (SITC Section $37.37 an ounce in 1970 and $410.25 an ounce in
5), electrical machinery and electronics (SITC Section 1988. The reserve levels for 1970 and 1988 refer to
72), transport equipment (SITC Section 73), and oth- the end of the year indicated and are in current
ers, defined as the residual. SITC Revision 1 data dollars at prevailing exchange rates. Because of dif-
are used for the year 1968, whereas the equivalent ferences in the definition of international reserves,
data in Revision 2 are used for the year 1988. in the valuation of gold, and in reserve manage-
Table 18. Balance of payments and reserves ment practices, the levels of reserve holdings pub-
lished in national sources do not have strictly com-
The statistics for this table are mostly as reported parable significance. Reserve holdings at the end
by the IMF but do include recent estimates by of 1988 are also expressed in terms of the number
World Bank staff and, in rare instances, the Bank's of months of imports of goods and services they
own coverage or classification adjustments to en- could pay for.
hance international comparability. Values in this The summary measures are computed from
table are in U.S. dollars converted at current ex- group aggregates for gross international reserves
change rates. and total imports of goods and services in current
The current account balance after official transfers is dollars.
the difference between (a) exports of goods and
services (factor and nonfactor) as well as inflows of
Table 19. Official development assistance from
unrequited transfers (private and official) and (b) OECD & OPEC members
imports of goods and services as well as all unre- Official development assistance (ODA) consists of net
quited transfers to the rest of the world. disbursements of loans and grants made on con-
Current account balance before official transfers is the cessional financial terms by official agencies of the
current account balance that treats net official un- members of the Development Assistance Commit-
requited transfers as akin to official capital move- tee (DAC) of the Organisation for Economic Co-
ments. The difference between the two balance of operation and Development (OECD) and members
payments measures is essentially foreign aid in the of the Organization of Petroleum Exporting Coun-
form of grants, technical assistance, and food aid, tries (OPEC), to promote economic development
which, for most developing countries, tends to and welfare. Although this definition is meant to
make current account deficits smaller than the fi- exclude purely military assistance, the borderline
nancing requirement. is sometimes blurred; the definition used by the
Net workers' remittances cover payments and re- country of origin usually prevails. ODA also in-
ceipts of income by migrants who are employed or cludes the value of technical cooperation and assis-
expect to be employed for more than a year in their tance. All data shown are supplied by the OECD,
new economy, where they are considered resi- and all U.S. dollar values are converted at official
dents. These remittances are classified as private exchange rates.
unrequited transfers and are included in the bal- Amounts shown are net disbursements to devel-
ance of payments current account balance, oping countries and multilateral institutions. The
whereas those derived from shorter-term stays are disbursements to multilateral institutions are now
included in services as labor income. The distinc- reported for all DAC members on the basis of the
tion accords with internationally agreed guide- date of issue of notes; some DAC members previ-
lines, but many developing countries classify ously reported on the basis of the date of encash-
253
ment. Net bilateral flows to low-income countries ex- agencies, and autonomous public bodies. Publicly
clude unallocated bilateral flows and all guaranteed loans are external obligations of private
disbursements to multilateral institutions. debtors that are guaranteed for repayment by a
The nominal values shown in the summary for public entity. These two categories are aggregated
ODA from high-income OECD countries were in the tables. Private nonguaranteed loans are exter-
converted at 1980 prices using the dollar GDP de- nal obligations of private debtors that are not guar-
flator. This deflator is based on price increases in anteed for repayment by a public entity.
OECD countries (excluding Greece, Portugal, and Use of IMF credit denotes repurchase obligations
Turkey) measured in dollars. It takes into account to the IMF for all uses of IMF resources, excluding
the parity changes between the dollar and national those resulting from drawings in the reserve
currencies. For example, when the dollar depreci- tranche. It is shown for the end of the year speci-
ates, price changes measured in national curren- fied. It comprises purchases outstanding under the
cies have to be adjusted upward by the amount of credit tranches, including enlarged access re-
the depreciation to obtain price changes in dollars. sources, and all of the special facilities (the buffer
The table, in addition to showing totals for stock, compensatory financing, Extended Fund,
OPEC, shows totals for the Organization of Arab and oil facilities), Trust Fund loans and operations
Petroleum Exporting Countries (OAPEC). The do- under the Enhanced Structural Adjustment facili-
nor members of OAPEC are Algeria, haq, Kuwait, ties. Use of IMF credit outstanding at year-end (a
Libya, Qatar, Saudi Arabia, and United Arab Emir- stock) is converted to U.S. dollars at the dollar-
ates. ODA data for OPEC and OAPEC are also SDR exchange rate in effect at year-end.
obtained from the OECD. Short-term debt is debt with an original maturity
of one year or less. Available data permit no dis-
Table 20. Official development tinctions between public and private nonguaran-
assistance: receipts teed short-term debt.
Net disbursements of ODA from all sources consist of Total external debt is defined for the purpose of
loans and grants made on concessional financial this Report as the sum of public, publicly guaran-
terms by all bilateral official agencies and multilat- teed, and private nonguaranteed long-term debt,
eral sources to promote economic development use of IMP credit, and short-term debt.
and welfare. They include the value of technical Table 22. Flow of public and private
cooperation and assistance. The disbursements external capital
shown in this table are not strictly comparable with
those shown in Table 19 since the receipts are from Data on disbursements and repayment of principal
all sources; disbursements in Table 19 refer only to (amortization) are for public, publicly guaranteed,
those made by high-income members of the OECD and private nonguaranteed long-term loans. The
and members of OPEC. Net disbursements equal net flow estimates are disbursements less the repay-
gross disbursements less payments to the origina- ment of principal.
tors of aid for amortization of past aid receipts. Net Table 23. Total external public and private debt
disbursements of ODA are shown per capita and and debt service ratios
as a percentage of GNP.
The summary measures of per capita ODA are Data on total long-term debt outstanding and disbursed
computed from group aggregates for population in this table cover public and publicly guaranteed
and for ODA. Summary measures for ODA as a debt and private nonguaranteed debt. Debt service
percentage of GNP are computed from group to- as a percentage of exports of goods and services is one of
tals for ODA and for GNP in current U.S. dollars. several conventional measures used to assess the
ability to service debt. The average ratios of debt
Table 21. Total external debt service to GNP for the economy groups are
The data on debt in this and successive tables are weighted by GNP in current dollars. The average
from the World Bank Debtor Reporting System, ratios of debt service to exports of goods and ser-
supplemented by World Bank estimates. That sys- vices are weighted by exports of goods and ser-
tem is concerned solely with developing econo- vices in current dollars.
mies and does not collect data on external debt for Table 24. External public debt
other groups of borrowers or from economies that and debt service ratios
are not members of the World Bank. The dollar
figures on debt shown in Tables 21 through 25 are External public debt outstanding and disbursed repre-
in U.s. dollars converted at official exchange rates. sents public and publicly guaranteed loans drawn
The data on debt include private nonguaranteed at year-end, net of repayments of principal and
debt reported by twenty-five developing countries write-off s. For estimating external public debt as a
and complete or partial estimates for an additional percentage of GNP, the debt figures are converted
twenty-three countries. into U.S. dollars from currencies of repayment at
Public loans are external obligations of public end-of-year official exchange rates. GNP is con-
debtors, including the national government, its verted from national currencies to U.S. dollars by
254
applying the conversion procedure described in The projections of population for 2000, 2025, and
the technical note to Tables 2 and 3. the year in which the population will eventually
Interest payments are actual payments made on become stationary (see definition below) are made
the outstanding and disbursed public and publicly for each economy separately. Information on total
guaranteed debt in foreign currencies, goods, or population by age and sex, fertility, mortality, and
services; they include commitment charges on un- international migration is projected on the basis of
disbursed debt if information on those charges is generalized assumptions until the population be-
available. comes stationary.
Debt seruice is the sum of actual repayments of A stationary population is one in which age- and
principal (amortization) and actual payments of in- sex-specific mortality rates have not changed over
terest made in foreign currencies, goods, or ser- a long period, during which fertility rates have re-
vices on external public arid publicly guaranteed mained at replacement level; that is, when the net
debt. Procedures for estimating total long-term reproduction rate (NRR) equals 1 (the note to Table
debt as a percentage of GNP, average ratios of debt 27 defines NRR). In such a population, the birth
service to GNP, and average ratios of debt service rate is constant and equal to the death rate, the age
to exports of goods and services are the same as structure is constant, and the growth rate is zero.
those described in the note to Table 23. Population projections are made age cohort by
The summary measures are computed from age cohort. Mortality, fertility, and migration are
group aggregates of debt service and GNP in cur- projected separately and the results are applied it-
rent dollars. eratively to the 1985 base-year age structure. For
the projection period 1985 to 2005, the changes in
Table 25. Terms of external public borrowing mortality are country specific: increments in life
Commitments refer to the public and publicly guar- expectancy and decrements in infant mortality are
anteed loans for which contracts were signed in based on previous trends for each country. When
the year specified. They are reported in currencies female secondary school enrollment is high, mor-
of repayment and converted into U.S. dollars at tality is assumed to decline more quickly. Infant
average annual official exchange rates. mortality is projected separately from adult mortal-
Figures for interest rates, maturities, and grace pe- ity.
riods are averages weighted by the amounts of the Projected fertility rates are also based on pre-
loans. Interest is the major charge levied on a loan vious trends. For countries in which fertility has
and is usually computed on the amount of princi- started to decline (termed "fertility transition"),
pal drawn and outstanding. The maturity of a loan this trend is assumed to continue. It has been ob-
is the interval between the agreement date, when a served that no country with a life expectancy of
loan agreement is signed or bonds are issued, and less than 50 years has experienced a fertility de-
the date of final repayment of principal. The grace cline; for these countries fertility transition is de-
period is the interval between the agreement date layed, and then the average decline of the group of
and the date of the first repayment of principal. countries in fertility transition is applied. Coun-
Public loans with variable interest rates, as a percent- tries with below-replacement fertility are assumed
age of public debt, refer to interest rates that float to have constant total fertility rates until 1995-
with movements in a key market rate; for example, 2000 and then to regain replacement level by 2030.
the London interbank offered rate (LIBOR) or the International migration rates are based on past
U.S. prime rate. This column shows the borrow- and present trends in migration flows and migra-
er's exposure to changes in international interest tion policy. Among the sources consulted are esti-
rates. mates and projections made by national statistical
The summary measures in this table are offices, international agencies, and research insti-
weighted by the amounts of the loans. tutions. Because of the uncertainty of future migra-
Table 26. Population growth and projections tion trends, it is assumed in the projections that
net migration rates wifi reach zero by 2025.
The introduction outlines the changes to the next The estimates of the size of the stationary popu-
seven tables. lation are speculative. They should not be regarded as
Population growth rates are period averages cal- predictions. They are included to show the implica-
culated from midyear populations. tions of recent fertility and mortality trends on the
Population estimates for mid-1988 and estimates basis of generalized assumptions. A fuller descrip-
of fertility and mortality are made by the World tion of the methods and assumptions used to cal-
Bank from data provided by the U.N. Population culate the estimates is contained in the World
Division, the U.N. Statistical Office, and country Bank's World Population Projections, 1989-90
statistical offices. Estimates take into account the Edition.
results of the latest population censuses, which, in Table 27. Demography and fertility
some cases, are neither recent nor accurate. Note
that refugees not permanently settled in the coun- The crude birth rate and crude death rate indicate re-
try of asylum are generally considered to be part of spectively the number of live births and deaths
the population of their country of origin. occurring per thousand population in a year. They
255
come from the sources mentioned in the note to auxiliary nurses, as well as paraprofessional per-
Table 26. sonnel such as health workers, first aid workers,
Women of childbearing age as a percentage of popula- traditional birth attendants, and so on. The inclu-
tion refers to women age 15 to 49. sion of auxiliary and paraprofessional personnel
The total fertility rate represents the number of provides more realistic estimates of available nurs-
children that would be born to a woman if she ing care. Because definitions of doctors and nurs-
were to live to the end of her childbearing years ing personnel vary-and because the data shown
and bear children at each age in accordance with are for a variety of yearsthe data for these two
prevailing age-specific fertility rates. The rates indicators are not strictly comparable across coun-
given are from the sources mentioned in Table 26. tries.
The net reproduction rate (NRR), which measures Data on births attended by health staff show the
the number of daughters a newborn girl wifi bear percentage of births recorded where a recognized
during her lifetime, assuming fixed age-specific health service worker was in attendance. The data
fertility and mortality rates, reflects the extent to are from WHO and supplemented by UNICEF
which a cohort of newborn girls will reproduce data; they are based on national sources. The data
themselves. An NRR of 1 indicates that fertility is are derived mostly from official community reports
at replacement level: at this rate women wifi bear, and hospital records, and some reflect only births
on average, only enough daughters to replace in hospitals and other medical institutions. Some-
themselves in the population. As with the size of times smaller private and rural hospitals are ex-
the stationary population, the assumed year of cluded, and sometimes even relatively primitive
reaching replacement-level fertility is speculative local facilities are included. The coverage is there-
and should not be regarded as a prediction. fore not always comprehensive, and the figures
Married women of childbearing age using contraception should be treated with extreme caution.
refers to women who are practicing, or whose hus- The percentage of babies with low birth weight re-
bands are practicing, any form of contraception. lates to children born weighing less than 2,500
Contraceptive usage is generally measured for grams. Low birth weight is frequently associated
women age 15 to 49. A few countries use measures with maternal malnutrition and tends to raise the
relating to other age groups, especially 15 to 44. risk of infant mortality and to lead to poor growth
Data are mainly derived from demographic and in infancy and childhood, thus increasing the inci-
health surveys, contraceptive prevalence surveys, dence of other forms of retarded development.
World Bank country data, and Mauldin and Se- The figures are derived from both WHO and
gal's article "Prevalence of Contraceptive Use: UNICEF sources and are based on national data.
Trends and Issues" in volume 19 of Studies in Fam- The data are not strictly comparable across coun-
ily Planning (1988). For a few countries for which tries since they are compiled from a combination of
no survey data are available, program statistics are surveys and administrative records that may not
used; these include India and several African have representative national coverage.
countries. Program statistics may understate con- The infant mortality rate is the number of infants
traceptive prevalence because they do not measure who die before reaching one year of age, per thou-
use of methods such as rhythm, withdrawal, or sand live births in a given year. The data are from
abstinence, or of contraceptives not obtained the U.N. publication Mortality of Children under Age
through the official family planning program. The 5: Projections, 1950-2025 as well as from the World
data refer to rates prevailing in a variety of years, Bank.
generally not more than two years before the year The daily calorie supply per capita is calculated by
specified in the table. dividing the calorie equivalent of the food supplies
All summary measures are country data in an economy by the population. Food supplies
weighted by each country's share in the aggregate comprise domestic production, imports less ex-
population. ports, and changes in stocks; they exclude animal
Table 28. Health and nutrition feed, seeds for use in agriculture, and food lost in
processing and distribution. These estimates are
The estimates of population per physician and per from the Food and Agriculture Organization.
nursing person are derived from World Health Or- The summary measures in this table are country
ganization (WHO) data and are supplemented by figures weighted by each country's share in the
data obtained directly by the World Bank from na- aggregate population.
tional sources. The data refer to a variety of years,
Table 29. Education
generally no more than two years before the year
specified. The figure for physicians, in addition to The data in this table refer to a variety of years,
the total number of registered practitioners in the generally not more than two years distant from
country, includes medical assistants whose medi- those specified; however, figures for females
cal training is less than that of qualified physicians sometimes refer to a year earlier than that for over-
but who nevertheless dispense similar medical ser- all totals. The data are mostly from Unesco.
vices, including simple operations. Nursing per- Primary school enrollment data are estimates of
sons include graduate, practical, assistant, and children of all ages enrolled in primary school. Fig-
256
ures are expressed as the ratio of pupils to the pared by national statistical offices. The results are
population of school-age children. Although many coordinated by the U.N. Statistical Office (UNSO)
countries consider primary school age to be 6 to 11 with support from other international agencies,
years, others do not. The differences in country particularly the Statistical Office of the European
practices in the ages and duration of schooling are Communities (Eurostat) and the Organisation for
reflected in the ratios given. For some countries Economic Co-operation and Development
with universal primary education, the gross enroll- (OECD). The World Bank, the Economic Commis-
ment ratios may exceed 100 percent because some sion for Europe, and the Economic and Social
pupils are younger or older than the country's Commission for Asia and the Pacific also contrib-
standard primary school age. ute to this exercise.
The data on secondary school enrollment are cal- A total of sixty-four countries participated in ICP
culated in the same manner, but again the defini- Phase V. and preliminary results are now available
tion of secondary school age differs among coun- for fifty-seven. For one country (Nepal), total GDP
tries. It is most commonly considered to be 12 to 17 data were not available, and comparisons were
years. Late entry of more mature students as well made for consumption only; two countries with
as repetition and the phenomenon of "bunching" populations of less than 1 millionLuxembourg,
in final grades can influence these ratios. with 81.3 as its estimated index of GDP per capita;
The tertiary enrollment ratio is calculated by di- and Swaziland, with 13.6have been omitted
viding the number of pupils enrolled in all postsec- from this table. Data for the remaining seven coun-
ondary schools and universities by the population tries, all Caribbean, are expected soon.
in the 20-24 age group. Pupils attending vocational Although the GDP per capita figures are pre-
schools, adult education programs, two-year com- sented as indexes to the U.S. value, the underlying
munity colleges, and distance education centers data are expressed in U.S. dollars. However, these
(primarily correspondence courses) are included. dollar values, which are different from those
The distribution of pupils across these different shown in Tables 1 and 3 (see the technical notes for
types of institutions varies among countries. The these tables), are obtained by special conversion
youth populationthat is, 20 to 24 yearshas been factors designed to equalize purchasing powers of
adopted by Unesco as the denominator since it currencies in the respective countries. This conver-
represents an average tertiary level cohort even sion factor, commonly known as the purchasing
though people above and below this age group power parity (PPP), is defined as the number of
may be registered in tertiary institutions. units of a country's currency required to buy the
Primary net enrollment is the percentage of school- same amounts of goods and services in the domes-
age children who are enrolled in school. Unlike tic market as one dollar would buy in the United
gross enrollment, the net ratios correspond to the States. The computation of PPPs involves obtain-
country's primary school age group. This indicator ing implicit quantities from national accounts ex-
gives a much clearer idea of how many children in penditure data and specially collected price data
the age group are actually enrolled in school, with- and revaluing the implicit quantities in each coun-
out the number being inflated by over- (or under-) try at a single set of average prices. The PPP rate
age children. thus equalizes dollar prices in every country, and
The primary pupil-teacher ratio is the number of intercountry comparisons of GDP based on them
pupils enrolled in school in a country, divided by reflect differences in quantities of goods and ser-
the number of teachers in the education system. vices free of any price-level differentials. This pro-
The summary measures in this table are country cedure is designed to bring intercountry compari-
enrollment rates weighted by each country's share sons in line with intertemporal real value
in the aggregate population. comparisons that are based on constant price se-
ries.
Table 30. Income distribution and ICP The figures presented here are the results of a
estimates of GDP two-step exercise. Countries within a region or
The data in this table refer to the ICP estimates of group such as the OECD are first compared using
GDP and the distribution of income or expenditure their own group average prices. Next, since group
accruing to percentile groups of households average prices may differ from each other, making
ranked by total household income, per capita in- the countries belonging to different groups not
come, or expenditure. comparable, the group prices are adjusted to make
The first column presents preliminary results of them comparable at the world level. The adjust-
the U.N. International Comparison Program (ICP), ments, done by UNSO, are based on price differ-
Phase V, for 1985. ICP recasts traditional national entials observed in a network of "link" countries
accounts through special price collections and dis- representing each group. However, the linking is
aggregation of GDP by expenditure components. done in a manner that retains in the world compar-
More comprehensive ICP results are expected to ison the relative levels of GDP observed in the
be available by the mid-1990s. The figures given group comparisons.
here are subject to change and should be regarded The two-step process was adopted because the
as indicative only. ICP Phase V details are pre- relative GDP levels and ranking of two countries
257
may change when more countries are brought into Similarly, the scope of the indicator is limited for
the comparison. It was felt that this should not be certain countries, and data for other countries are
allowed to happen within geographic regions; that not fully comparable. Because households vary in
is, that the relationship of, say, Ghana and Senegal size, a distribution in which households are
should not be affected by the prices prevailing in ranked according to per capita household income,
the United States. Thus overall GDP per capita rather than according to total household income, is
levels are calculated at "regional" prices and then superior for many purposes. The distinction is im-
linked together. The linking is done by revaluing portant because households with low per capita
GDPs of all the countries at average "world" incomes frequently are large households, whose
prices and reallocating the new regional totals on total income may be high, whereas many house-
the basis of each country's share in the original holds with low household incomes may be small
comparison. households with high per capita income. Informa-
Such a method does not permit the comparison tion on the distribution of per capita household
of more detailed quantities (such as food consump- income exists for only a few countries and is infre-
tion). Hence these subaggregates and more de- quently updated. Where possible, distributions are
tailed expenditure categories are calculated using ranked according to per capita income; where this
world prices. These quantities are indeed compa- is not possible, distributions are ranked by per cap-
rable internationally, but they do not add up to the ita expenditure, household income, or household
indicated GDPs because they are calculated at a expenditure. Since the size of household is likely
different set of prices. to be small for low-income households (for in-
Some countries belong to several regional stance, single-person households and couples
groups. Some groups have priority; others are without children), the distribution of household
equal. Thus fixity is always maintained between income may overstate the income inequality. Also,
members of the European Communities, even since household savings tend to increase faster as
within the OECD and world comparison. For Fin- income levels increase, the distribution of expendi-
land and Austria, however, the bilateral relation- ture is inclined to understate the income inequal-
ship that prevails within the OECD comparison is ity. The World Bank's Living Standards Measure-
also the one used within the global comparison. ment Study and the Social Dimensions of
However, a significantly different relationship Adjustment project, covering Sub-Saharan African
(based on Central European prices) prevails in the countries, are assisting a few countries in improv-
comparison within that group, and this is the rela- ing their collection and analysis of data on income
tionship presented in the separate publication of distribution.
the European comparison.
For further details on the ICP procedures, read- Table 31. Urbanization
ers may consult the ICP Phase IV report, World The data on urban population as a percentage of total
Comparisons of Purchasing Power and Real Product for population are from the U.N. publication Prospects of
1980 (New York: United Nations, 1986). World Urbanization, 1988, supplemented by data
The income distribution data cover rural and ur- from the World Bank.
ban areas for all countries. The data refer to differ- The growth rates of urban population are calcu-
ent years between 1979 and 1988 and are drawn lated from the World Bank's population estimates;
from a variety of sources. These include the Eco- the estimates of urban population shares are calcu-
nomic Commission for Latin America and the Ca- lated from the sources just cited. Data on urban
ribbean, the Luxembourg Income Study, the agglomeration in large cities are from the U.N. 's
OECD, the U.N.'s NationalAccounts Statistics: Com- Patterns of Urban and Rural Population Growth, 1980.
pendium of Income Distribution Statistics, 1985, the Because the estimates in this table are based on
World Bank, and national sources. Data for many different national definitions of what is urban,
countries have been updated, and some of the in- cross-country comparisons should be interpreted
come distribution data previously published have with caution. Data on urban agglomeration in
been deleted because they refer to years long past. large cities are from population censuses.
In many countries the collection of income distri- The summary measures for urban population as
bution data is not systematically organized or inte- a percentage of total population are calculated
grated with the official statistical system. The data from country percentages weighted by each coun-
are derived from surveys designed for other pur- try's share in the aggregate population; the other
poses, most often consumer expenditure surveys, summary measures in this table are weighted in
that also collect information on income. These sur- the same fashion, using urban population.
veys use a variety of income concepts and sample
designs, and in many cases their geographic cover- Table 32. Women in development
age is too limited to provide reliable nationwide This table provides some basic indicators disaggre-
estimates of income distribution. Although the gated to show differences between the sexes that
data presented here represent the best available illustrate the condition of women in society. The
estimates, they do not avoid all these problems measures reflect the demographic status of women
and should be interpreted with caution. and their access to health and education services.
258
Statistical anomalies become even more apparent these two series are drawn from diverse national
when social indicators are analyzed by gender, be- sources and collected by the World Health Organi-
cause reporting systems are often weak in areas zation (WHO), although many national adminis-
related specifically to women. Indicators drawn trative systems are weak and do not record vital
from censuses and surveys, such as those on pop- events in a systematic way. The data are derived
ulation, tend to be about as reliable for women as mostly from official community reports and hospi-
for men; but indicators based largely on adminis- tal records, and some reflect only deaths in hospi-
trative records, such as maternal and infant mor- tals and other medical institutions. Sometimes
tality, are less reliable. More resources are now be- smaller private and rural hospitals are excluded,
ing devoted to develop better information on these and sometimes even relatively primitive local facil-
topics, but the reliability of data, even in the series ities are included. The coverage is therefore not
shown, stifi varies significantly. always comprehensive, and the figures should be
The risk of dying ly age 5 (also referred to as the treated with extreme caution.
under-5 mortality rate) shows the probability of a Clearly, many maternal deaths go unrecorded,
newborn baby dying before reaching age 5. The particularly in countries with remote rural popula-
rates are derived from life tables based on esti- tions; this accounts for some of the very low num-
mated current life expectancy at birth and on in- bers shown in the table, especially for several Afri-
fant mortality rates. In general throughout the can countries. Moreover, it is not clear whether an
world more males are born than females. Under increase in the number of mothers in hospital re-
good nutritional and health conditions and in flects more extensive medical care for women or
times of peace, male children under 5 have a more complications in pregnancy and childbirth
higher death rate than females. These columns because of poor nutrition, for instance. (Table 28
show that female-male differences in the risk of shows data on low birth weight.)
dying by age 5 vary substantially. In industrial These time series attempt to bring together read-
market economies, female babies have a 23 percent ily available information not always presented in
lower risk of dying by age 5 than male babies; the international publications. WHO warns that there
risk of dying by age 5 is actually higher for females are inevitably gaps in the series, and it has invited
than for males in some lower-income economies. countries to provide more comprehensive figures.
This suggests differential treatment of males and They are reproduced here, from the 1986 WHO
females with respect to food and medical care. publication Maternal Mortality Rates, supplemented
Such discrimination particularly affects very by the UNICEF publication The State of the World's
young girls, who may get a smaller share of scarce Children 1989, as part of the international effort to
food or receive less prompt costly medical atten- highlight data in this field. The data refer to any
tion. This pattern of discrimination is not uni- year from 1977 to 1984.
formly associated with development. There are The education indicators, based on Unesco
low- and middle-income countries (and regions sources, show the extent to which females have
within countries) where the relative risk of dying equal access to schooling.
by age 5 for females relative to males approximates Persistence to grade 4 is the percentage of children
the pattern found in industrial countries. In many starting primary school in 1970 and 1984, respec-
other countries, however, the numbers starkly tively, who continued to the fourth grade by 1973
demonstrate the need to associate women more and 1987. Figures in italics represent earlier or later
closely with development. The health and welfare cohorts. The data are based on enrollment records.
indicators in both Table 28 and in this table's ma- The slightly higher persistence ratios for females in
ternal mortality column draw attention, in particu- some African countries may indicate male partici-
lar, to the conditions associated with childbearing. pation in activities such as animal herding.
This activity still carries the highest risk of death All things being equal, and opportunities being
for women of reproductive age in developing the same, the ratios for females per 100 males should
countries. The indicators reflect, but do not mea- be close to 100. However, inequalities may cause
sure, both the availability of health services for the ratios to move in different directions. For ex-
women and the general welfare and nutritional ample, the number of females per 100 males will
status of mothers. rise at secondary school level if male attendance
Life expectancy at birth is defined in the note to declines more rapidly in the final grades because of
Table 1. males' greater job opportunities, conscription into
Maternal mortality refers to the number of female the army, or migration in search of work. In addi-
deaths that occur during childbirth, per 100,000 tion, since the numbers in these columns refer
live births. Because deaths during childbirth are mainly to general secondary education, they do
defined more widely in some countries to include not capture those (mostly males) enrolled in tech-
complications of pregnancy or the period after nical and vocational schools or in full-time appren-
childbirth, or of abortion, and because many preg- ticeships, as in Eastern Europe.
nant women die because of lack of suitable health All summary measures are country data
care, maternal mortality is difficult to measure con- weighted by each country's share in the aggregate
sistently and reliably across countries. The data in population.
259
Data sources
Production U.N. Department of International Economic and Social Affairs. Various years. Statistical
and domestic Yearbook. New York.
absorption . Various years. World Energy Supplies. Statistical Papers, series J. New York.
International Comparison Program Phases IV (1980) and V (1985) reports, and data from
ECE, ESCAP, Eurostat, OECD, and U.N.
FAQ, IMF, UNIDO, and World Bank data; national sources.
Fiscal and International Monetary Fund. 1989. Government Finance Statistics Yearbook. Vol. 11.
monetary Washington, D.C.
accounts . Various years. International Financial Statistics. Washington, D.C.
U.N. Department of International Economic and Social Affairs. Various years. World Energy
Supplies. Statistical Papers, series J. New York.
IMF data.
Core International Monetary Fund. Various years. International Financial Statistics. Washington,
international D.C.
transactions U.N. Conference on Trade and Development. Various years. Handbook of International Trade
and Development Statistics. Geneva.
U.N. Department of International Economic and Social Affairs. Various years. Monthly
Bulletin of Statistics. New York.
Various years. Yearbook of International Trade Statistics. New York.
FAQ, IMF, U.N., and World Bank data.
External Qrganisation for Economic Co-operation and Development. Various years. Development
finance Co-operation. Paris.
1988. Geographical Distribution of Financial Flows to Developing Countries. Paris.
IMF, OECD, and World Bank data; World Bank Debtor Reporting System.
Human Institute for Resource Development/Westinghouse. 1987. Child Survival: Risks and the Road
resources to Health. Columbia, Md.
Mauldin, W. Parker, and Sheldon J. Segal. 1988. "Prevalence of Contraceptive Use: Trends
and Issues. "Studies in Family Planning 19, no. 6: 335-53.
Sivard, Ruth. 1985. WomenA World Survey. Washington, D.C.: World Priorities.
U.N. Department of International Economic and Social Affairs. Various years. Demographic
Yearbook. New York.
Various years. Population and Vital Statistics Report. New York.
Various years. Statistical Yearbook. New York.
1980. Patterns of Urban and Rural Population Growth. New York.
1984. Recent Levels and Trends of Contraceptive Use as Assessed in 1983. New York.
1988. Mortality of Children under Age 5: Projections 1950-2025. New York.
1989. Prospects of World Urbanization. New York.
1989. World Population Prospects: 1988. New York.
U.N. Educational Scientific and Cultural Organization. Various years. Statistical Yearbook.
Paris.
UNICEF. 1989. The State of the World's Children 1989. Oxford: Oxford University Press.
World Health Organization. Various years. World Health Statistics Annual. Geneva.
1986. Maternal Mortality Rates: A Tabulation of Available Information, 2nd edition.
Geneva.
Various years. World Health Statistics Report. Geneva.
FAQ and World Bank data.
260
The World Bank
More than a billion people in the developing world live in poverty, and millions in
Sub-Saharan Africa and Latin America are experiencing increasing deprivation.
Ten years after World Development Report 1980 looked at the cuses and extent of
poverty, this thirteenth annual Report returns to the subject to examine what has
gone right and wrong and to look toward the future. It recommends that countries
and the development community pursue a twofold strategy for reducing poverty
a strategy that has already shown promise.
The first necessity is to promote broadly based growth that will generate in-
come-earning opportunities for the poor. The second is to ensure, by improving
access to education, health care, and other social services, that the poor can take
advantage of these opportunities. Transfers and other special arrangements are
needed for those who are not able to benefit fully from the increased opportuni-
tiesfor example, the needy aged, the ill, and the disabledand those buffeted by
economic shocks and setbacks. But the main elements of the strategy are designed
to enable the poor to make effective use of their principal assettheir ability to
work. The Report urges the development community to allocate larger shares of aid
budgets to those countries which have shown in their policies a genuine commit-
ment to reducing poverty.
The world economy emerged from the shocks and transformations of the
1980s in better shape than might have been expected. The current trends, if they
continue, would create a good economic foundation for meeting the challenge of
poverty and creating brighter prospects for large populations in the developing
world in the last years of the twentieth century. Great progress has already been
made in the past three decades. Even during the so-called lost decade of the 98Os,
living standards continued to improve for most of the world's poor. What is needed
now is political commitmenton the part of the developing countries, the irdus-
- trial -countries, and the international community.
Like previous issues, this Report contains a World Development Indicators
-annex with comprehensive, up-to-date data on social and economic development in
more than 180 countries and territories. These data will also be available on diskette
for use with personal computers.
Cover design by Joyce C. Petruzzelli ISBN 0-19-520851-X (PB)
ISBN 0-19-520850-1 (HC)