Front Cover
 Title Page
 Table of Contents
 List of tables and figures
 List of contributors
 Part 1: Global and regional population...
 Global and regional population...
 Part 2: Global and regional food...
 The outlook for world food and...
 Global and regional food demand...
 Global and regional food supply,...
 Global and regional food supply,...
 Part 3: Selected issues affecting...
 Future supplies of land and water...
 The growth potential of existing...
 Prospects of meeting future food...
 Plant breeding and biotechnology...
 Climate change and agriculture:...
 Back Cover
 Reprint permission notice

Group Title: Population and food in the early twenty-first century : meeting future food demands of an increasing population
Title: Population and food in the early twenty-first century
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00085347/00001
 Material Information
Title: Population and food in the early twenty-first century meeting future food demands of an increasing population
Physical Description: xiv, 239 p. : ill. ; 28 cm.
Language: English
Creator: Islam, Nurul, 1929-
Publisher: International Food Policy Research Institute
Place of Publication: Washington D.C
Publication Date: c1995
Subject: Food supply -- Forecasting -- Congresses   ( lcsh )
Food consumption forecasting -- Congresses   ( lcsh )
Population forecasting -- Congresses   ( lcsh )
Twenty-first century -- Forecasts -- Congresses   ( lcsh )
Alimentos (industria)   ( larpcal )
Precos agricolas   ( larpcal )
Aliments -- Approvisionnement -- Prévision -- Congrès   ( rvm )
Aliments -- Consommation -- Prévision -- Congrès   ( rvm )
Prévision démographique -- Congrès   ( rvm )
Vingt et unième siècle -- Prévisions -- Congrès   ( rvm )
Genre: bibliography   ( marcgt )
conference publication   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references.
Statement of Responsibility: edited by Nurul Islam.
General Note: Papers presented at a roundtable held Feb. 1994.
 Record Information
Bibliographic ID: UF00085347
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 32665892
lccn - 95022133
isbn - 0896293319

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
    List of tables and figures
        Page vii
        Page viii
        Page ix
        Page x
        Page xi
        Page xii
        Page xiii
    List of contributors
        Page xiv
        Page 1
        Page 2
        Page 3
        Page 4
    Part 1: Global and regional population projections
        Page 5
        Page 6
    Global and regional population projections to 2025
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Part 2: Global and regional food demand and supply prospects
        Page 23
        Page 24
    The outlook for world food and agriculture to year 2010
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
    Global and regional food demand and supply prospects
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
    Global and regional food supply, demand, and trade prospects to 2010
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
    Global and regional food supply, demand, and trade prospects to 2010
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
    Part 3: Selected issues affecting food supply
        Page 141
        Page 142
    Future supplies of land and water for world agriculture
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
    The growth potential of existing agricultural technology
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
    Prospects of meeting future food needs through new technology
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
    Plant breeding and biotechnology for meeting future food needs
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
    Climate change and agriculture: Research findings and policy considerations
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
    Back Cover
        Page 240
        Page 241
    Reprint permission notice
Full Text

and Food
in the Early
Meeting Future Food Demand
of an Increasing Population

Edited by Nurul Islam


As part of its publications program, the International Food Policy Research Institute issues
proceedings of workshops or conferences, literature reviews, topical papers, and other special
reports IFPRI believes may be of interest to its audience. These occasional papers undergo peer
review prior to publication.



Edited by Nurul Islam

International Food Policy Research Institute (IFPRI)
Washington, D.C.

Copyright 1995 International Food Policy
Research Institute

All rights reserved. Sections of this report may be
reproduced without the express permission of but with
acknowledgment to the International Food Policy
Research Institute.

Library of Congress Cataloging-in-Publication Data

Population and food in the early twenty-first century:
meeting future food demand of an increasing popula-
tion / edited by Nurul Islam
p. cm. (Occasional papers)
Includes bibliographical references.
ISBN 0-89629-331-9
1. Food supply-Forecasting-Congresses.
2. Food consumption forecasting-Congresses.
3. Population forecasting-Congresses. 4. Twenty-
first century-Forecasts-Congresses. I. Islam,
Nurul, 1929- II. Series: Occasional papers
(International Food Policy Research Institute)
TX345.P64 1995 95-22133
338.1'9'0112---dc20 CIP


Foreword xiii
List of Contributors xiv
Chapter 1. Overview 1

PART I: Global and Regional Population Projections 5
Chapter 2. Global and Regional Population Projections to 2025 7
Comment on Chapter 2: How Reliable Are the Population Projections? 17

PART II: Global and Regional Food Demand and Supply Prospects 23
Chapter 3. The Outlook for World Food and Agriculture to Year 2010 25
Chapter 4. Global and Regional Food Demand and Supply Prospects 49
Chapter 5. Global and Regional Food Supply, Demand, and Trade Prospects to 2010 61
Appendix to Part II: Comparisons of Alternative Supply and Demand Projections 85
Comments on Part II:
Methodological Issues 91
Global Outlook 103
OECD Perspective 109
Eastern Europe and the Former Soviet Union 113
Asia 117
Sub-Saharan Africa 121
Latin America and the Caribbean 127

Middle East/North Africa 131

PART III: Selected Issues Affecting Future Food Supply 141
Chapter 6. Future Supplies of Land and Water for World Agriculture 143
Comment on Chapter 6: Future Supplies of Land and Water Are Fast Approaching Depletion 161
Chapter 7. The Growth Potential of Existing Agricultural Technology 167
Comment on Chapter 7: Growth Potential of Existing Technology Is Insufficiently Tapped 191
Chapter 8. Prospects of Meeting Future Food Needs Through New Technology 207
Chapter 9. Plant Breeding and Biotechnology for Meeting Future Food Needs 221
Chapter 10. Climate Change and Agriculture-Research Findings and Policy Considerations 229

Tables and Figures


Chapter 2. Global and Regional Population Projections to 2025
2.1 Population growth, 1900-90 7
2.2 Population growth, 1990-2100 8
2.3 Projected population growth 1990-2025, based on United Nations medium variant 12
2.4 Projected population growth for 2025, based on United Nations low and high variants 14
2.5 Urban population growth, 1950-2025 14
2.6 Effect of AIDS on projected population growth, 1985-2005 15
Comment on Chapter 2: How Reliable Are the Population Projections?
1. Population projects 18
Chapter 3. The Outlook for World Food and Agriculture to Year 2010
3.1 Per capital production of cereals (with rice included in milled terms) 27
3.2 Population, per capital food supplies and agricultural production (all products) 28
3.3 Per capital food supplies for direct human consumption 32
3.4 Total cereals, possible developed country outcomes in a world context (million tons,
with rice milled) 34
3.5 Outcome for all cereals (including rice in milled form), 93 developing countries 36
3.6 Feed use of cereals and livestock production 38
3.7 Meat and milk production and consumption, 93 developing countries 40
3.8 Roots, tubers, and plantains (starchy foods) in the developing countries 41
3.9 Growth rate of production and demand for pulses and oilcrops in the developing countries 42
3.10 Value of agricultural trade, all developing countries, 1961-63 to 1988-90 42
3.11 Land and irrigation in crop production use, developing countries 43
3.12 Cereal yields in major agro-ecological land classes and intercountry differences, for
developing countries, excluding China 44
3.13 Intercountry gaps in average yields for wheat and rice, developing countries, excluding China 46
Chapter 4. Global and Regional Food Demand and Supply Prospects
4.1 Historical and projected grain production and consumption growth rates, 1960-2010 52
4.2 Historical and projected grain net imports, 1960-2010 53
4.3 Historical and projected per capital grain-consumption levels, 1960-2010 54
4.4 Share of wheat, coarse grains, and rice of total grain consumption, 1960-2010 56
4.5 Agricultural model variables 58
4.6 Model regions 59
Chapter 5. Global and Regional Food Supply, Demand, and Trade Prospects to 2010
5.1 Average annual growth rates of food production, by region 62

5.2 Average annual growth rates in total domestic use of food crops and livestock products,
1966-89 63
5.3 Average share of food, feed, and other uses to total domestic use of food crops,
1966-89 66
5.4 Ratio of foodgrain exports and imports to production, 1966-89 68
5.5 Countries, regions, and commodities covered by the IFPTSIM model 69
5.6 Projected world price levels under alternative scenarios, 1988-2010 71
5.7 Baseline projections for production and consumption of various commodities, by country
or region, 1988, 2000, and 2010 72
5.8 Per capital demand of various commodities under alternative scenarios, by region, 1988-2010 74
5.9 Production and consumption of various commodities under alternative scenarios, by region, 2010 78
Appendix to Part II: Comparisons of Alternative Supply and Demand Projections
1. Data for 1989-91 and projections to 2010 for all cereals (rounded to the nearest million tons,
with rice milled) 86
2. Annual percentage growth rates of production and total use of all cereals 86
Comment on Part II: Methodological Issues
1. Adoption and innovation scenarios 92
2. Philippine Rice Yields, 1964-90 92
3. Summary of FAO projections of rice yields for developing countries, 1969-2010, ranked and
grouped by average rice yield (paddy) 93
4. Average annual rice yields in Asia by country, 1950-87 94
5. Average annual rice yield growth rates in Asia by country, 1950-87 95
6. Per capital annual consumption, Japanese nonfarm households, 1991 96
7. Total cereal imports for selected Asian countries, 1964-91 (three-year averages), and net
imports for developed and developing countries 96
8. Historical and projected per capital cereal consumption levels, 1960-2010 97
9. Per capital food supply for direct human consumption and implicit income elasticities 97
10. Selected rice and wheat income elasticity estimates for Bangladesh, India, Indonesia,
Pakistan, and Sri Lanka 98
11. Time-series information on growth rates for income for Bangladesh, India, Indonesia,
Pakistan, and Sri Lanka, 1970-90 100
12. Family calorie availability and calorie intakes per capital by food group and expenditure
quartile for Kenya and the Philippines 100
Comment on Part II: Global Outlook
1. Comparison of product and geographical scope for Chapters 3-5 103
2. Comparison of modeling approaches 104
3. Comparison of projected real prices 106
4. Comparison of cereal consumption and production growth 107
5. Comparison of cereal net imports by country group 107
Comment on Part II: OECD Perspective
1. World food production and trade 109

2. World grain trade assistance 110
3. World cereal production indicators 111
Comment on Part II: Eastern Europe and the Former Soviet Union
1. Comparative base data and growth rates for grain from three projections for Eastern
Europe and the former Soviet Union 113
2. Projections for cereal production, domestic consumption, and net trade for Eastern
Europe and the former Soviet Union for 2010 114
Comment on Part II: Asia
1. Assumed annual growth rates of gross domestic product 117
2. Human consumption demands for Asia, 2020 118
Comment on Part II: Latin America and the Caribbean
1. Share of production in the Americas, by region and agricultural product, 1990 127
2. Share of world production, by region and cereal, 1989-91 128
3. Share of LAC agricultural exports to the United States, 1982 and 1990 128
Comment on Part II: Middle East/North Africa
1. Shifts in sources of protein and energy in human diets, 1969-71 to 1988-90 (three-year
averages), for selected countries of the Middle East/North Africa region and four
industrialized countries 131
2. Shifts in per capital production and consumption of all grains for feed and food, including
pulses, in selected countries of Middle East/North Africa 132
3. Per capital production of all grains, including pulses, in the years 2000 and 2010, based on
population projections and annual growth rates in production of 0, 1, 2, and 3 percent
from 1988-92 levels 134
4. Shifts in livestock units and feed, 1968-72 to 1988-92 (five-year averages), in selected countries 135
5. Diet compositions of ruminant animals, by feed category, for 1968-72 and 1988-92 (five-year
means), for selected countries of the Middle East/North Africa 135
6. Projected surpluses and deficits in feeds to the years 2000 and 2010, given estimated
domestic feed production in 1988-92 and annual increases of 1, 2, and 3 percent per year,
in 12 countries 136
Chapter 6. Future Supplies of Land and Water for World Agriculture
6.1 Present and potential cropland in 92 developing countries 146
6.2 Land degradation categories 152
6.3 Degradation losses of potential land productivity in dry areas 153
Comment on Chapter 6: Future Supplies of Land and Water Are Fast Approaching Depletion
1. World grain production and fertilizer use, 1950-93 165
Chapter 7. The Growth Potential of Existing Agricultural Technology
7.1 Estimates of potential for expanding cultivated area 168
7.2 Growth rates of arable and permanent crop area, and of irrigated area, 1961-90 170
7.3 Increments to arable area, 1974-89 171
7.4 Technological options for achieving sustainable agriculture 172
7.5 Average area and yield of rice, wheat, barley, maize, sorghum, and millet, 1961-63 and
1989-91 177

7.6 Average yields in developed and developing countries, 1989-91, and world
national record 178
7.7 Growth rates of area and yield by region, 1961-91, for rice, wheat, barley, maize,
sorghum, and millet 179
7.8 Area, yield, fertilizer use per hectare, and percent area under modem varieties of wheat 181
7.9 Average fertilizer consumption per hectare, 1961-63 and 1988-90 183
7.10 Area, yields, and growth rates for roots and tubers, food legumes, leguminous oilseeds,
1961-63 and 1989-91 184
7.11 Growth rate of fertilizer consumption per hectare, 1961-90 186
Comment on Chapter 7: Growth Potential of Existing Technology Is Insufficiently Tapped
1. Cereal yields by region, 1990 191
2. Cereal yields in selected countries, 1990 192
3. Average maize yields in production test plots and farmer plots in Ghana, 1988 194
4. Average yields of maize and sorghum varieties in production test plots and farmer plots
in Upper Eastern Region, Ghana, 1988 195
5. Some phosphate rock deposits in the development countries 196
6. Impact of direct application of phosphate rock on crop yields in Mali and Columbia 197
7. N:P205:K20 Ratio, 1990 199
8. Fertilizer demand, 1989-90 to 2010 201
9. Growth in global fertilizer use 201
Chapter 8. Prospects of Meeting Future Food Needs Through New Technology
8.1 Maximum production in grain equivalent of six classes of agricultural land ranked
according to potential productivity 210
8.2 Theoretical maximum production of grain equivalents (per hectare) of the
continents and the world 210
8.3 Average world production per hectare for eight cereal crops, 1990-92 210
8.4 Rates of yield gain in wheat before and after yield takeoff in selected countries 211
8.5 Rates of yield gain in rice before and after yield takeoff in some countries 212
8.6 Rates of yield gain in maize before and after yield takeoff in selected countries 212
Chapter 10. Climate Change and Agriculture-Research Findings and Policy Considerations
10.1 Welfare effects by country group of agricultural impacts of climate change, based on
general circulation models 230
10.2 Price change from base resulting from climate change, based on general circulation models 230


Chapter 2. Global and Regional Population Projections to 2025
2.1 World population growth estimates (1750-1990) and projections (1990-2150) 8
2.2 The demographic transition in the developing countries 9
2.3 Determinants of population growth 10
2.4 Total fertility rate estimates (1950-90) and projections (1990-2025), by region 11
2.5 Life expectancy estimates (1950-90) and projections (1990-2025), by region 12
2.6 Population growth estimates (1950-90) and projections (1990-2025), by region 13
2.7 Population size estimates (1950-90) and projections (1990-2025), by region 14
Comment on Chapter 2: How Reliable Are the Population Projections?
1. Sub-Saharan Africa, population projections made in 1985 and 1994 19
2. Latin America and the Caribbean, population projections made in 1985 and 1994 19
3. East and Southeast Asia, population projections made in 1985 and 1994 19
4. Southwest Asia, population projections made in 1985 and 1994 19
5. Developed countries, population projections made in 1985 and 1994 19
6. South Asia, population projections made in 1985 and 1994 19
7. North Africa, population projections made in 1985 and 1994 20
8. Components of growth 20
9. Percentage distribution of the population (X axis) by five-year age group (Y axis) for
males and females, 1990, 2010, and 2025 21
Chapter 3. The Outlook for World Food and Agriculture to Year 2010
3.1 Growth rates of production (world, all products), moving eight-year periods, 1961-92 26
3.2 Growth rates of production (world, cereals), moving eight-year periods, 1961-92 26
3.3 Growth rates of production (all developing countries, all products), moving eight-year
periods, 1961-92 30
3.4 Growth rates of production (developing countries, low-dependence group, all products),
moving eight-year periods, 1961-92 30
3.5 Growth rates of production (developing countries, high-dependence group, all products),
moving eight-year periods, 1961-92 31
3.6 Growth rates of production (developing countries, high-dependence group, excluding China,
all products), moving eight-year periods, 1961-92 31
3.7 Chronic undernutrition in developing countries 33
Comment on Part II: Middle East/North Africa
1. Diet compositions of ruminant animals, by feed category, 1968-72 to 1988-92, for selected
countries of the Middle East/North Africa 136
2. Projected surpluses and deficits in livestock units or feeds, with population projections
for years 2000 and 2010, for 12 countries 137
Chapter 6. Future Supplies of Land and Water for World Agriculture
6.1 Hypothetical long-run global supply curves for agricultural land and water 144

Comment on Chapter 6: Future Supplies of Land and Water Are Fast Approaching Depletion
1. World grain area, 1950-93 161
2. Former Soviet Union grain harvested area, 1950-93 161
3. Grain production, Brazil, 1950-93 162
4. Grain harvested area for Japan, South Korea, and Taiwan, 1950-93 163
5. Grain production for Japan, South Korea, and Taiwan, 1950-93 163
6. World irrigated area per person, 1950-91 164
7. World irrigated area, 1950-91 164
8. World fertilizer use per person, 1950-93 166
Comment on Chapter 7: Growth Potential of Existing Technology Is Insufficiently Tapped
1. Effect of combined fertilizer and residue management on millet yield, Niger 195
2. Annual growth rate of fertilizer consumption in South America and South Asia, 1970-92 198
3. Trends in maize yield under different treatments on an acidic red loam (Paleustalf) at
Ranchi in eastern India 199
4. The effect of nitrogen fertilizer application on nitrate leaching and yield 200
5. Nitrogen fertilizer use per hectare of arable land and land under permanent crops, 1990 200
Chapter 8. Prospects of Meeting Future Food Needs Through New Technology
8.1 Yield categories and delineation of yield gaps 1 and 2 208
8.2 Three main research tasks that relate to improving crop yields 209
8.3 Wheat yields in China, 1949-92 213
8.4 Rice yields in China, 1949-92 213
8.5 A gradient of biotechnologies ranging from the more simple (near bottom) to the more
complex (top) 217


International Food Policy Research Institute (IFPRI), in
collaboration with the United Nations Food for Popula-
tion Activities and with support from the Rockefeller
Foundation, held a roundtable in February 1994, Popula-
tion and Food in the Early Twenty-First Century: Meeting
Future Food Demand of an Increasing Population. The
roundtable was held at a time that fell in between the Food
and Agriculture Organization of the United Nations and
the World Health Organization (FAO/WHO) Interna-
tional Conference on Nutrition and the United Nations
International Conference on Population and Develop-
ment. The IFPRI roundtable, with its focus on the pros-
pects of a long-term balance between food demand and
supply, provided a link between the two conferences, as
the adequacy of food supply at affordable prices for future
populations is a crucial element in a strategy designed to
alleviate poverty and accelerate growth, in the context of
an increasing population.
During the same period, comprehensive studies on the
assessment of the outlook for long-term food demand and
supply were under way in FAO, The World Bank, and
IFPRI. These projections not only dealt with supply and
demand prospects for the world as a whole, but also for
different groups of countries in both developed and devel-
oping regions. Particular attention was paid to Eastern
Europe, the former Soviet Union, and China in view of
significant policy and institutional changes under way in
these regions. The roundtable brought together the
authors of the three sets of long-term studies for a joint
examination of methodology, assumptions, and results of
their projections, with a view to arriving at a broad con-
sensus and identifying the areas of agreement and reasons
why results from the various projections might differ. In
addition, papers were presented on population projec-
tions, natural-resource constraints, and prospects of tech-
nological progress. The roundtable brought together

experts on the United States, European Union, Eastern
Europe, and the former Soviet Union, and on different
regions in the developing world with a view to obtaining
their views and judgment on both projections and analysis
of issues and their policy implications.
This volume contains the papers presented at the
roundtable as well as comments provided by the various
discussants, with an overview by the editor, Nurul Islam.
It is published as an input to the IFPRI conference, 2020
Vision for Food, Agriculture, and the Environment, and
is, in addition, considered as a contribution to the debate
that is likely to continue on the challenges, opportunities,
and constraints affecting the long-term food supply and
demand outlook in future years. Many unsettled questions
remain, especially on natural-resource and environmental
constraints that require further analysis and research. These
questions are discussed in this volume.
We acknowledge, with thanks, the financial contribu-
tion from the Rockefeller Foundation toward the organi-
zation of the roundtable. We especially are grateful to
S. W. Sindig of the Rockefeller Foundation for his per-
sonal and active interest. We also thank Nafis Sadik of the
United Nations Fund for Population Activities (UNFPA)
and M. Alauddin of the Commission for Population on
Welfare for active participation, and all the authors and
discussants, for the effort they have made in preparing the
papers and comments, and the time they have taken to
participate in the roundtable. We particularly appreciate
their efforts in revising the papers and in patiently re-
sponding to queries from the editor during the volume's
production. We hope that the publication of this volume
justifies their efforts.

Per Pinstrup-Andersen
Director General

List of Contributors

George I. Abalu is a senior regional adviser in Food and
Agricultural Policy and Planning at the United Nations
Economic Commission for Africa, Addis Ababa,
Mercedita Agcaoili is a postdoctoral fellow in the Envi-
ronment and Production Technology Division of the
International Food Policy Research Institute.
Nikos Alexandratos is chief of the Global Perspective
Studies Unit in the Economic and Social Policy De-
partment of the Food and Agriculture Organization of
the United Nations.
John Bongaarts is director of the Research Division of
The Population Council.
Eduard R. Bos is a demographer in the Population,
Health, and Nutrition Department of The World Bank.
Howarth E. Bouis is a research fellow in the Food
Consumption and Nutrition Division of the Interna-
tional Food Policy Research Institute.
Lester R. Brown is president of Worldwatch Institute.
Balu Bumb is a senior scientist in the Economics Re-
search and Development Division of the International
Fertilizer Development Center.
Pierre Crosson is a senior fellow and in the Energy and
Natural Resources Division of Resources for the Future.
S. Mahendra Dev is an associate professor at the Indira
Gandhi Institute of Development Research, Bombay,
Donald N. Duvick is an affiliate professor of plant breed-
ing in the Department of Agronomy at Iowa State
Behjat Hojjati is a research analyst in the Environment
and Production Technology Division of the Interna-
tional Food Policy Research Institute.
Merlinda D. Ingco is an economist in the Commodity
Policy and Analysis Unit of The World Bank's Inter-
national Economics Department.
Nurul Islam is a research fellow emeritus at the Interna-
tional Food Policy Research Institute.

D. Gale Johnson is a professor in the Department of
Economics at the University of Chicago.
William H. Meyers is a professor-in-charge of the Center
for Agricultural and Rural Development, Iowa State
Donald O. Mitchell is a senior economist in the Com-
modity Policy and Analysis Unit of The World Bank's
International Economics Department.
Thomas L. Nordblom is an agricultural economist of the
Pasture, Forage, and Livestock Program at the Inter-
national Center for Agricultural Research in the Dry
Areas, Aleppo, Syria.
Patrick O'Brien is director of the Commodity Econo-
mics Division of the U.S. Department of Agriculture's
Economic Research Service.
Peter A. Oram is a research fellow emeritus in the
Environment and Production Technology Division of
the International Food Policy Research Institute.
Kirit S. Parikh is an associate professor at the Indira
Gandhi Institute of Development Research, Bombay,
Donald L. Plucknett is a retired senior scientific advisor
for the Consultative Group on International Agricul-
tural Research at The World Bank.
Carlos Pomareda is director of Program I, in Agricul-
tural Policy Analysis and Planning, at the Inter-Ameri-
can Institute for Cooperation on Agriculture,
Coronado, Costa Rica.
John Reilly is deputy director for technology in the
Resources and Technology Division of the U.S. De-
partment of Agriculture's Economic Research Service.
Mark W. Rosegrant is a research fellow in the Environ-
ment and Production Technology Division of the In-
ternational Food Policy Research Institute.
Farouk Shomo is an economic research associate in the
Pasture, Forage, and Livestock Program at the Inter-
national Center for Agricultural Research in the Dry
Areas, Aleppo, Syria.

Chapter 1

By Nurul Islam

The experts, researchers, and policymakers as well as the
public in general seem to alternate between moods of
pessimism and optimism, anxiety and complacency,
about the world food situation and outlook. Frequently,
short-term food shortages or surpluses, or movements in
food prices, influence the perception of the future world
food situation. During the world food crisis of 1973,
marked by a sudden and substantial rise in food prices that
resulted in acute distress-and even starvation and famine
in several low-income developing countries-there was
great apprehension about the future food supply. The year
1973 was seen by many as a harbinger of things to come,
signaling the emergence of a long-run world food short-
age. During the late 1970s and 1980s, as the food situation
improved, food prices fell. Faced with an abundance, the
food-exporting developed countries resorted to measures
to curtail production or at least to reduce price support and
subsidy programs. Optimism about the future food situ-
ation was revived. This assessment was further reinforced
by the declining trend in the real prices of wheat and rice
over the past several decades.
As the 1980s rolled into the 1990s, concerns arose that
the increasing scarcity of land and water resources and
environmental degradation, including erosion, pollution,
and loss of biodiversity, might constrain the expansion of
food production in both developed and developing coun-
tries. Furthermore, climate change, including global
warming, might adversely affect the prospects of produc-
tion growth. The late 1980s had witnessed a fall in per
capital food production in developed countries and a slow-
down in the rate of growth of production and yield in
developing countries. It was argued that low cereal prices
throughout the 1980s were not so much a sign of abun-
dance as of inadequate demand due to a lack of income
and employment opportunities.
Optimists countered these concerns by asserting that
technological breakthroughs will continue to stimulate
production growth, that the impact of global warming is
highly uncertain, and that the dangers of resource scarcity
and environmental degradation are exaggerated and can
be mitigated by improved policies, institutions, and tech-
nological progress.

As IFPRI was planning its roundtable, "Population
and Food in the Early Twenty-First Century: Meeting
Future Food Demand of an Increasing Population," the
preparations were under way for the International Confer-
ence on Population and Development to be held in late
1994, with a focus on the consequences of increasing
population and measures to meet them. The anticipated
pressure of increasing population on future food supplies
contributed to a new awareness about the need to examine
the long-term demand and supply outlook for food. This
was an occasion to look beyond the short term to examine
whether the developments of the 1980s and early 1990s
were transient phenomena or had in them seeds of long-
term change.
The papers in this volume examine food demand and
supply prospects up to 2010 and explore the following
questions: What is the projected increase in world popu-
lation and in the different regions? What is the likely
demand for food, especially cereals? Will per capital food
production and consumption continue to increase? What
are the constraints, such as land and water, on increasing
food supply overall and in different regions? Will the
increased food supply be available at constant or rising
prices? What are the prospects of technological progress,
either intensification of known technology or break-
throughs in conventional plant breeding or biotechnol-
ogy? What are the implications of global warming for
future food supply?
At the roundtable, participants generally agreed that
the world food supply in 2010 would probably meet
global demand, but that regional problems could occur.
South Asia and Sub-Saharan Africa were recognized as
the most vulnerable regions. The key to future food sup-
plies was seen as increased productivity, that is, yields
must continue to rise; to accomplish this, sustained sup-
port for investment in agriculture, including research ex-
penditures, would be needed. The participants also
identified areas where knowledge is inadequate. For ex-
ample, uncertainty continues to surround the possibilities
of technological progress as well as the environmental
consequences of agricultural intensification. Sources of
growth in yield are inadequately understood, as are the

2 Population and Food in the Early Twenty-First Century

effects of changes in income distribution and urbanization
on food demand. The need for further analysis and re-
search was obvious.
In this context, an effort is made in this volume to take
a step forward in furthering analysis of and insight into
the evolving world food situation. The following provides
an overview of the major issues raised and conclusions
For the past several decades, the rate of growth in
world food production exceeded the population
growth rate. This was true not only for developed
and developing countries, as two separate groups,
but also for developing regions, except Sub-Saharan
Africa. Moreover, the trend over the last three dec-
ades shows an increase in average per capital food
consumption for all developing regions except Sub-
Saharan Africa.
The annual increase in world population has continu-
ally risen and is expected to peak during the decade
of the 1990s, remaining high for a couple of decades
before declining. Additions to the world population
will be around 933 million and 921 million, respec-
tively, during the two decades up to 2010. Nearly all
of this will occur in the developing world. The growth
rate is expected to decline from 2.1 percent a year
during 1985-90 to 1.5 percent during 2005-2010.
About 90 percent of the rate of increase in aggregate
food (cereal) demand from now to 2010 will be due
to population increase. During this period, per capital
growth in income is expected to be around 3.4 per-
cent. The result of a slowdown in the rate of popula-
tion growth combined with a modest increase in per
capital income will be a slow growth in per capital
cereal consumption. For the developing countries as
a group, per capital consumption is expected to in-
crease at an annual rate ranging from 2.2 to 2.4
percent during 1990-2010, that is, from 237 kilo-
grams a year during 1989-91 to around 250-255
kilograms in 2010.
The main driving force behind projected growth in
per capital cereal consumption is growth in the indi-
rect demand for cereals, that is, use of cereals as a
feed for livestock. Average per capital direct (human)
consumption of cereals for the world as a whole
registers a very small or no increase. However, indi-
rect per capital consumption of cereals is projected to
increase by about 80 percent, albeit from a low level
of 38 kilograms per capital to 57 kilograms.
The annual growth rate of food production in devel-
oping countries is expected to range from 1.8 to 2.1
percent against a rate of increase of 2.2-2.4 percent
in domestic consumption. This will lead to an in-
crease in imports from about 90 million tons during
1989-91 to more than double this volume by 2010.

A given change in growth rate of production or
demand causes a much larger change in growth of
net import volumes, since the latter constitutes a
relatively small proportion (about 10 percent) of
domestic production. By 2010, both the absolute
volume of net imports and the ratio of imports to
domestic production are projected to rise in develop-
ing countries.
* The projected changes in food supply, demand, and
prices are very sensitive to the assumptions made
about trends in population growth and agricultural
productivity. A small increase in population growth
rate above the projected rate or a small decline in the
yield growth rate below the projection raises the
price of cereals in world markets, often significantly,
and hence adversely affects the per capital consump-
tion in developing countries. For example, a 25
percent decline in the growth rate in yield per hectare
leads to a rise in the world price by 70 percent for
wheat, 40 percent for rice, 50 percent for maize, and
58 percent for other coarse grains; as a result, per
capital consumption falls by 4 percent for all devel-
oping countries. A 20 percent increase in population
growth rate causes increased demand and a rise in
world price of 30 percent for wheat, 18 percent for
rice, 11 percent for maize, and 18 percent for other
coarse grains.
* South Asia and Sub-Saharan Africa, with the highest
rates of increase in their net cereal imports, are the
two regions that face the greatest challenge in meet-
ing food demand. Net imports in Sub-Saharan Africa
could increase as much as 4 times and in South Asia
as much as 10 times. In spite of increased cereal
imports by 2010, per capital cereal consumption is
expected to rise only slightly in these two regions; in
Sub-Saharan Africa, it may even decline, according
to one projection. The two regions' current per capital
consumption is the lowest among developing coun-
tries-lower in Sub-Saharan Africa than in South
Asia--and will remain so even to 2010.
* The prospects of financing the rising cereal import
costs of developing countries depend on an increase
in their export earnings or "import capacity," and in
food aid. The export earnings are linked to prospects
of economic growth and trade liberalization in de-
veloped countries, their principal export markets.
The future of food aid depends on the availability of
food surpluses in food-exporting developed coun-
tries and on the latter's willingness-in the face of
competing demands for development assistance-to
provide food aid under the new rules of the Uruguay
Round regime.
* The growth in production and net exports of devel-
oped countries is expected to be adequate to meet the

Chapter 1: Overview 3

rising imports of developing countries. This will be
facilitated, first, by a very small increase in per capital
cereal consumption in developed countries, given
their already high level of direct and indirect con-
sumption, and, second, by growth in their produc-
tion, which, though lower than in the past, is
nonetheless expected to exceed consumption
* A major change in the net trade position of Eastern
Europe and the former Soviet Union is projected. By
2010, their per capital domestic absorption of cereals
will decline because of more efficient use of cereals
as animal feed and a reduction in the prevailing high
level of postharvest losses. Moreover, the recent
substantial decline in their consumption of cereals
and livestock, following the elimination of subsi-
dies, is expected to recover very slowly up to 2010
and only to the average level of 1989-91. As a
consequence, the formerly centrally planned econo-
mies will no longer be net importers and are likely
to generate net export surpluses that will partly offset
the increase in cereal imports of developing coun-
tries. At the same time, research and investment in
developed countries must continue in order to ensure
the necessary increase in their production and ex-
ports in a manner that meets their rising concerns
about the environmental effects of agricultural inten-
* The two most important factors contributing to fu-
ture increase in food supply in developing countries
are expansion of land under cultivation (including an
increase in irrigation) and increase in yield.
The amount of land currently under cultivation for
all agricultural crops is about 750 million hectares.
The land area in the same countries, not in agricul-
tural use but with rainfed crop production potential,
is more than twice this amount.
Not all this potential agricultural land will be avail-
able for crop production because of increased com-
petition for land for human settlements,
industrialization, and urbanization. The land with
potential for crop production that is available is
unevenly distributed among developing countries
and regions. Also, bringing land under cultivation
can entail high economic and environmental costs,
requires infrastructure development, and in many
cases, requires eradication or alleviation of livestock
and human diseases.
Water may become more of a constraint than land.
Area available for additional irrigation is estimated
to be 50 percent above the currently irrigated area,
and 80 percent of that will be in developing coun-
tries. Increasing competition for water in nonagricul-
tural uses, rising costs of irrigation projects, and the

likely adverse environmental effects of these proj-
ects all combine to limit the rate of expansion of
irrigation in developing countries.
* Because of constraints on the expansion of land and
water resources, growth in future food supply de-
pends predominantly on growth in yields. At the
same time, research and development efforts to ex-
pand the quantity and improve the quality of land and
water resources deserve high priority no less than
yield-enhancing technologies.
* Because of a multiplicity of factors, yield growth in
cereal production in developing countries has been
declining in recent years. While the rate of growth in
yield is on the decline, absolute increases in yield
over time have been on the whole positive. The
future rate of growth in yield, while generally of
lower magnitude than in the past, will not suffer a
continuous decline. All projections confirm that ab-
solute yields will increase over time.
* A recent disturbing development relates to the de-
cline in per capital world cereal production from the
mid-1980s onward. In the net cereal-exporting coun-
tries, which are mostly developed countries, there
has been a fall in both total and per capital cereal
production. This is most likely a reflection of a
demand constraint. Many developed-country ex-
porters resorted to supply management, but food aid
flows continued to rise-an unlikely occurrence in
the event of supply constraint. In the importing coun-
tries, the per capital growth rate in cereal production
has declined during this period but has not turned
negative. The reasons for this decline, which is not
necessarily a reflection of long-term trends, require
deeper analysis and watchful monitoring. In earlier
decades, there were instances of a pause or decline
in growth rate in per capital production (temporarily
during the 1960s and 1970s), but recovery soon
How to achieve a 1.5-2.0 percent annual rate of
growth in yield of major cereals in the next 20 years
or so constitutes a major challenge in developing
countries. This requires a wide diffusion and im-
proved efficiency of use of existing technology.
Adaptive and strategic as well as maintenance re-
search, with a greater emphasis on location-specific
research, will play a major role not only in maintain-
ing yields as new regions and farmers adopt new
techniques but also in improving potential yields. A
constant stream of plant varieties, even with similar
yield levels, needs to be generated for the breeding
of improved varieties. As food production expands
to "less favorable" or "low potential" areas with
poor soil or limited water availability, the research
needs of such areas assume increasing importance,

4 Population and Food in the Early Twenty-First Century

even though returns on research in less favorable
areas may be lower, slower to be realized, and more
uncertain. Future genetic improvement will rest on
conventional plant breeding, including hybrid seeds,
and on biotechnology. Prospects for biotechnology
to provide a significant breakthrough in yield in the
next 10-15 years are limited; its major near-term
contribution will be to provide greater resistance to
pests and diseases as well as enhanced stability by
reducing periodic decline in yields.
* The wide diffusion and efficient use of existing
technology and the generation of new technology
require a significant upgrading of education, train-
ing, and extension as well as rural infrastructure-a
difficult and time-consuming task. It is not certain
that required increases in food and agricultural pro-
duction can be secured without any environmental
costs; tradeoffs between environmental costs and
increased food production in specific regions cannot
be entirely ruled out. This emphasizes the critical
need for designing appropriate environmentally
friendly technology.
* In view of the challenges to expand and exploit the
world's food production potential, especially in de-
veloping countries, and a long gestation lag of 10-15
years between the initiation of research and the
achievement of its results, the priority given to agri-
cultural research and investment needs to be raised,
not lowered. The low cereal prices currently prevail-
ing in the world market have generated a sense of
complacency leading to a stagnation or even a de-
cline in public-sector agricultural investment, in-
cluding research expenditures, both national and

Food security in developing countries cannot be
viewed with reference solely to cereals or even to all
food crops. Nor does it involve the pursuit of national
food self-sufficiency, irrespective of comparable
cost considerations. The agriculture sector as a
whole provides income and employment and hence
entitlements or access to food, including access to
imports financed by agricultural export earnings.
The projected food demand and supply balance re-
fers to effective "market" demand, not to nutrition
or food needs based on energy and protein require-
ments. The projections imply a doubling of per cap-
ita income by 2010 and, provided the distribution of
income remains unchanged, a doubling of the per
capital income of the poor. In fact, in poor countries,
no significant increase in per capital consumption is
expected. Meeting nutritional needs would require
not only adequate food supply but also a significant
increase in effective demand or entitlements for the
Chapters 2-10 in this volume are based on papers
presented by the chapters' authors at the IFPRI round-
table. These chapters are organized into three parts: Part
I, Global and Regional Population Projections (Chapter
2); Part II, Global and Regional Food Demand and Supply
Prospects (Chapters 3-5); and Part III, Selected Issues
Affecting Future Food Supply (Chapters 6-10). Chapters
2, 6, and 7 are each followed by a comment by another
expert on the issue addressed in the particular chapter.
Each of the six comments in Part II address all three
chapters in this section. An appendix to Part II that com-
pares the alternative demand and supply projections
offered by the authors of Chapters 3-5 is also included.

Global and Regional Population Projections

Chapter 2

By John Bongaarts

The unprecedented speed with which the world's popula-
tion has grown over the last four decades has resulted in
a more than doubling of the number of inhabitants, bring-
ing the total to 5.3 billion by 1990. Despite substantial and
partially successful efforts to reduce population growth in
the developing countries, the climax of this expansion will
come over the next three decades. Long-range projections
suggest that the world population will continue to grow
throughout the next century, eventually reaching totals of
11 to 12 billion or higher.
There is little doubt that this population "explosion"
represents one of the most important processes affecting
adversely the current and future welfare of humans and
threatening the natural environment. Sheer numbers of
people put pressure on limited environmental resources,
thus leading to resource depletion and pollution as well as
difficulties in raising standards of living, especially in
poor countries. Population densities have reached very
high levels in some countries (for example, 800 people
per square kilometer in Bangladesh), and densities will
more than double over the next century in most countries
in Africa, Asia, and Latin America. However, even in
countries where population densities are still relatively
low-such as in substantial parts of Sub-Saharan Africa
and Latin America-rapid population growth itself can
pose an important obstacle to development. Expansion of
the stock and quality of human and physical capital is the
critical ingredient of development. However, in countries
with high population growth rates, large investments are
needed just to prevent the stock of capital per capital from
deteriorating. Desirable growth in standards of living
requires that such investments grow more rapidly than
population itself-a task that many poor countries find
very difficult to achieve. This adverse consequence of
rapid population growth operates independently from the
effects of high density, although there are of course many
countries in which these two factors are both present.
After a brief review of long-range trends in population,
this chapter focuses on the most recent medium-range
projections (to 2025) made by the Population Division of
the United Nations (United Nations 1993a). To put these

projections into perspective, trends in key demographic
variables since 1950 are also examined. Although projec-
tions are available for all individual countries, the present
analysis only occasionally refers to them and instead
focuses on regional aggregates. The following six regions
are included: Asia (excluding West Asia, Japan, Australia,
and New Zealand), Latin America (including the Carib-
bean), Sub-Saharan Africa, and West Asia/North Africa
in the developing world; and the Organization of Eco-
nomic Cooperation and Development (OECD) countries
and Commonwealth of Independent States (CIS)/Eastem
Europe in the developed world.1 The chapter concludes
with a brief assessment of the demographic impact of the
Acquired Immune Deficiency Syndrome (AIDS) epi-
demic in Africa.

Long-Range Trends
For most of human history, population growth remained
near zero, when averaged over long periods. Although
historical records are incomplete and far from reliable, the
modem expansion of human numbers began, very slowly,
in the eighteenth century with the decline of the death rate
in Europe and North America. As shown in Table 2.1, by
the year 1900 the population of the world numbered
approximately 1.6 billion, of which a third lived in the
regions now referred to as developed countries. Growth
accelerated in the twentieth century and has been espe-
cially rapid in the developing countries since 1950 (Table
2.1). More people have been added to the world since
1950 than in all of human history before the middle of this
century (Merrick 1989).

Table 2.1-Population growth, 1900-90
Population Increase
1900 1950 1990 1950-90
(billions) (percent)
Developing countries 1.07 1.68 4.08 143
Developed countries 0.56 0.84 1.21 44
World 1.63 2.52 5.30 110
Source: Merrick 1989.

8 Population and Food in the Early Twenty-First Century

The acceleration in growth is well demonstrated by the
shortening of the time intervals needed to add successive
billions to the world population. The first billion was
reached early in the nineteenth century, the second billion
took 120 years, the third 33 years, the fourth 14 years, and
the fifth (between 1974 and 1987) just 13 years. Ifprojec-
tions turn out to be accurate, the next 3 billion will be
added at an even faster pace, each taking just over a
decade, to reach 8 billion before the year 2020.
Long-range population projections suggest continued
growth for the century ahead (United Nations 1993a; Bos
et al. 1992). Between 1990 and 2100, as shown in Table
2.2, the population of developed countries is expected to
rise by 0.3 billion, reaching a total of 1.5 billion (Bos et
al. 1992).

Table 2.2--Population growth, 1990-2100
Population Increase
1990 2025 2100 1990-2100
(billions) (percent)
Developing countries 4.08 7.07 10.20 150
Developed countries 1.21 1.40 1.50 24
World 5.30 8.47 11.70 121
Sources: Bos et al. 1992; United Nations 1993a.

Growth in the developing countries is projected to be
much more rapid, adding 6.1 billion by the end of the next
century. The global total may be 11.7 billion in 2100.
Figure 2.1 presents the additions per decade to the
population of the world and to the developing countries
and developed countries for the past two centuries as well
as for the next century. The most striking feature of this

Figure 2.1-World population growth estimates
(1750-1990) and projections (1990-
1000 190o
1750 1800 1850 1900 1950 2000 2050 2100 2150
mDeveloped Countries E Developing Countries
Sources: Data for 1750-1900, United Nations 1973; data for
1920-40, United Nations 1966; data for 1950-2025,
United Nations 1993a; data for 2030-2150, Bos et al.
1992; Merrick, 1989.

figure is the peak in population increments that will occur
between 1990 and 2020. Additions to population during
these three successive decades are expected to be 933,
921, and 900 million, respectively. Nearly all of this
growth is projected to take place in the developing coun-
tries. There are two reasons for this: annual population
growth rates are more than four times higher in developing
countries than in developed countries, and the population
size of the developing countries is much larger than that
of the developed countries. The net result is that only 1 in
17 of the additions to world population between 1990 and
2025 will be in the developed countries.
It is important to note that the long-range trends in
absolute growth plotted in Figure 2.1 are not the same as
those for the rate of growth. The most important differ-
ence is that the world's growth rate peaked in the late
1960s. Between 1970 and the present, growth rates have
been declining at the same time that annual population
additions have continued to rise. These facts, perhaps
puzzling at first, are consistent with one another because
the slowly declining growth rates are applied to rapidly
expanding population totals. It is only after 2020 that
declines in the growth rate are sufficient to bring about a
significant reduction in absolute annual population

The Demographic Transition
The world population increases because each year more
births occur than deaths. In the early 1990s, the absolute
increase in population was 93 million per year, the differ-
ence between 144 million births and 51 million deaths.
Similarly, the world's growth rate (1.7 percent per year)
equaled the birth rate (26 births per 1,000 people per year)
minus the death rate (9 deaths per 1,000 people per year).
At the country and regional levels, population growth can
also be affected by migration, but for large aggregates of
population, such as those used in this analysis, migration
is a sufficiently minor factor that it may be ignored for
present purposes.
Demographers use the term "demographic transition"
to refer to the broad pattern of secular change in birth and
death rates that accompanies a population's development
from a traditional agricultural base into a modern indus-
trial society. Population size is nearly constant before and
after this transition, but in the intervening period (which
can extend to well over a century) rapid growth occurs.
Formally, five stages in the demographic transition can be
identified (Bongaarts 1994a):
Pretransition. Birth and death rates are both high and
population growth is near zero. This situation pre-
vailed in most of the world before the onset of the
industrial age.
Early transition. The death rate declines as a result
of a lower incidence of epidemics and famines, and
of improvements in standards of living, nutrition,

Chapter 2: Global and Regional Population Projections to 2025

and public health measures. Since fertility remains
high, population growth accelerates.
Midtransition. As the economic and social value of
children declines, couples begin to adopt contracep-
tion to limit family size, thus initiating a reduction
in the birth rate. The death rate continues to decline
and population growth reaches its maximum.
Late transition. The death rate reaches a minimum,
and with continued declines in fertility the rate of
population growth, while still positive, slows.
Posttransition. A new equilibrium between births
and deaths is established, resulting in near zero
growth in population size.
This broad description of the demographic transition
is consistent with a large variety of actual patterns of
change over time. The timing of the onset, the duration of
the different phases, and levels of birth and death rates
differ from region to region and from country to country.
In the developed countries, the transition started in the
second half of the eighteenth century with a sustained
mortality decline. The onset of the fertility decline
occurred in the late nineteenth and early twentieth centu-
ries. The reduction in the birth rate since then has not been
continuous, as an unusually low fertility rate during the
depression of the 1930s was followed by the postwar
"baby boom." The developed countries have now
reached the last two transition stages. Today the average
rate of population growth of more developed countries is
only 0.5 percent per year, and in a few countries-such as
Germany, Sweden, and Hungary-population growth de-
clined slightly in the late 1980s.
The demographic transition in Africa, Asia, and Latin
America started much later than in the industrialized
countries (Figure 2.2). A sustained reduction in mortality
was not observed until the end of the nineteenth century
and the first half of the twentieth century. Low-cost public

Figure 2.2-The demographic transition in the
developing countries
Births/Deaths per 1,000
I Birth Rate

1900 1950

Sources: Pre-1950 data, author's calculations; 1950 onward,
United Nations 1993a.

health measures have yielded very rapid mortality reduc-
tions, especially in the decades since World War II. By
the late 1960s, the average death rate had dropped to 15
per 1,000 people per year, which, together with a still
largely unchanged birth rate of 40 per 1,000 people per
year, yielded an annual growth rate of 25 per 1,000 people
per year or 2.5 percent. This rate of natural increase is
substantially above those observed historically in Euro-
pean populations in midtransition; a difference that is in
part attributable to relatively higher levels of fertility in
contemporary developing countries. Since the late 1960s,
birth rates have declined rapidly, except in Sub-Saharan
Africa. The steepest reductions occurred in East and
Southeast Asia, where socioeconomic development and
family planning programs have been conducive to rapid
changes in reproductive behavior.
As a result of variations in the timing and speed of the
declines in mortality and fertility, countries in the devel-
oping world today are at widely differing stages of the
transition. Much of Sub-Saharan Africa has yet to experi-
ence a significant fertility decline and is hence still in the
early transitional stage. In contrast, the most advanced
countries, such as Hong Kong and Singapore, have nearly
completed the transition. Most developing countries are
at present in the mid- and late transitional stages.

Population Projection Procedures
The most general assumption underlying population pro-
jections made by the United Nations and the World Bank
is that populations of individual countries will over time
proceed through the successive transition stages, eventu-
ally completing the demographic transition (United Na-
tions 1993a; Bos et al. 1992). In principle, a variety of
procedures could be used to calculate the future popula-
tion totals implied by this assumption. One could, for
example, project a reduction in the population growth rate
from its current level to zero at a specified time in the
future, or one could make specific assumptions about
future trajectories of birth and death rates. In reality,
recent projections almost invariably rely on much more
complex methodologies. A full description of them is
beyond the scope of this chapter, but it is helpful to
comment briefly on the most critical concepts and variables.
The chain of causation linking population growth to
successive layers of determinants is summarized in Figure
2.3. As is clear from the preceding discussion, the first
step in analyzing this set of relationships is to recognize
that the growth rate equals the birth minus the death rate
(plus the net migration rate, not shown). These so-called
"vital" rates of population are, in turn, determined by
individual-level measures of fertility and mortality. The
most widely used fertility indicator is the total fertility
rate, which equals the number of births a woman would
have by the end of her reproductive years if she experi-
enced age-specific fertility rates of a given year. Mortality

10 Population and Food in the Early Twenty-First Century

Figure 2.3-Determinants of population growth
Population Growth Rate

Birth D
Rate R;

I Age I


Fertility Expectancy

Family Public
Planning Socioeconomic Health
Programs Development Programs

is usually measured by the life expectancy at birth, which
equals the average number of years a newborn would live
if subjected to the age-specific mortality rates prevailing
in a given year. Unfortunately, there is no simple direct
relationship between the total fertility rate and the birth
rate, or between life expectancy and the death rate. The
reason is that birth and death rates are also influenced by
the age composition of the population, which in turn is
determined by past levels of fertility and mortality. As a
consequence, two populations can have the same death (or
birth) rates, while at the same time having quite different
levels of mortality (or fertility). This is the explanation for
the surprising fact that the death rate in the developing
countries (9.1 per 1,000 people per year) is slightly lower
than in the developed countries (9.7 per 1,000 people per
year) in 1990-95. Life expectancy in the developed coun-
tries is, as expected, higher than in the developing coun-
tries (75 versus 62 years); and, other things being equal,
this should have led to a substantial difference in the death
rate. However, the much older age structure of the devel-
oped countries puts enough upward pressure on the death
rate to make it slightly higher than the rate in developing
countries. An examination of trends and differences in
mortality and fertility should therefore rely on measures
such as the total fertility rate and life expectancy, rather
than on the cruder and potentially misleading birth and
death rates. (The adjective "crude" is systematically used
in the demographic literature when referring to these rates.)
The age structure of a population clearly plays an
important role in determining birth, death, and population
growth rates. Its evolution over time should therefore be
tracked to make accurate projections. Complex calcula-
tions are required to achieve this objective. Fortunately,
standard software is available for personal computers to
handle this menial task, thus greatly simplifying the con-
struction of population projections. The user still has to
provide a variety of input variables, the most crucial of
which are the expected future trajectories of fertility and

The final layer of determinants of growth is a set of
socioeconomic and programmatic factors (Figure 2.3).
Changes in social and economic variables (such as levels
of education, urbanization, and income) are the basic
driving forces of the demographic transition and, in the-
ory, the transition can be completed without active gov-
ernment efforts to influence fertility and mortality. In
practice, all governments are deliberately trying to im-
prove health conditions. The implementation of public
health programs has provided access to immunization,
sanitation, and other health services that have brought
about unprecedented increases in life expectancy. Most
developing-country governments now also try to acceler-
ate fertility declines by implementing family planning
programs. Nearly all of these programs are voluntary and
provide contraceptive information, supplies, and services
to assist individuals in implementing their reproductive
Unfortunately, current understanding of the links be-
tween development and family planning and public health
programs on the one hand, and development and fertility
and mortality on the other, is limited. It is therefore not
possible to quantify these links with sufficient degrees of
precision to incorporate them routinely in population pro-
jections. However, analysts may well take knowledge of
a country's conditions into account when specifying
trends in fertility and mortality for projections, and the
Population Division of the United Nations has in fact done
this in preparing the projections summarized below.

Fertility and Mortality Trends
and Assumptions
The most critical assumptions incorporated in projections
are those related to the future trajectories of fertility and
mortality. For the near future, extrapolation of past trends
can suffice, but additional assumptions are required to
produce satisfactory medium- and long-range projections.
Past and projected future levels of fertility for the
period from 1950 to 2025 are presented in Figure 2.4. The
total fertility was high and virtually stable at above six
births per woman on average in the 1950s (the slight
temporary decline in the late 1950s in Asia was due to a
large famine in China). This level of fertility reflects a near
absence of birth control, a condition that prevailed for
centuries before the middle of this century. Around 1970,
a decline in fertility started nearly simultaneously in Asia,
Latin America, and West Asia/North Africa. The rate of
decline since the 1970s has been most rapid in Asia, in
large part because of an exceptionally sharp reduction in
China. Sub-Saharan Africa is the only region that has yet
to experience significant reproductive change (there are
however a few countries, such as Botswana, Kenya, Zim-
babwe, and South Africa, where a fertility decline has
been initiated). As a result of these divergent past trends,
fertility levels in 1990-95 differed widely among regions

Chapter 2: Global and Regional Population Projections to 2025

Figure 2.4-Total fertility rate estimates (1950-90) and projections (1990-2025), by region
Births per Woman




- Sub-Saharan Africa

Latin America
- OECD Countries

- -West Asia/North Africa
SCIS/Eastern Europe

Sources: United Nations 1993a; Population Council data bank.

from a high of 6.4 births per woman (bpw) in Sub-Saharan
Africa to 4.7, 3.2, and 3.1 bpw in West Asia/North Africa,
Asia, and Latin America, respectively. The average for
developing countries in 1990-95 stands at 3.6 bpw. In
contrast, the average for developed countries was already
low (2.8 bpw) in the early 1950s and has since declined
to 1.9 bpw in the early 1990s. In the 1950s and 1960s,
fertility levels in the OECD countries differed little from
those in CIS/Eastern Europe, but starting in the 1970s the
fertility in the former has averaged a few tenths of a bpw
below that of the latter.
Future trends in fertility are based on the assumption
that the total fertility rate will over time reach and then
remain at the so-called "replacement" level in all regions,
except in the OECD countries. Replacement fertility is
just above 2 bpw and represents the level at which each
generation just replaces the previous one, thus leading to
zero population growth. Below-replacement fertility pro-
duces, in the long run, negative population growth. As is
evident from Figure 2.4, the total fertility rates in Asia,
Latin America, and CIS/Eastern Europe are expected to
reach the replacement level before 2025. Sub-Saharan
Africa and West Asia/North Africa are also on a trajectory
toward replacement fertility, but this level is not expected
to be reached until some time after 2025. The OECD
countries now have below-replacement fertility and this
is expected to remain so until 2025. Sub-Saharan Africa
is the only region in which the future trend in fertility is
sharply different from the past, and for this reason the
projection assumption must be considered the most specu-

Mortality levels have also changed rapidly over the
past few decades (see Figure 2.5). The developing coun-
tries have experienced exceptional improvements in life-
expectancy; from an average of 41 years in 1950-55 to 62
years today. By the early 1990s, Latin America had
reached mortality levels similar to those prevailing in the
developed countries in the 1950s, and Asia and West
Asia/North Africa are not far behind. Sub-Saharan Africa
has had the lowest life-expectancy levels and the slowest
rate of improvement. As a result, its life-expectancy, at 51
years in 1990-95, is still substantially below that of the
other developing-country regions. As expected, the
OECD countries and CIS/Eastern Europe already had
achieved relatively low levels of mortality by 1950 but
have nevertheless seen significant further improvements
since then. Life expectancy in the OECD countries now
stands at 77 years, while a stagnation during the 1970s and
1980s has prevented CIS/Eastern Europe from achieving
similar improvements.
Projections of future life expectancies by the United
Nations assume continued improvements over time in all
regions. However, increments will be increasingly diffi-
cult to achieve as countries reach ever higher levels of life
expectancy. The maximum, to be reached in the twenty-
first century, is assumed to be 85 years. By 2025, mortality
conditions in Asia, Latin America, and West Asia/North
Africa are expected to be similar to those that prevailed in
the OECD countries in the 1970s. Sub-Saharan Africa will
continue to lag, in part because the continent is most
heavily affected by the AIDS epidemic. This subject will
be discussed further in a later section.





12 Population and Food in the Early Twenty-First Century

Figure 2.5--Life expectancy estimates (1950-90) and projections (1990-2025), by region
Life Expectancy (years)


1975 2000

- Sub-Saharan Africa

- Latin America
-OECD Countries

- West Asia/North Africa
CIS/Eastem Europe

Sources: United Nations 1993a; Population Council data bank.

Projection Results
The implications of assumed trends in fertility and mor-
tality for future population growth and size are summa-
rized in Figures 2.6 and 2.7. Population growth rates are
expected to decline modestly in all regions between 1990
and 2025, with the most rapid declines in Asia, Latin
America, and West Asia/North Africa (Figure 2.6A). The
current wide disparities between the growth rates of dif-
ferent regions are projected to continue until 2025, with
the highest growth rates in Sub-Saharan Africa (now more
than 3 percent per year) and the lowest in the developed
countries. The absolute annual increases resulting from
these growth rates are plotted in Figure 2.6B. Despite
reductions in growth rates, absolute population increases
are expected to rise in Sub-Saharan Africa and West
Asia/North Africa and to change little in Latin America.
Future additions will be far larger in Asia than in other
regions, although Sub-Saharan Africa will not be far
behind by 2025.
Projections of population size until 2025 are plotted in
Figure 2.7. The population of the world is expected to
grow from 5.3 billion in 1990 to 8.5 billion in 2025.
Nearly all of this increase will occur in the developing
countries, which are expected to grow by 3 billion (from
4.1 to 7.1) over this period. The projection for the devel-
oped countries indicates an increase from 1.2 to 1.4 billion
between 1990 and 2025, thus adding just 0.2 billion. Asia,
with over half of the world's inhabitants in 1990, is by far
the largest region, and this will remain so until 2025, when

its population is expected to reach 4.5 billion. Projections
for the different regions are shown in Table 2.3.

Table 2.3-Projected population growth, 1990-
2025, based on United Nations
medium variant
Population Increase
1990 2025 1990-2025
(billions) (percent)
Asia 2.87 4.49 56
Latin America 0.44 0.70 59
West Asia/North Africa 0.27 0.56 107
Sub-Saharan Africa 0.50 1.30 160
OECD countries 0.80 0.91 14
CIS/Eastern Europe 0.40 0.48 20
World 5.29 8.47 60
Source: United Nations 1993a.
One important consequence of the wide regional vari-
ation in the growth rates (evident from the data in the last
column of Table 2.3) is that the regional composition of
the world will shift. The shares that Asia (54 percent) and
Latin America (8.3 percent) have of the world population
will change little, but Sub-Saharan Africa's share could
rise from 9.4 percent in 1990 to 15.3 percent in 2025, and
the developed countries could lose an equivalent amount,
declining from 22.6 to 16.4 percent over the same time
These projection results are all taken from the medium
variant of the United Nations, which represents the most

Chapter 2: Global and Regional Population Projections to 2025

Figure 2.6-Population growth estimates (1950-90) and projections (1990-2025), by region
Percent per Year





- Sub-Saharan Africa

- -Latin America
- OECD Countries

- -West Asia/North Africa
"CIS/Eastern Europe

Millions per Year

1975 2000

Asia Latin America
Sub-Saharan Africa OECD Countries
Sources: United Nations 1993a; Population Council data bank.

likely scenario. The United Nations also makes alternative
"low" and "high" projections, These other variants pro-
vide a range within which the actual population trajectory
is most likely to fall (no probabilities have been esti-
mated). The only differences between these variants relate
to the future trend in fertility. Compared with the medium
projection, the low variant's fertility is assumed to decline
more rapidly and to a lower (below replacement) level. In
the high variant, the fertility decline is slower and remains

- -West Asia/North Africa
-CIS/Eastern Europe

above replacement. The resulting population projections
for 2025 are shown in Table 2.4.
The high and low variants differ only modestly-less
than 10 percent-from the medium projection for 2025.
The past record of United Nations projections is quite
good (Lee 1990) and, barring unforeseen extraordinary
disasters, these variants can therefore be used with con-
siderable confidence as upper and lower limits for future
population trajectories. The relatively small differences





14 Population and Food in the Early Twenty-First Century

Figure 2.7-Population size estimates (1950-90) and projections (1990-2025), by region

0 "

1975 2000


-Developing Country

l Latin America
FOECD Countries

mWest Asia/North Africa BSub-Saharan Africa
mCIS/Eastern Europe -World Total

Sources: United Nations 1993a; Population Council data bank.

between the variants suggest that the medium projections
are quite robust and, for most planning purposes, these
projections are likely to be more accurate than forecasts
of other economic or social variables.

Table 2.4-Projected population growth for
2025, based on United Nations low
and high variants
Population Deviation from
Projection 2025 Medium Variant
Low High Low High
(billions) (percent)
Asia 4.15 4.81 -7.7 7.1
Latin America 0.66 0.76 -5.6 9.2
West Asia/North Africa 0.53 0.60 -6.1 7.2
Sub-Saharan Africa 1.21 1.38 -6.7 6.2
OECD 0.85 0.98 -7.1 7.2
CIS/Eastern Europe 0.43 0.52 -9.9 7.0
World 7.85 9.08 -7.3 7.2

Source: United Nations 1993a.

The discussion thus far has focused on population
aggregates and has not dealt with issues of population
distribution within countries or regions. One distribu-
tional issue, the proportion of population living in urban
areas, deserves a brief comment because of its policy
significance. In all populations, urban areas are growing
more rapidly than rural areas. Urban populations can grow
for three reasons: an excess of births over deaths, immi-
gration from rural areas, and reclassification of areas from

rural to urban. The urban proportion has risen rapidly
since 1950 and, as shown in Table 2.5, this trend is
expected to continue until 2025.

Table 2.5-Urban population growth, 1950-2025
Urban Population
1950 1990 2025
Developing countries 17 34 57
Developed countries 54 73 84
World 29 43 61
Source: United Nations 1993a.

This urbanization trend, together with overall popula-
tion growth, has led to explosive growth of urban areas
and especially of metropolitan areas. On the other hand,
rural growth has been slower than for the population as a
whole, and in many developing countries rural population
growth is expected to virtually cease early in the next
century (United Nations 1993b). This will have signifi-
cant implications for trends in land-use patterns.

The Demographic Impact of
the AIDS Epidemic
Previous projections by the United Nations have been
criticized for ignoring the AIDS epidemic. To address this
shortcoming, the latest projections, made in 1992, have
been revised to take explicitly into account the demo-
graphic effects of the epidemic (United Nations 1993a).
This adjustment has resulted in substantial changes in the

Chapter 2: Global and Regional Population Projections to 2025

projections for some countries, and a brief summary of a
few key results are presented next.
Since the epidemic's demographic effect is expected
to be very small in most countries, the United Nations'
adjustments have been limited to countries in which the
prevalence of HIV (the AIDS virus) exceeded 1 percent
in 1990. By this criterion, 15 African countries were
selected as having the most severe epidemics. Most of
these countries are located in East and Central Africa, and
together they account for 90 percent of infected individu-
als in Sub-Saharan Africa. For each of these countries, two
separate projections were made: one without and the other
with the epidemic. To obtain the latter, the United Na-
tions' demographic projection model was combined with
an epidemiological model developed by the World Health
Organization to forecast AIDS mortality at the country
level (Chin and Lwanga 1991; Heuveline et al. 1992).
Projections were made until 2005, when the epidemic is
expected to reach its peak impact. The key results for this
group of 15 countries are shown in Table 2.6.

Table 2.6-Effect of AIDS on projected popula-
tion growth, 1985-2005
1985-90 2000-05
With Without With Without

Life expectancy (years) 50.20 51.40 51.20 57.70
Death rate
(per 1,000 people) 15.80 15.20 13.70 10.80
Growth rate (percent) 3.21 3.27 2.88 3.13
Source: United Nations 1993a.
Although the epidemic had a limited effect in the 1980s,
its impact will grow with the size of the epidemic, and after
the turn of the century, mortality measures will be affected
substantially. Comparing projections with and without
AIDS, the average life expectancy is reduced by more than
6 years (57.7 minus 51.2) and the death rate is higher by 2.9
deaths per 1,000 people (13.7 minus 10.8). The population
growth rate is lowered significantly from 3.13 to 2.88
percent per year, but growth is clearly expected to remain
high, as the birth rate is largely unaffected.
These results are averages that conceal substantial vari-
ations among countries. An examination of country-spe-
cific projections indicates that Zambia is expected to be the
most seriously affected by the epidemic. In 2000-05, Zam-
bia's life expectancy is projected to be 15 years lower than
it would have been in the absence of AIDS, and the death
rate is expected to be 77 percent higher (that is, 16.6 rather
than 9.4 per 1,000 people). However, even in this extreme
case, population growth rate would still be 2.6 percent per
year rather than the 3.2 percent that would have been
observed without the epidemic. If these country-specific
projections are accurate, then in Sub-Saharan Africa as a
whole the population growth rate would be reduced by

about 0.1 percent in 2000-05, thus having only a marginal
impact on population projections for the continent.
It is important to note that these results hinge on two
crucial assumptions: that epidemics in countries outside
this group of 15 African countries will remain relatively
small, and that the African epidemics reach a maximum
in 2000-05. These assumptions are not implausible, based
on our current understanding of the epidemic, but if they
turn out to be incorrect, its demographic impact could
become substantially larger. After 2005, the rates of HIV
infection are assumed to decline quickly. This may well
be an unrealistic expectation, thus leading to some over-
estimation of the population size after 2005, especially in
Sub-Saharan Africa.

In the future, the actual population trajectories of regions
will almost certainly deviate to some extent from those
now projected, because current forecasts rely on poten-
tially inaccurate assumptions. Deviations could in a few
instances exceed 10 percent by 2025, but for the world as
a whole the error is likely to be smaller. There are two
reasons that substantial population growth in the develop-
ing countries is all but assured. The first is that fertility is
still well above the replacement level in all major devel-
oping-country regions. Although declines are now under
way, replacement fertility is not expected to be reached
until well after the year 2000. Even if fertility drops
eventually below two births per woman in some regions,
the average fertility for the entire period 1990 to 2025 is
likely to make a positive contribution to population
growth. The second reason is the existence of population
momentum, which is the tendency of developing-country
populations to continue growing for decades even if re-
placement fertility could be achieved immediately. There
is nothing that in practice can be done about its underlying
cause, a young age structure, although it can be slowed
somewhat by delaying the average age of childbearing. In
sum, for all practical purposes, it is wise to assume that
the population expansion over the next few decades will
not be very different from the medium variant of the
current United Nations projections (or from quite similar
World Bank projections).
The addition of around 3 billion new inhabitants of the
globe by the year 2025 will no doubt put enormous
pressure on world agriculture to provide improved diets
for this expanding population. A simple numerical illus-
tration documents the magnitude of the task ahead and the
role of population growth in the developing countries. Per
capital cereal production has risen at an average rate of 1
percent per year over the past three decades in developing
countries (FAO 1991). Continuation at this rate of im-
provement until 2025 would require that cereal produc-
tion expand by 131 percent between 1990 and 2025.
Without population growth, this increase need only be 42

16 Population and Food in the Early Twenty-First Century

percent. In other words, population growth could be re-
sponsible for well over half of the growth in cereal output
that is likely to be needed by 2025.
The many adverse consequences of a continued expan-
sion of human numbers make further efforts to lower
fertility, and hence population growth, highly desirable.
If a comprehensive approach is adopted and implemented
with vigor (this should include the strengthening of family
planning programs as well as efforts to reduce high de-
mand for children and population momentum), significant
additional reduction in fertility can be achieved (Bon-
gaarts 1994b). However, even in that case, it is unlikely
that population totals will be substantially below the
United Nations low variant.

1. "West Asia/North Africa" refers in this chapter to the
following 17 countries: Algeria, Cyprus, Egypt, Iraq,
Jordan, Kuwait, Lebanon, Libya, Morocco, Oman,
Saudi Arabia, Sudan, Syria, Tunisia, Turkey, United
Arab Emirates, and Yemen.

Bongaarts, J. 1994a. Demographic transition. In Encyclo-
pedia of the environment, ed. R. A. Eblen and W. R.
Eblen, 132. Boston: Houghton-Mifflin.
1994b. Population policy options in the devel-
oping world. Science 263 (5148): 771-776.

Bos, E., M. T. Vu, A. Levin, and R. Bulatao. 1992. World
population projections, 1992-93 edition. Baltimore:
Johns Hopkins University Press for the World Bank.
Chin, J., and S. Lwanga. 1991. Estimation and projection
of adult AIDS cases: A simple epidemiological model.
Bulletin of the World Health Organization 69 (4):
FAO (Food and Agriculture Organization). 1991. Agro-
stat-PC databank. Rome: FAO.
Heuveline, P., L. Heligman, J. Chin, and H. Strickler.
1992. The projected impact of AIDS on population and
demographic variables. Paper presented at the Annual
Meeting of the Population Association of America,
Denver, Colorado, 30 April to 2 May.
Lee, R. 1990. Long-range global population forecasts: A
critical appraisal. In Resources, environment, and
population, ed. K. Davis and M. Bernstein.
Merrick, T. 1989. World population in transition. Popu-
lation Bulletin 41 (2).
United Nations. 1966. World population prospects as
assessed in 1963. Population Studies 41. New York.
1973. The determinants and consequences of
population trends, vol. 1. Population Studies 50. New
1993a. Worldpopulationprospects: The 1992
revision. New York.
1993b. World urbanization prospects: The
1992 revision. New York.

Comment on Chapter 2

By Eduard R. Bos

Demographer Philip Hauser once said about population
projections, "Predictions, estimates, projections, fore-
casts; the fine academic distinction among these terms is
lost upon the user of demographic statistics. So long as
numbers that purport to be possible future populations are
published, they will be regarded as forecasts or predic-
tions" (Haub 1987). While there is considerable truth in
this from the user's perspective, there is, however, one
characteristic that from the producer's position makes a
demographic projection better than a prediction. That is,
population projections of the kind the United Nations, the
World Bank, and others make are replicable calculations.
In other words, if the assumptions about future trends in
vital rates and migration are made known and agreed on,
and a starting population is identified, the resulting popu-
lation projection should always be undisputed (assuming
the computations have been done correctly, which, with
the wide availability of computer programs for this pur-
pose, is no longer the problem it was in the past).
The issue then becomes a matter of agreeing on the
composition and size of the starting population, and of the
credibility of future trends in vital rates. These two factors
are the sole grounds for the uncertainty that are inherent
in population projections. The current demographic char-
acteristics of a population are assembled from data
sources such as censuses, registration systems, and sur-
veys that are frequently incomplete or inconsistent. Ad-
justments are often made not only to the total size of a base
population but also to the distribution of the population
by age and sex, the level of mortality, and the pattern of
fertility. Not infrequently, the base population estimates
from which the projections are made are themselves pro-
jections from earlier empirical data. To indicate to the user
how much confidence one may have in the data, a listing
of the sources of the base data, and of the adjustments
made to them, should therefore be a required element of
demographic estimates. The second source of uncertainty
concerns the assumptions of the components of future
population growth-fertility, mortality, migration. As the
future is unknowable, the right approach is to fully explain
and justify the assumptions. Such justification involves
analyzing past trends and indicating how future trends

may be either similar or different. Once these two sources
of forecast error in projections have been addressed, users
will be in a position to decide on the likelihood of the
John Bongaarts' study presents a detailed account of
past trends and determinants of population growth, and
shows how past and current conditions have created the
momentum for future population increases. The popula-
tion figures shown by Bongaarts are for large regions
created by aggregating individual country estimates and
projections. Because of obvious space limitations, no
information on the reliability of the base data is given;
instead, the source of most of the data is indicated to be
the United Nations' 1992 projections. This, unfortunately,
contains only fragmentary indications of the base sources
of the data. Although, in large aggregates, errors in base
data are often offsetting, it should be recognized that some
of the developing country regions, especially the Sub-Sa-
haran Africa aggregate, are combinations of quite uncer-
tain demographic statistics of individual countries. Some
of the basic assumptions of population projections made
by the United Nations-such as fertility rates declining
toward replacement level, and continued improvement in
mortality-are partly explained by Bongaarts, but to
evaluate how plausible the projections are, more specific
information is needed. This is not a shortcoming of the
Bongaarts study, but rather of the United Nations projec-
tions, which suffer from a lack of transparency regarding
their assumptions.
The presentation of the projection results by Bongaarts
is an excellent summary of demographic trends in the
recent past and of expectations for the near future and does
not offer grounds for serious disagreements. Instead, in
this discussion, I will try to expand a bit on the issues
presented by Bongaarts. First, through examination of
previous efforts and of the components of future popula-
tion growth, I examine the question of whether a sense of
confidence can be obtained about the projections. Second,
I elaborate on Bongaarts' projections through examining
the structures of the projected populations.
How certain are the projections? On a global scale,
where migration is not a factor, only differences between

18 Population and Food in the Early Twenty-First Century

future and projected fertility and mortality levels would
be responsible for the projections being inaccurate. Of
these, fertility trends have been most difficult to project
accurately. Some have argued that the decline in fertility
projected by the United Nations and World Bank is too
slow and that the projected populations are therefore too
high (Seckler and Cox 1994). Others have taken the
opposite view and have argued that the projected decline
in fertility is "too optimistic" (Stycos 1994).
Seckler and Cox (1994) and others who propose a
faster decline have analyzed recent trends in fertility rates
(TFRs) and assume that such trends will continue because
the underlying reasons for the decline in fertility, such as
socioeconomic development and cultural changes, will
continue. Their regression analysis of trends found that a
quadratic expression fit the past TFR. However, it cannot
be assumed that this trend will hold at lower levels of
fertility, as the TFR would quickly approach zero. The
essential questions not answered by Seckler and Cox are,
at which level does the quadratic trend no longer apply,
and which functional form will the TFR take then? The
answer implicit in the United Nations and World Bank
projections is that the TFR decline will start following a
geometric pattern when fertility drops to below 3. In the
case of the World Bank projections, this pattern has been
empirically confirmed with historical data; the extent to
which the future will be similar to the past is of course
much more difficult to assess.
Those who question the continued decline in fertility
have argued that a stagnation in the future decline in
fertility is more plausible because those most wanting to
adopt family planning have already done so; reaching
rural and less-educated populations will require greater
effort. Moreover, fertility trends have at times reversed
direction without obvious explanations. The main prob-
lem with this argument is that almost all surveys con-
ducted in the past few years in the Demographic and
Health Survey program show no such deceleration or
reversal at TFR levels above 3, even in settings that may
be considered difficult for fertility reduction from a socio-
economic perspective, such as Bangladesh.
The United Nations projections address this issue im-
plicitly by showing three different paths population may
follow in the future: "high" (slow fertility decline), "me-
dium" (moderate fertility decline), and "low" (rapid fer-
tility decline). There is a tendency among users to consider
mainly the medium scenario as the most likely. However,
as the future course of fertility is not agreed upon, the high
and low projections are useful illustrations of population
growth under alternative patterns of fertility change.
At the World Bank, we also make population projec-
tions, as has already been mentioned. Our projections for
larger regions are without exception very close to those
produced by the United Nations Population Division as
well as those of the U.S. Bureau of the Census, although

for individual countries we sometimes have different re-
sults. The figures for the world and for regions I use are
from the projections made at the World Bank. The fact
that projections made by different agencies are close is of
course no evidence for their accuracy, which can only be
discovered in due time. It is, however, possible to examine
previous projections to see how well the current popula-
tion has been projected, as one predictor of the accuracy
of future demographic projections. Table 1 compares
global projections published at four different times: 1980,
1985, 1990, and 1994. The work for these projections was
usually carried out in the year preceding the year of
publication; the 1994 projections are still preliminary and
not yet published. The 1980 projections were made for
only a 20-year time span. From this comparison, it can be
concluded, first, that in the short term (10-15 years from
the base year), the projections have matched current popu-
lation quite well and, second, that there have been no
events in the past decade to change the projected popula-
tion for the intermediate term, that is, the first two decades
of the twenty-first century. The one projection that is
consistently higher than the others is the one published in
1990. As these projections are the aggregates of individual
country projections, it is not immediately clear why the
1990 global aggregate produced higher population totals,
but a possibility is the upturn in fertility in China in the
late 1980s, which has since then been reversed.

Table 1-Population projections
Published Projections in
Year 1994 1990 1985 1980
1980 n.a. n.a. 4.442 4.416
1985 n.a. 4.844 4.835 4.808
1990 5.271 5.285 5.261 5.224
1995 5.701 5.744 5.701 5.650
2000 6.126 6.204 6.147 6.075
2005 6.546 6.663 6.581 n.a.
2010 6.967 7.112 6.994 n.a.
2015 7.376 7.556 7.395 n.a.
2020 7.775 7.993 7.787 n.a.
Source: Bos et al. 1994.
Note: n.a. means not available.

As the projection of the world is the addition of indi-
vidual country projections, an overprojection in some
regions will offset an underprojection in other regions. To
inspect how the current 1994 projections differ from those
made almost 10 years earlier for regions of the world, the
total projected populations by region are compared in
Figures 1-7. The figures show that in some of the devel-
oping regions the projected population is now lower than
what was projected in 1985, but in other regions projec-
tions are higher. Population projections are now lower in

Comment on Chapter 2: How Reliable Are the Population Projections?

Figure 1-Sub-Saharan Africa, population
projections made in 1985 and 1994

1990 1995 2000 2005 2010 2015 2020
01985 *1994

Source: Bos et al. 1994.

Figure 2-Latin America and the Caribbean,
population projections made in
1985 and 1994

0.4 L I
1990 1995 2000 2005 2010 2015 2020
*1985 *1994

Source: Bos et al. 1994.

Figure 4-Southwest Asia, population
projections made in 1985 and 1994
n (billions)

1990 1995 2000 2005 2010 2015 2020
01985 *1994

Source: Bos et al. 1994.

Figure 5-Developed countries, population
projections made in 1985 and 1994

90 1995 2000 2005 2010 2015 2020
01985 *1994

Source: Bos et al. 1994.

Figure 3-East and Southeast Asia, population
projections made in 1985 and 1994

1.7 I
1990 1995 2000 2005 2010 2015 2020
01985 *1994

Figure 6-South Asia, population projections
made in 1985 and 1994

190 1995 2000 2005 2010 2015 2020
01985 *1994

Source: Bos et al. 1994.

Source: Bos et al. 1994.

20 Population and Food in the Early Twenty-First Century

Figure 7-North Africa, population projections
made in 1985 and 1994

0.10 I
1990 1995 2000 2005 2010 2015 2020
*1985 -1994

Source: Bos et al. 1994.

Figure 8-Components of growth

1990 1995 2000 2005 2010 2015 2020
Momentum *Standard ( Constant Mortality

Source: Bos et al. 1994.

the large South Asia region for which the 1994 projection
is consistently lower, and in the East Asia and Southeast
Asia regions, where the initial population is higher in the
1994 projections but is below the 1985 projections after
2000. Thus, the projections for the regions have been less
consistent than the global projections. If country-level
projections were examined, the level of error would be
even more substantial in some cases.
Another way of examining reliability of population
projections is to look at the components of population
change. At the global level, where migration is zero,
population projections may be inaccurate because either
assumed mortality or fertility trends, or both, are not
accurate. In his study, Bongaarts mentioned the concept
of momentum, which is the growth due to age structure,
as opposed to population growth resulting from fertility
exceeding mortality. If fertility and mortality were as-
sumed to be in exact balance-couples would be having
exactly enough surviving children to replace them-
selves-from 1995 onward, population would increase
from the 1995 projected population of 5.7 billion to a little
more than 7.0 billion because of the momentum built into
the young age structures (Figure 8). That is, more than 60
percent of the projected increase in population between
1995 and 2020 is due to age-structure momentum. Of the
remaining 40 percent, two-thirds is due to fertility being
above replacement and the other one-third is due to as-
sumed improvements in mortality. While momentum is
not entirely invincible (see Bongaarts 1994b for some
options), it most likely will drive future population
growth. Thus, for the same reason that age-structure mo-
mentum is an obstacle for countries wanting to reduce
future population growth, momentum enhances the con-
fidence one can have in the general trend shown in popu-
lation projections. As the number of births in a given year
is the product of the population and the birth rate, knowing
the first part of this equation with some degree of confi-
dence significantly reduces the uncertainty of the projec-

tion. Only catastrophic mortality that would affect the
existing population could significantly alter the upward
trend in total population.
The second issue to be addressed concerns projected
changes in age structures between the current and future
populations. An important purpose of population projec-
tions is not only to present data on the future number of
people but also to be the basis of projections of other
quantities. Population projections are frequently used for
planning and project purposes, such as planning for edu-
cational facilities, family planning logistics, the demand
for health care, or future social security beneficiaries. In
most of these cases demographic projections are used, but
with an interest only in particular components or age
groups of the total population. As all population projec-
tions are so-called "cohort-component" projections, in
which age-specific rates are applied to age cohorts, they
are well suited to provide this information.
The projections of total population that Bongaarts pre-
sented might be called "summary measures" of population
projections in that they are derived from the projections of
the age groups that make up the population. To indicate the
projected changes in age structure that are expected in the
next few decades, Figure 9 shows age pyramids, that is,
graphical representations of the distribution by sex and age
for developing countries and developed countries for 1990,
2010, and 2025. The pyramids show the percentage of the
population in each five-year age group. It can be seen that
substantial changes will be taking place in the age structure
of the populations. For many socioeconomic planning pur-
poses, such changes may be of greater importance than
changes in the absolute population. The age structure
changes are such that by 2010 demographers will no longer
be able to talk of the age pyramid of developed countries,
but instead may be referring to the graphical representation
of the age structure as the age rectangle. For many pur-
poses, these changes will be more consequential than the
projected changes in the total population.

Comment on Chapter 2: How Reliable Are the Population Projections? 21

Figure 9-Percentage distribution of the population (X axis) by five-year age group (Y axis) for males
and females, 1990, 2010, and 2025

Developing Countries 1990


14 12 10 8 6 4 2 0 2 4 6 8 10 12 14
mMales E Females

Developed Countries 1990

10 8 6 4 2
mMales o Females

Developing Countries 2010

12 10 8

6 4 2 0 2 4 6 8 10 12

Developed Countries 2010


8 6

4 2
E3 Females

0 2 4 6 8 10

Developing Countries 2025


Developed Countries 2025


7 6 5 4 3 2 1 0 2 4 6 8 10 12
IMales Eo Females

Source: Bos et al. 1994.

10 8 6 4 2 0 2 4 6 8 10
*Males OFemales

22 Population and Food in the Early Twenty-First Century

Bongaarts, J. 1994a. Global and regional population
projections to 2025. Paper prepared for the Round-
table on Population and Food in the Early Twenty-
First Century, 14-16 February, International Food
Policy Research Institute, Washington, D.C.
1994b. Population policy options in the de-
veloping world. Science 263 (February 11): 771-
Bos, E., M. T. Vu, E. Massiah, and R. A. Bulatao. 1994.
World population projections, 1994-95. Baltimore:
Johns Hopkins University Press.

Haub, C. 1987. Understanding population projections.
Population Bulletin 42-4. Washington, D.C.: Popu-
lation Reference Bureau.
Seckler, D., and G. Cox. 1994. Population projections
by the United Nations and the World Bank: Zero
population growth in 2035? Unpublished manuscript.
Stycos, J. M. 1994. Population, projections, and policy:
A cautionary perspective. EPAT/MUCIA Working
Paper 12.
United Nations. 1993. Worldpopulation prospects: The
1992 revision. New York.

Global and Regional Food Demand and Supply Prospects

Chapter 3

By Nikos Alexandratos

This chapter presents the prospects for world food and
agriculture to 2010, drawing on recent work for the FAO
study Agriculture: Towards 2010 (FAO 1993a; a revised
and expanded version has been published as Alexandratos
1995).1 The study is more than a mere "commodity
projections" exercise, as it covers a wide array of issues
(including those of the forestry and fisheries sectors) and
pays particular attention to the issues related to food
security, natural resources, the environment, and sustain-
ability. For reasons of space, only the demand and supply
prospects for the basic food commodities are covered
here. Yet the importance of other agricultural commodi-
ties should not be underestimated in any analysis of the
food security prospects of the developing countries, par-
ticularly those that depend on production for export of, for
example, sugar, tropical beverages, and raw materials for
rural income generation and for financing food imports.
After a brief summary of the methodology, the chapter
draws attention to the progressive slowdown of world
agricultural growth and to the possibility that such a
slowdown will continue in the future and examines the
significance of these developments for the problems of
food, population, and food security. This is followed by a
presentation of the projections for the per capital food
supplies and their implications for food and nutrition in
the developing countries. The next section presents the
outlook for the cereal and livestock sectors in both the
developed and the developing countries. The rest of the
chapter deals only with developing-country outcomes-
possible developments in their other (noncereal) basic
food crops, possible developments in their agricultural
trade balance, and possible land-yield developments un-
derpinning future production growth. The final section
draws some overall conclusions.

Highlights of the Methodology
All the analyses and projections were carried out for a
large number of commodities or commodity groups (52
crop and livestock products for food demand analysis, 32
for supply-and-demand analysis, and 40 for production
analysis) and for practically all countries individually.2

The overall approach of this study was to start with
projections of demand using Engel demand functions and
exogenous assumptions of population and gross domestic
product (GDP) growth. The former are from the 1990
United Nations assessment of population prospects, and
the latter are largely based on the World Bank's baseline
GDP projections for 1992-2002 (World Bank 1993).
Production projections were then developed by first ex-
amining provisional targets for production, for each com-
modity and country, that were derived from simple rules
about future self-sufficiency and trade levels. These tar-
gets then underwent several rounds of iterations and ad-
justments based in part on consultations with specialists
on the different countries and disciplines, with particular
reference to what are considered to be "feasible" levels
of land use, yields, and trade. Accounting consistency
controls at the commodity, land-resource (developing
countries only), country, and world levels had to be re-
spected throughout. In addition, but only for the cereal,
livestock, and oilseed sectors, a formal flex-price model
(the FAO World Food Model, FAO 1993b) was used to
provide starting levels for the iterations and to keep track
of the implications for all variables of the changes in any
one variable introduced in the successive rounds of in-
spection and adjustment. The model is composed of sin-
gle-country modules and world-market feedbacks leading
to national and world market clearing through price ad-
The results of the model projections (whether the sin-
gle Engel demand functions or the flex-price model) were
also subjected to many rounds of adjustments by special-
ists on countries and many disciplines, particularly during
the phase of analyzing the scope for production growth
and trade. The projections presented here are definitely
not "trend extrapolations."

Overall Agricultural Growth in
Relation to Population

The world agricultural growth rate has been slowing down
in the last three decades. It was 3.0 percent per year in the

26 Population and Food in the Early Twenty-First Century

1960s, 2.3 percent per year in the 1970s, and 2.0 percent
per year in the period 1980-92. The slowdown has been
more pronounced since the mid-1980s. Indeed, 1984-92
is the first eight-year period of all eight-year moving
periods from 1961 to 1992 when the growth rate of world
per capital production (all products) fell to zero, while the
decline in the per capital production of cereals was the
deepest (Figures 3.1 and 3.2). These developments have
been interpreted by many (for example, Brown 1994) as
a turn for the worse and have given rise to expressions of

concern about the capability of world agricultural produc-
tion to keep up with the growth of world population.
The significance of this slowdown for the world food
security issue, that is, the access to food of the population
groups with inadequate nutrition, cannot, however, be
deduced from developments at the global level. The slow-
down in the production growth rate of cereals originated
predominantly in the major net cereal-exporting countries
(United States, Canada, European Community-12, Aus-
tralia, Argentina, Thailand), which accounted for 36 per-

Figure 3.1-Growth rates of production (world, all products), moving eight-year periods, 1961-92
Growth Rate (percent per year)

1 Per Capita Production


-1 I I I I I I l Y 5 56
A -& o\ ^ 4 A4 b \ b S c' c.9

Figure 3.2--Growth rates of production (world, cereals), moving eight-year periods, 1961-92
Growth Rate (percent per year)

4 -

Total Production

Per Capita Production


q ilc l l ll /bN/4b /1b 4qNN

Chapter 3: The Outlook for World Food and Agriculture to Year 2010 27

cent of world production in the mid-1980s. Their produc-
tion was 602 million tons3 in 1990-92 (824 kilograms per
capital down from 621 million tons in 1984-86 (888
kilograms per capital In contrast, production in the rest
of the world rose over the same period from 1,040 million
tons to 1,135 million tons and per capital production
remained largely unchanged. Some of the major exporting
countries had to resort on several occasions during this
period to supply-management measures and reduction of
incentives to producers to rein in the growth of produc-
tion. It is therefore difficult to interpret the decline in their
production as indicative of the onset of a process whereby
production constraints became more binding and effective
demand could not be met at the going price. It is more
likely that the sluggish growth of export demand for their
output dominated the picture.
The growth rate of gross world agricultural output may
slow down further to 1.8 percent per year in 1988-90 to
2010. Even so, it will continue to outpace the 1.5 percent
per year growth rate of world population, although the
growth rate of per capital production may also be lower
than in the past. No food welfare conclusions may be
deduced from this expected decline, mainly because per
capital food consumption is a bounded variable, with a
(loosely defined) ceiling not to be exceeded for physi-
ological reasons. Therefore, if the growth rate of world
per capital consumption falls and eventually tends to zero,
it may be a good thing (indicating that a high proportion
of people are well fed) or a bad thing. Which is which
cannot be discerned without more disaggregation and
resorting to other criteria. This is not only because a world
indicator averages out good and bad outcomes, but also
because it can create the illusion that the world situation
is a zero sum game: if the world average does not increase,
then any gains of one group must be at the expense of
another group. This need not be the case, however, as
developments in the per capital production of cereals (past
and projected) show (Table 3.1). This variable increases
in both the developed and the developing countries, yet
the world average remains flat.
Given the limited value of the world average produc-
tion as an indicator of changes in food welfare, the analy-
sis must focus on the cases when a slowdown in the

growth rate of production may be conceived as threaten-
ing the food welfare (1) if it takes place in those countries
that have low nutritional levels and in which agriculture
is a major source of employment and income for a good
part of the population, and (2) if the slowdown of produc-
tion growth were to be associated with rising food prices
affecting the poor everywhere.
On all of these counts, a slowdown in the growth rate
of production in the developed countries may not be
interpreted as threatening their food welfare. It is more
likely the result of demand constraints: more is not pro-
duced because it is not needed, although this statement
applies less to the former centrally planned economies
(CPEs) of Eastern Europe in their current state of transi-
tion. However, developed countries account for some 50
percent of world gross output (see Table 3.2). Therefore,
the situation is one in which about one-half of the world
output need only grow at a rate in the vicinity of the very
lowest rate of the developed countries' population.
In contrast, a slowdown in the agricultural growth rate
of the developing countries may often be interpreted as a
threat to their food security because of their high eco-
nomic dependence on agriculture, low levels of per capital
food supplies, and limited capacity to increase supplies
through imports.
Of the 93 developing countries of the study, 31 may be
classified as having relatively "low" dependence on agri-
culture for employment and income generation because
they have less than one-third of their economically active
population in agriculture. This group has a population of
730 million or 19 percent of the developing countries' total
and average per capital food supplies of 2,910 calories per
day; and all countries in it but five are in the medium-high
range of this variable (2,600 to 3,300 calories per day).
The remaining 62 developing countries of the study are
in the "high" dependence group, that is, they have more
than one-third of their economically active population in
agriculture, and all but four have per capital food supplies
of less than 2,600 calories, with the majority of them
nearer the 2,000-calorie mark. The group average is 2,370
calories or 2,230 calories, if China is excluded. Moreover,
the majority of the countries in this group have compara-
tively low net food imports per capital, and some depend

Table 3.1-Per capital production of cereals (with rice included in milled terms)
Production Populationa ProductionlCapita
Countries 1979-81 1990-92 2010b 1980 1991 2010 1979-81 1990-92 2010
(million tons) (millions) (kilograms)
Developed countries 793 873 1,016 1,170 1,262 1,406 678 692 722
Developing countries 651 883 1,318 3,277 4,125 5,744 200 214 229
World 1,444 1,756 2,334 4,447 5,387 7,150 325 326 326
Source: United Nations 1993.
81992 United Nations assessment.
bProduction projections from Table 3.4.

28 Population and Food in the Early Twenty-First Century

Table 3.2-Population, per capital food supplies and agricultural production (all products)

Population Per Capita Food Supplies
Region 1990 2010 1980-90 1990-2010 1969-71 1988-90 2010
(millions) (percent per year) (calories per day)
World 5,295 7,150 (7,209) 1.8 1.5 (1.6) 2,430 2,700 2,870
Developing countries 3,984 5,668 (5,758) 2.1 1.8 (1.9) 2,120 2,470 2,730
Group 1, (low dependency) 731 1,049 (1,054) 2.3 1.8 (1.8) 2,490 2,910 3,160
Group 2, (high dependency) 3,253 4,619 (4,704) 2.1 1.8 (1.9) 2,040 2,370 2,630
Group 2, excluding China 2,120 3,234 (3,333) 2.4 2.1 (2.3) 2,070 2,230 2,420
Group 2, excluding China and India 1,274 2,045 (2,110) 2.6 2.4 (2.5) 2,090 2,230 2,380
Developed countries 1,253 1,406 (1,373) 0.7 0.6 (0.5) 3,200 3,400 3,470
Former CPEs 389 433 (435) 0.7 0.5 (0.6) 3,310 3,380 3,380
Others 864 973 (938) 0.7 0.6 (0.4) 3,140 3,410 3,510
Developing countries, by region
Sub-Saharan Africa 486 874 (915) 3.1 3.0 (3.2) 2,140 2,100 2,170
Near East/North Africa 310 513 (493) 2.9 2.6 (2.4) 2,380 3,010 3,120
South Asia 1,157 1,679 (1,728) 2.2 1.9 (2.0) 2,040 2,220 2,450
East Asia 1,597 2,008 (2,001) 1.6 1.2 (1.2) 2,020 2,600 3,060
East Asia, excluding China 464 623 (630) 2.0 1.5 (1.5) 2,100 2,500 2,890
Latin America and the Caribbean 435 593 (622) 2.1 1.6 (1.7) 2,500 2,690 2,950
Source: United Nations 1993.
Notes: Historical growth rates are ordinary least square estimates from annual data.
aPopulation data and projections from the 1992 United Nations assessment (United Nations 1993). The projections of the 1990 United
bNinety-three developing countries of the study accounting for 98.5 percent of total population in the developing countries.

heavily on agricultural exports for their balance of pay-
ments, and hence for their capacity to import food.
These country groups can now be examined to ascer-
tain whether their agricultural growth experience after
1984 represents a turn for the worse. The relevant data
(growth rates of aggregate and per capital production, for
all products, for all moving eight-year periods) are plotted
in Figures 3.3-3.6. As these figures show, (1) for the
developing countries as a whole, the growth rates of per
capital agricultural production (all products) have not gen-
erally been lower in recent eight-year moving periods
compared with earlier ones (Figure 3.3); (2) the same
observation applies also to the low- and high-dependency
groups of countries (Figures 3.4 and 3.5); and (3) for the
most vulnerable group of countries (those with high de-
pendence on agriculture, excluding China), the per capital
growth rates of recent years have actually been higher than
in earlier periods (Figure 3.6).
In the light of this evidence, it is difficult to accept the
position that developments in recent years have marked a
turning point for the worse. It is more appropriate to
consider that trends in recent years, just like those of the
longer-term past, have been characterized by the persis-
tence of low and totally inadequate per capital growth rates
of production in the countries whose food security would
profit significantly from higher production. Putting the
issue in these terms goes to the heart of any analysis of the
nature and significance of agricultural growth trends,
which must always be evaluated in relation to needs rather
than in abstract.

The preceding analysis of the historical experience
may now be used to examine the significance of the
further slowdown projected for the growth rate of world
agricultural production. The relevant historical data and
projections are shown in Table 3.2. For the low-depen-
dence group of the developing countries (group 1 in Table
3.2), the growth rate for 1988-90 to 2010 (2.4 percent per
year) is projected to be well below that of the preceding
two decades 1970-92 (3.0 percent per year). However,
this deceleration was well under way in the 1980s and
group l's population growth also slowed down consider-
ably, resulting in the difference between the past and the
future growth of per capital production almost disappear-
ing. The conclusion is that for this group of countries the
projected slower growth in agriculture may not be inter-
preted as a structural change (compared with the past) in
whatever fundamental relationships exist between the
growth rate of agriculture and food security. This can also
be seen from the prospective developments in these coun-
tries' per capital food supplies for direct human consump-
tion (projected to rise to 3,160 calories per day).
The data and projections in Table 3.2 show that no
slowdown in the growth of per capital production is fore-
seen for the high-dependence group, excluding China.
Indeed, if India is excluded from this group, the projec-
tions imply a modest acceleration in the growth rate of
production. This is not to say that agricultural growth will
be sufficiently high to support significant improvements
in food security. It only indicates that the growth rate of
production may not be deteriorating in relation to that of

Chapter 3: The Outlook for World Food andAgriculture to Year 2010

Growth Rates of Gross Agricultural Production
Total Per Capita
1970-80 1980-92 1970-92 1988-90 to 2010 1970-80 1980-92 1970-92 1988-90 to 2010
(percent per year)
2.3 2.0 2.2 1.8 0.5 0.3 0.4 0.2
3.0 3.3 3.3 2.6 0.7 1.2 1.1 0.8
3.4 2.7 3.0 2.4 0.9 0.4 0.6 0.5
2.8 3.6 3.4 2.7 0.6 1.5 1.3 0.8
2.4 3.1 2.9 2.7 0.0 0.7 0.4 0.4
2.3 2.8 2.7 2.8 -0.3 0.2 0.0 0.3
1.8 0.7 1.2 0.7 0.9 0.0 0.4 0.2
1.3 0.5 0.7 0.4 0.5 -0.2 -0.1 -0.1
2.0 0.8 1.4 0.8 1.2 0.1 0.6 0.4

1.0 2.6 1.8 3.0 -1.9 -0.5 -1.2 -0.2
3.0 3.2 3.1 2.7 0.3 0.3 0.3 0.3
2.6 3.5 3.1 2.6 0.3 1.2 0.9 0.6
3.6 4.0 4.1 2.7 1.6 2.4 2.4 1.5
4.2 3.4 3.9 2.6 1.6 1.4 1.7 1.0
3.2 2.2 2.6 2.3 0.7 0.1 0.4 0.6

Nations assessment (which underlie the food and agriculture projections in this paper) are shown in parentheses.

the population. An improvement in the growth rate of per
capital production was already evident in the 1980s com-
pared with the 1970s, and nearly all the countries of
Sub-Saharan Africa are in this group. The issue of future
progress depends in large measure on the assumptions
made about the longer-term success of policy reforms in
this region.
The provisional conclusion is that the significance of the
projected further slowdown in world agricultural growth for
food security should be interpreted with care. This is be-
cause it takes place in those countries whose food security
may not be greatly dependent on the growth of their agri-
culture, that is, the developed countries, and those among
the developing countries with low dependence on agricul-
ture and relatively high per capital food supplies.
However, the persistence of low (though higher than
in the past) growth rates in the per capital production of
the vulnerable countries does indicate growth in agricul-
ture that is totally inadequate for improving food security.
For this group of countries, the issue is not whether per
capital production may be slightly higher than in the past
but why it cannot be projected to be significantly higher,
for example, 1.0 to 1.5 percent per year. The why should
be intuitively apparent to the reader. For example, for
Sub-Saharan Africa a growth rate of 1 percent per year in
per capital production would mean a greater than 4.0
percent growth rate in aggregate output.4 Although this
aggregate output growth rate may not be impossible to
achieve, both demand and production constraints militate
against such an optimistic assumption.

On the demand side, the constraints are the low (ex-
ogenous) GDP growth, assumed to be 0.7 percent per year
per capital 5 the large share of the roots and tubers sector
in total consumption; and the poor world market prospects
for the main exportables. On the production side, only one
major region (East Asia) sustained average production
growth rates of 4.0 percent per year over a period of two
decades. With all the caveats attached to statements about
the future, it is reasonable to doubt whether Sub-Saharan
Africa could replicate this feat, given its much more
adverse production environment, including both agro-
ecological conditions (for example, the limited role of
irrigation in the region and the region's higher vulnerabil-
ity to weather factors) and weaker infrastructure (institu-
tional, research, and physical; see Platteau 1990).
The preceding discussion was geared to making the point
that the world, as a whole, could continue to increase
agricultural production at a rate faster than that of popula-
tion growth and in any case at a rate that could match the
growth of effective demand, probably at nonincreasing
prices.6 However, it must be emphasized that this conclu-
sion should in no way be interpreted as providing any
answers, let alone optimistic answers, to some fundamental
issues, such as (1) whether the future evolution of demand
and supply will be of a pattern and magnitude to meet
normative objectives of nutrition and income generation for
the poor, in short to ensure food security for all; (2) whether
future growth in production can be achieved at zero envi-
ronmental cost or, in any case, at acceptable tradeoffs
between more production, the environment, and sustain-

30 Population and Food in the Early Twenty-First Century

Figure 3.3--Growth rates of production (all developing countries, all products), moving eight-year
periods, 1961-92
Growth Rate (percent per year)

Total Production

Per Capita Production


Figure 3.4-Growth rates of production (developing countries, low-dependence group, all products),
moving eight-year periods, 1961-92
Growth Rate (percent per year)

4 Total Production


2 -
Per Capita Production

0 I11I

-1 (b Ib I 4II
-1' 4 Iq1 N-'1 'q 0 I
~,6 A6i' 9 qZ Z9~Z9h 98

ability; and (3) the extent to which constraints related to
natural resource scarcities and the environment represent
obstacles to making progress toward food security.
In relation to this latter point, the conclusion that the
world as a whole has the potential to increase output faster
than population and meet the growth of effective demand
at probably nonincreasing prices in no way implies that
resource and environmental constraints are not important

obstacles to agricultural growth and improved food secu-
rity at the local level, particularly in low-income countries
that are not integrated, or are imperfectly integrated, into
the world economy. So long as such limited integration
persists, countries with limited agricultural resources and
high economic dependence on them will continue to be
constrained in their efforts to improve food security by
genuine resource and environmental constraints. This

Chapter 3: The Outlook for World Food and Agriculture to Year 2010 31

Figure 3.5-Growth rates of production (developing countries, high-dependence group, all products),
moving eight-year periods, 1961-92
Growth Rate (percent per year)

3""'^*^ Total Production --

Per Capita Production


0 ~ ~ ~ A\---4-----0--P--5----gib-------q-----
1 p I I I I I I I I I I I I

Figure 3.6-Growth rates of production (developing countries, high-dependence group, excluding
China, all products), moving eight-year periods, 1961-92

Growth Rate (percent per year)

( ,-" V" 6 ," A' S, < ," C?" V, ," V, b9 ,-- \ A A % V'- Z, Vo" o < ,` ? 1

kind of situation can coexist, as indeed it does at present, sion of these issues, including issues of world fo
side-by-side with the availability of perhaps ample poten- lation balance beyond 2010, see Alexandratos 1
tial for increasing production in the world as a whole. This
issue is underlined here to dispel any notion that simple Prospects for Food and Nutrition
global answers can be given to the question of whether This section presents the projections for per ca
progress toward improved food security is impeded by plies for direct consumption of the cereals and
increasing natural resource scarcities. (For further discus- commodities (kilograms per year) as well as th(


Lpita sup-
)se of per

32 Population and Food in the Early Twenty-First Century

capital food supplies from all food commodities (calories
per day). Subsequently, it explores what these projections
may imply for developments in the incidence of chronic
undernutrition. The term "direct food consumption" cor-
responds to the food balance sheet concept of food sup-
plies from production and net imports of food
commodities for direct human consumption, that is, netted
out for nonfood uses (for example, feed, seed) and waste.

Per Capita Food Supplies: Cereals,
Livestock, and Total Food

The per capital direct food consumption of cereals may
increase only slightly in the developing countries and the
world as a whole (Table 3.3). This is the net result of the
projections carried out for each of the individual countries
and cereal commodities (wheat, milled rice, maize, barley,
sorghum, millet, other coarse grains).7 The little growth
in per capital direct consumption of cereals results from
(1) demand diversification toward other products in the
countries with relatively high levels of direct consumption
of cereals per capital, generally more than 200 kilograms;
and (2) a combination of inadequate growth in per capital
incomes-and, in some cases, supply constraints-in
countries with still low levels of cereals per capital. Supply

constraints were taken into account in the process of
revising and adjusting the first-round projection results in
recognition of the fact that, in the low-income countries
with high population shares in agriculture, the consump-
tion per capital is likely to be determined in a major way
by the growth of production supplemented by limited
potential to increase net food imports.
The nonfood use of cereals, mostly for animal feed, in
the developing countries will continue to grow. Their per
capital use of cereals for all purposes is projected to grow
from 235 kilograms in 1988-90 to 254 kilograms in 2010.
This increment is less than half as large as that of the
preceding two decades. It is noted, however, that much of
the growth in the historical period had occurred in the
1970s (190 kilograms in 1969-71, 220 kilograms in
1979-81,235 kilograms in 1988-90), the period of rapid
growth in their cereal imports. In contrast, the per capital
use of cereals for all purposes in the developed countries
may decline somewhat, because of rather significant de-
clines in the group of the former CPEs. The projected
decline reflects the assumption that, with the policy re-
forms, the grain-to-meat ratios and the postharvest losses
(included in the total use statistics) will be reduced. It also
reflects the prospect that, even after recovery from the
deep decline in the early years of the reform, the per capital

Table 3.3--Per capital food supplies for direct human consumption
Commodity 1969-71 1988-90 2010 1969-71 1988-90 2010
(kilograms per capital)
World All developing countries
Cereals (for food) 146.3 164.6 167.0 145.3 170.5 173.0
Cereals (all uses) (305.0) (331.0) (325.0) (190.0) (235.0) (254.0)
Meat 26.0 31.9 37.0 10.5 16.4 25.0
Milk 74.6 75.3 72.0 27.4 35.9 42.0
All food (calories per day) 2,430.0 2,700.0 2,860.0 2,210.0 2,470.0 2,730.0
Sub-Saharan Africa Near East/North Africa
Cereals (for food) 115.3 114.3 121.0 183.3 212.9 210.0
Cereals (all uses) (140.0) (133.0) (140.0) (295.0) (376.0) (386.0)
Meat 10.5 9.5 10.0 13.0 18.4 23.0
Milk 28.1 27.6 26.0 54.0 59.4 67.0
All food (calories per day) 2,140.0 2,100.0 2,170.0 2,380.0 3,010.0 3,120.0
East Asia South Asia
Cereals (for food) 151.0 200.5 206.0 148.5 155.8 163.0
Cereals (all uses) (192.0) (272.0) (319.0) (167.0) (177.0) (181.0)
Meat 8.6 20.3 40.0 3.8 4.2 6.0
Milk 3.5 6.8 9.0 36.8 53.5 63.0
All food (calories per day) 2,020.0 2,600.0 3,040.0 2,040.0 2,220.0 2,450.0
Latin America and the Caribbean All developed countries
Cereals (for food) 118.8 128.9 139.0 148.6 146.3 141.0
Cereals (all uses) (224.0) (260.0) (296.0) (583.0)a (637.0)a (633.0)a
Meat 33.3 39.4 49.0 63.3 80.5 87.0
Milk 83.3 92.0 96.0 188.4 199.1 198.0
All food (calories per day) 2,500.0 2,690.0 2,950.0 3,200.0 3,400.0 3,470.0
Former CPEs Other developed countries
Cereals (for food) 195.9 171.5 169.0 127.5 135.0 128.0
Cereals (all uses) (686.0)a (828.0)a (720.0)a (538.0) (550.0) (590.0)
Meat 49.9 71.6 70.0 69.2 84.6 94.0
Milk 189.0 178.9 176.0 188.1 208.1 208.0
All food (calories per day) 3,310.0 3,380.0 3,380.0 3,140.0 3,410.0 3,510.0
aData and projections of the former CPEs are before conversion of the former Soviet Union's production data from bunker to clean
weight (see Table 3.4). Revised data for former CPEs: 1969-71, 652 kilograms; 1988-90, 790 kilograms; 2010, 693 kilograms.

Chapter 3: The Outlook for World Food and Agriculture to Year 2010 33

meat and milk consumption in the former Soviet Union in
2010 would still be somewhat below the levels of the
prereform period. Such declines may be compensated for
the former CPEs as a whole by modest increases in the
corresponding consumption levels in Eastern Europe.
These prospective developments in the former CPEs may
have significant impacts on the world cereal markets and
hence for the developing countries.
Of the other commodities in Table 3.3, the modest
further growth projected for the per capital consumption
of milk in the developing countries reflects, among other
things, the prospect that subsidized surpluses from the
developed countries will be rarer in the future compared
with the past. Thus, a major factor that underpinned the
rapid growth ofmilk consumption in the past may be much
less important in the future. A related factor has to do with
the fact that the most rapidly growing developing countries,
in East Asia, are only modest consumers of milk.

Implications for Nutrition in the
Developing Countries
The FAO and World Health Organization (WHO) docu-
ments for the 1992 International Conference on Nutrition
provided estimates indicating that some 800 million peo-
ple in the developing countries are chronically undernour-
ished (FAO/WHO 1992). This means that, on average in
the course of a year, their food consumption is below that
required to maintain body weight and support light activ-
ity. This threshold level (ranging from 1,780 calories per
day in Asia to 1,985 calories per day in Latin America) is
equal to 1.54 times the basal metabolic rate. These esti-
mates are based on the assumption that the distribution of

per capital food consumption (in terms of calories) is
log-normal with a mean equal to the per capital calories of
the food balance sheets and a standard deviation as esti-
mated (or guesstimatedd") generally from household sur-
veys (FAO 1992).
The per capital food projections (all food commodities)
in terms of calories presented in Table 3.3 can, therefore,
be used to understand how the incidence of chronic un-
dernutrition may evolve in the future, assuming the stand-
ard deviation of the distribution were to remain constant.8
The results are shown in Figure 3.7.
Predictably, chronic undemutrition is likely to persist
at significant levels (some 650 million, 11 percent of the
projected population of the developing countries) because
many countries, and the whole region of Sub-Saharan
Africa, are projected to make little progress in raising their
very low per capital food supplies. However, significant
reductions are projected for East Asia, the Near
East/North Africa,9 and Latin America and the Caribbean,
and more modest ones for South Asia. The conclusion is
that the problem of chronic undernutrition in terms of
absolute numbers affected will likely be shifting from
South Asia to Sub-Saharan Africa, no matter that the latter
region's population in 2010 will still be half as large as
that of South Asia.

Prospects for the Cereal and
Livestock Sectors in Developed and
Developing Countries
Cereal production accounts for 30 percent of gross agri-
cultural output of the developing countries and 22 percent
of the developed countries. The shares of livestock pro-

Figure 3.7-Chronic undernutrition in developing countries
Million Persons1







245 (35)





42 (24)

1969-71 1979-81 1988-90
*Sub-Saharan Africa QENear East/North Africa
0 South Asia MLatin America and the Caribbean
1Numbers in parentheses are percentages of total population.

o East Asia

34 Population and Food in the Early Twenty-First Century

duction are 27 percent (developing countries) and 52
percent (developed countries), while those of the other
food crops (for example, roots and tubers, pulses, sugar,
oilseeds, fruit, and vegetables) are 36 percent and 24
percent, respectively. In addition, the developing coun-
tries produce most of the world's output of nonfood crops
(tropical beverages and raw materials). The differing mar-
ket prospects for each commodity sector are an important
determinant of the growth rate of total agricultural pro-
duction in each country group.

The Cereal Sector

Developed Countries
The possible developments of cereal production in devel-
oped countries to the year 2010 are summarized in Table
3.4 and compared with those for the developing countries.
Total cereal production in the developed countries may
grow by some 180 million tons, a 21 percent increase (or
0.9 percent per year) over the 850 million tons of 1988-90.
This is a lower rate of growth than the average achieved
in the last 20 years (1.5 percent per year). However, most
of the past increase had occurred in the 1970s and the
growth rate had fallen to 0.8 percent per year in the 1980s.
Thus, the trends established in the more recent past may
prevail in the next 20 years, but with significant differ-
ences among the major groups of countries. In particular,
North America and Oceania are likely to increase their
share in the total production of the developed countries
while Western Europe's total cereals production in year
2010 may be no more than 7 percent above the 1988-90
levels, compared with an increase of 43 percent between

1969-71 and 1988-90. The region's domestic use of
cereals may grow in the future, all of it for feed, compared
with the decline experienced in the 1980s, because cereals
would become more competitive with cereal substitutes,
if the reforms under way were to be fully implemented.
One of the major forces that might shape the future is
the possibility that total domestic use of cereals in the
former CPEs of Eastern Europe may be somewhat below
that of the prereform, three-year average 1988-90, for the
reasons discussed earlier. These developments in the total
use of cereals in the former CPEs of Eastern Europe, in
combination with recovery and resumption of modest
growth in their production, are likely to have profound
implications for world food markets. Eastern Europe will
most likely be a progressively growing net exporter of
cereals, and the former Soviet Union will probably be a
much smaller net importer and could progressively be-
come 100 percent self-sufficient or even turn into a net
exporter.10 Given the uncertainties concerning the possi-
ble developments in this group of countries, the working
assumption used in this study is that net export availabili-
ties of Eastern Europe could be some 5 million tons, while
the net import requirements of the former Soviet Union
could be zero. These assumed net trade outcomes for the
former CPEs, together with those of the developing coun-
tries and the other developed countries (Japan, Israel,
South Africa), define the scope for the growth of net cereal
exports from the three major net exporting regions (West-
ern Europe, North America, Oceania). Their combined net
exports may grow from 158 million tons in 1988-90 to
190 million tons by the year 2010. Only about one-half of
the growth of net import requirements of the developing

Table 3.4-Total cereals, possible developed country outcomes in a world context (million tons, with
Main Net Exporting Regions Other Western
W. Europe N. America Oceania Subtotal Countries
1969-71 143.6 243.3 14.6 401.5 21.2
1979-81 178.5 341.9 21.7 542.1 24.3
1988-90 206.4 313.5 22.9 542.8 22.8
1991-92c 206.4 365.7 22.4 594.5 17.4
2010 n.a. n.a. n.a. 679.0 31.0
Total Domestic Use
1969-71 166.3 191.6 6.3 364.2 35.5
1979-81 189.1 202.3 7.3 398.8 46.8
1988-90 182.6 227.3 8.6 418.5 52.0
2010 200.0 279.0 11.0 490.0 63.0
Net Trade
1969-71 -23.8 49.6 9.0 34.8 -14.6
1979-81 -11.4 129.4 14.7 132.6 -22.9
1988-90 24.4 118.7 14.5 157.5 -28.6
2010 n.a. n.a. n.a. 189.0 -32.0
Notes: n.a. means not available.
aNumbers in parentheses are revised data following the change in reporting cereals production in the former Soviet Union from bunker
been roughly adjusted to reflect this change.
bData and projections for all developing countries, that is, including those not included among the 93 countries covered individually in
cNet imports of all developing countries after deducting 30 million tons of net exports from the exporting developing countries from the
dThe latest production data for the two-year average 1991-92 are shown here to underline the recovery of North American production

Chapter 3: The Outlook for World Food and Agriculture to Year 2010

countries will appear as increasing the demand for net
exports, the other half being cancelled by the assumed
disappearance of the former CPEs as a major net importer.
How the shares of the three exporting regions in the
above-indicated increment in the net import requirements
of the rest of the world will evolve is uncertain. Most
studies that have addressed this question suggest that,
following the Common Agricultural Policy reform and
the agreement on agriculture of the Final Act of the
Uruguay Round, Western Europe's net exports of cereals
may in the future be somewhat below present levels.
Therefore, all the above-mentioned increments in the com-
bined net exports of the three-country groups, and some
more, will likely accrue to North America and Oceania.

Developing Countries
Table 3.5 shows the overall picture for cereal production in
developing countries. The possible developments in pro-
duction and consumption shown in Table 3.5 mean that the
net cereal import requirements of the developing countries
would continue to grow, though slowly--more like the path
of the 1980s than the explosive one of the 1970s. Thus, net
cereal imports may grow from the 90 million tons of 1988-
90 to some 160 million tonsil in 2010 and the aggregate
cereal self-sufficiency ratio may decline a little to 90 per-
cent. About one-half of the total increment in the net import
requirements would be for the Near East/North Africa. The
balance would be mostly for Latin America and Sub-Saha-
ran Africa, and only a minor part would be for South Asia
and East Asia, assuming China (mainland) would continue
to be only a small net importer.

The prospect that Sub-Saharan Africa's net cereal im-
port requirements may more than double to nearly 20
million tons may be viewed with alarm given the region's
difficult balance-of-payments situation and heavy de-
pendence on food aid. This possible outcome suggests a
continued and possibly expanded role for food aid in the
future in this region. Still, the region's net import require-
ments are, and will likely remain, a small part of the total
cereal deficit of the developing countries. Imports are also
entirely inadequate for the region's own needs, since even
if its production increased by 3.4 percent per year (as
assessed by this study), total availabilities would still be
grossly inadequate to raise more than marginally the very
low consumption levels. Thus, while it is a matter of
concern that the region's net cereal imports may more than
double in the next 20 years, the real concern should be
how to define a superior outcome for the region. Such an
outcome would be composed of higher growth of agricul-
tural output (though not necessarily of cereal), exports,
and overall incomes and would lead to higher demand
despite, most likely, even higher net food imports.12
Currently total domestic use of cereals in the develop-
ing countries amounts to 930 million tons (with rice
included in milled terms), of which 90 million tons comes
from net imports from the developed countries. Direct
food consumption absorbs some 670 million tons (72
percent of the total, but with wide regional differences, for
example, from 88 percent in South Asia to 56 percent in
the Near East/North Africa). Animal feed accounts for
some 160 million tons (17 percent of the total, again with
wide regional differences), while the remaining 100 mil-

rice milled)
Former CPEs All Developed All Developing
E. Europe F. Soviet Uniona Subtotal Countries Countriesb World

55.1 168.9 (157.5) 224.0 (213.0) 647.0 (635.0) 482.0 1,129.0 (1,117.0)
68.3 169.6 (158.3) 238.0 (227.0) 804.0 (793.0) 652.0 1,457.0 (1,444.0)
80.5 204.0 (189.0) 285.0 (270.0) 850.0 (834.0) 848.0 1,698.0 (1,679.0)
72.0 n.a. (172.0) n.a. (244.0) n.a. (855.0) 890.0 (1,745.0)
n.a. n.a. (n.a.) 318.0 (306.0) 1,028.0 (1,016.0) 1,318.0 2,346.0 (2,334.0)

58.0 169.2 (158.0) 227.0 (216.0) 627.0 (625.0) 498.0 1,125.0 (1,113.0)
77.6 214.7 (204.0) 292.0 (282.0) 738.0 (728.0) 720.0 1,458.0 (1,448.0)
81.3 238.8 (224.0) 320.0 (305.0) 791.0 (775.0) 931.0 1,721.0 (1,706.0)
85.0 228.0 (216.0) 313.0 (301.0) 866.0 (854.0) 1,480.0 2,346.0 (2,334.0)
-3.0 5.4 2.4 22.5 -20.4 n.a.
-9.2 -31.1 -40.3 69.4 -66.8 n.a.
-2.2 -34.2 -36.4 92.6 -90.0 n.a.
n.a. n.a. 5.0 162.0 -162.0d n.a.

weight, including foreign material and excess moisture, to clean weight (Shend 1993). The projections shown in parentheses have
this study.
net imports of the importing developing countries.
after the abnormally low level of 1988 due to the drought.

36 Population and Food in the Early Twenty-First Century

Table 3.5-Outcome for all cereals (including rice in milled form), 93 developing countries

3.5A--All regions combined

Production Consumption Net Imports Self-Sufficiency
(million tons) (percent)
1961-63 350 357 16 (18) 98
1969-71 480 492 17 (20) 98
1979-81 650 709 59 (67) 92
1988-90 845 918 80 (90) 92
2010 1,314 1,462 148(162) 90
a Numbers in parentheses are the net imports of all developing countries, including those not in the group of 93, some of which are
with net imports of 6 million tons in 1988-89.

3.5B-By region
Per Capita Total Feed Total Use Production Net Balancea
(kilograms) (million tons)
Sub-Saharan Africa
1969-71 115 31 1,224 37 36 -3
1979-81 113 40 n.a. 48 41 -8
1988-90 114 54 n.a. 63 54 -8
2010 121 110 n.a. 128 109 -19
Near East/North Africa
1969-71 183 33 10 54 46 -6 (-7)
1979-81 203 47 19 80 58 -23 (-24)
1988-90 213 63 32 112 73 -38 (-39)
2010 210 103 64 190 119 -71 (-72)
East Asia, including China
1969-71 151 173 21 220 216 -7 (-9)
1979-81 181 252 49 334 316 -19 (-25)
1988-90 201 320 74 435 419 -20 (-27)
2010 206 424 176 657 635 -22 (-35)
South Asia
1969-71 148 106 1 119 116 -5
1979-81 153 136 2 154 148 -1
1988-90 156 172 2 196 200 -5
2010 163 271 4 302 292 -10
Latin America and the Caribbean
1969-71 119 33 22 63 66 +3
1979-81 128 46 38 94 87 -8
1988-90 129 56 45 113 99 -10
2010 139 87 79 134 159 -25 (-26)
Note: SSR is self-sufficiency ratio; n.a. means not available.
a Numbers in parentheses are the net imports of all developing countries, including those not in the group of 93, some of which are
with net imports of 6 million tons in 1988-89.

lion tons is used for seed and industrial nonfood uses or 330 million tons by the year 2010 (see Table 3.6). Why
represents waste (postharvest to retail), feed use should grow at this relatively high rate is ex-
For the reasons already discussed, the use of cereals for plained below in the discussion of the prospective devel-
direct food consumption may grow at only 1.9 percent per opments of livestock production and consumption. It
year while that for feed would grow nearly twice as fast should be noted, however, that this growth rate is below
(3.7 percent per year) and may more than double to some that of earlier periods, when feed use of grains started from

Chapter 3: The Outlook for World Food and Agriculture to Year 2010 37

Per Capita
Consumption Growth Rate
Production All Uses Direct Food Other Uses of Production
(kilograms per capital) (percent per year)
165 169 131 38 1960-70 4.0
186 190 145 45 1970-80 3.1
201 220 162 58 1980-90 2.8
216 235 170 65 1988-90
228 254 173 80 to 2010 2.1
sizeable importers though minor producers. By far, the major net cereal importer among the countries not in the group of 93 is Taiwan,

Growth Rates
SSR Period Demand Production
(percent) (percent)

97.4 1961-90 2.6 1.9
85.5 1970-90 2.9 2.1
86.4 1980-90 3.2 3.4
85.5 1988-90 to 2010 3.4 3.4

86.9 1961-90 4.0 2.4
72.6 1970-90 4.3 2.4
65.4 1980-90 3.9 2.9
62.7 1988-90 to 2010 2.6 2.3

98.2 1961-90 4.1 4.0
94.5 1970-90 3.7 3.6
96.2 1980-90 2.7 3.1
96.7 1988-90 to 2010 2.0 2.0

97.3 1961-90 2.8 3.0
96.0 1970-90 2.7 3.0
102.0 1980-90 2.8 3.0
96.3 1988-90 to 2010 2.1 1.8

104.9 1961-90 3.6 2.9
92.9 1970-90 3.2 2.4
87.6 1980-90 1.8 0.8
86.5 1988-90 to 2010 2.4 2.3

sizeable importers though minor producers. By far, the major net cereal importer among the countries not in the group of 93 is Taiwan,

a low base and reached a peak growth rate of 7.2 percent the concern that the use of cereals for feed diverts supplies
per year in the 1970s, before slowing down to 3.6 percent that would otherwise be accessible to the poor? Increased
per year in the 1980s. demand for feed has traditionally originated in the middle-
Still, feed use may account in the year 2010 for some and high-income countries, in which both feed and direct
22 percent of total use, up from 17 percent today, which food use of cereals have been increasing and the latter is
itself was up from 11 percent 20 years ago. How valid is near "satisfactory" levels, though not for all population

38 Population and Food in the Early Twenty-First Century

groups. By contrast, the need to increase per capital direct
food use of cereals is to be found predominantly in the
low-income countries, where feed use of cereals accounts
for only a tiny proportion of total availabilities.
Under these circumstances, the link between the two,
often spatially separate, categories of demand (of the
middle-income countries, for both food and feed; of the
poor countries, for food) and eventual diversion from the
latter to the former could occur, in the first instance,
through the operation of the world markets. But the con-
ditions for this to happen are rather stringent: either global
supplies are quasiconstant, in which case the additional
demand for feed would raise prices, with the result that
part of the demand of the poor for direct consumption
would be priced out of the market; or supplies can be
increased but only at prices that are above those that would
otherwise prevail. In a global context, the increasing use
of cereals for feed was most likely supported by additional
supplies that were forthcoming at prices not significantly
above those that would otherwise have prevailed. One of
the key factors behind the declining, or nonincreasing, real
prices in the presence of increasing demand has been the
lowering of production costs following the diffusion of
productivity-increasing technology. It is possible that this
technological progress was partly linked to the expansion
of demand of cereals by the feed sector.
Food-feed competition can be a significant factor in
diverting supplies away from the poor at the level of
individual countries or regions within countries. This
would be the case of countries that face stringent produc-
tion constraints and have little scope for increasing sup-
plies through imports. In such cases, an increase in the
demand for feed would raise prices rather than production
and reduce the direct food consumption of the poor.
However, in the low-income regions of South Asia and
Sub-Saharan Africa the use of cereals for feed remains a
minuscule proportion of total use. Moreover, the projec-
tions show that this situation would not change much in

the future, as nearly all these regions' increases in demand
for cereals would be for direct food purposes. Therefore,
there is a prima facie case that the food-feed competition
may not really be a significant factor in preventing pro-
gress toward raising the per capital direct consumption of
cereals in the many countries with such consumption still
at inadequate levels. If anything, the historical experience
of many middle-income developing countries demon-
strates that increases in the feed use of cereals occurred in
parallel with increases in the per capital consumption for
direct food purposes. The general underlying factor has
been the growth in per capital incomes in combination with
the growth of domestic cereals production and/or im-
provements in the import capacity. It is the absence of
these factors that stand in the way of increasing per capital
direct consumption in many low-income countries rather
than the fact that feed use may be increasing in the middle-
and high-income countries.

The Livestock Sector

Developed Countries
Two major factors may influence outcomes in the live-
stock sector in developed countries. In the former CPEs
of Eastern Europe, the reported per capital consumption of
meat had reached levels that were not much below those
encountered in other developed countries with much
higher incomes.13 With the price reforms and falls in
purchasing power, the per capital consumption of both
meat and milk is going through a process of decline that
may not bottom out for some time. For the longer term,
the somewhat optimistic assumptions used here are (1)
that per capital consumption of the region as a whole may
just revert to near the reported prereform levels of about
70 kilograms (or whatever that level was), but with more
poultry meat and less beef (this is an optimistic assump-
tion because even by 2010 the per capital incomes of the
region are likely to be well below those of other countries

Table 3.6--Feed use of cereals and livestock production
Growth Rates
Feed Use of Cereals Livestock Productiona
1969-71 1988-90 2010 1970-90 1988-90 to 2010 1970-90 1988-90 to 2010
(million tons) (percent per year)
Developed countries 386 482 527 1.2 0.4 1.8 0.7
Former CPEsb 115 189 172 2.2 -0.4 2.1 0.3
Others 271 294 355 0.7 0.9 1.7 0.8
93 developing countries 55 154 327 5.5 3.7 5.3 3.7
(39) (71) (198) (5.4) (3.2) (4.3) (3.2)
aGrowth rates computed from a special, feed-specific volume index of livestock output, taking into account the fact that in the developing
countries much of the beef, mutton, and milk production comes from nongrainfed animals, while pigmeat, poultrymeat, and eggs come
predominantly from grainfed production systems (for relevant data, see Tyers and Anderson 1992, p. 386). The feed-intensity weighted
livestock production index is computed as follows: 0.3 (Beef+ Mutton) + 0.1 (Milk) + 1.0 (Pig + Poultry + Eggs).
bData and projections are before conversion of the former Soviet Union's cereals production data from bunker to clean weight (see
Table 3.4). Revised numbers are as follows: 1969-71, 111 kilograms; 1988-90,182 kilograms; and 2010, 166 kilograms.
CFigures in parentheses are for 92 countries, excluding China.

Chapter 3: The Outlook for World Food andAgriculture to Year 2010

with meat consumption at that level); (2) that also produc-
tion may revert to prereform levels and indeed be slightly
higher by the year 2010; and (3) that this projected pro-
duction can be achieved with a smaller input of cereals feed
(Table 3.6).
For the other developed countries as a whole, the only
likely significant change is that further growth in the per
capital consumption of meat will come from the poultry
sector, while per capital consumption of pork could stabi-
lize (after the rapid expansion of the past, which took place
mainly in Western Europe) and so would that of beef and
dairy products. Cereals will probably be more competi-
tively priced vis-A-vis cereal substitutes and growth of
feed use could resume at about the same rate as that of
livestock production.
The overall conclusion from these summary projections
for the livestock sector in the developed countries is that
their demand of cereals for feed would grow only modestly
(perhaps approximately 45 million tons in 1988-90 to 2010
compared with approximately 100 tons from 1969-71 to
1990) and that it will be the expansion of feed demand in
the developing countries (approximately 170 million tons)
that will dominate future developments.
Developing Countries
Structural change in the food consumption of the devel-
oping countries toward more livestock products will con-
tinue (Table 3.3). However, these countries' per capital
consumption of such products will still be well below
those of the high-income countries 20 years from now.
For some developing countries, the consumption of live-
stock products may not advance even in the longer-term
future to the stage where it would match the consumption
levels of the developed countries-for various reasons,
including those of ecology and culture. Prospective devel-
opments in China, such as a slowdown in the growth of
per capital meat consumption, decisively influence the
total outcome because this country accounts for 40 percent
of total meat consumption of the developing countries.14
For the developing countries, excluding China, the growth
rate of per capital demand for meat may be maintained at
1.1 percent per year, while the growth of their total de-
mand and production would be somewhat less than in the
past because of the lower growth of population (Table 3.7).
Concerning the milk sector, there may be a drastic
slowdown in the rate of increase in consumption. In contrast
to the prospective developments for the meat sector, the
slowdown is generalized and all regions may be affected.
The reasons are to be found in a combination of production
constraints and reduced availabilities of exports at highly
subsidized prices (including food aid) from the main export-
ing developed countries. The trade picture is likely to be
dominated by the reduced scope for subsidized production
and exports from the main exporting developed countries,
a policy trend likely to hold in the future, with consequent

upward pressure in world market prices; and the limited
import capacity, already apparent in the 1980s, of those
developing countries (mainly in the Near East/North Africa
region), which fueled the growth of consumption and im-
ports in the 1970s.
The feed use of cereals in the developing countries is
becoming an increasingly important component of total
cereals use. The bulk of the additional production of meat
in most countries will come from the pig and poultry
sectors. Unlike the case of their ruminant animal sector,
substantial increases in pig and poultry depend heavily on
the expansion of the grainfed production systems. This
general pattern would translate into the feed use of cereals
growing faster than the volume of livestock output. In
parallel, however, gains in feed use productivity would
tend to attenuate this pattern. Moreover, this pattern is less
pronounced or entirely absent in those countries (both
developed and developing) with relatively high shares of
the cattle-sheep sector in the total growth of livestock
production and ample grassland and other nongrain feed
By and large, and unless there are special country-spe-
cific circumstances, the use of cereal feed is likely to grow
at roughly the same rate as that of total livestock output
(Table 3.6). Finally, it must be noted that cereal feed is
only one component, and in many developing countries a
minor component, of the (largely unknown) total balance
between feed resources and livestock production.

Other Basic Food Crops in the
Developing Countries

Roots, Tubers, and Plantains
This category of basic foods is composed of a variety of
products, the main ones being cassava, sweet potatoes,
potatoes, yams, taro, and plantains (hereafter referred to
as "starchy foods"). For some of these products, much of
the production and apparent consumption are at the sub-
sistence level. Moreover, they are produced and con-
sumed predominantly in countries with poor statistical
services and are subject to above-average rates of waste
and postharvest losses. For these reasons, the data on
production and apparent consumption are subject to mar-
gins of error that are thought to be much larger than those
for other food products.15
Two dimensions of the starchy foods economy seem
to dominate all others when it comes to viewing develop-
ments in this sector: ecology and per capital incomes. In
other words, these products are the mainstay of diets of
poor people in countries characterized by ecological con-
ditions (mainly the forest zones in the humid tropics) that
make these foods the predominant subsistence crops. It is
precisely in the countries that combine both these charac-
teristics and also have a low per capital level of aggregate
food supplies that very high levels of apparent per capital

40 Population and Food in the Early Twenty-First Century

Table 3.7-Meat and milk production and consumption, 93 developing countries
Growth Rates
1969-71 1979-81 1988-90 2010 1970-90 1988-90 to 2010
(percent per year)
Total (million tons) 27.0 42.2 64.0 143.0 4.8 3.9
Per capital (kilograms) 10.5 13.0 16.4 25.0 2.6 2.0
Production (million tons) 28.5 42.6 64.8 143.0 4.6 3.8
By species (million tons)
Cattle 12.1 14.6 18.6 32.0 2.2 2.7
Sheep/Goat 3.0 3.6 4.9 10.0 2.8 3.1
Pig 9.7 16.8 28.3 64.0 6.1 4.0
Poultry 3.7 7.6 12.9 37.0 7.0 5.1
Total (million tons) 84.7 122.8 164.0 273.4 3.7 2.5
Per capital (kilograms) 27.4 32.1 35.9 41.7 1.7 0.7
Production (million tons) 78.0 107.3 147.3 247.6 3.5 2.5
Net Trade (million tons) -6.8 -16.2 -16.2 -25.8 5.8 2.2
Notes: Meat is measured by carcass weight, excluding offals. Production is that of indigenous meat, with live animal exports counted
as domestic meat production and live animal imports as meat imports, in their carcass weight equivalent. For milk, all data and
projections are for all milk and dairy products in liquid whole milk equivalents.

food consumption of roots are encountered. There are 15
countries in Sub-Saharan Africa (group "Africa-high" in
Table 3.8), each with more than 200 kilograms of apparent
per capital consumption (in fresh product weight), but with
some having levels of 350 to 400 kilograms (Ghana,
Gabon, Congo, and Uganda) and others having more than
400 kilograms (Zaire). Together these countries account
for 55 percent of Sub-Saharan Africa's population, and
they depend on roots for some 40 percent of their food
supplies. Smaller country groups in both Sub-Saharan
Africa and the Latin America and the Caribbean regions
have a medium-level dependence on roots for their total
food supplies (100 to 200 kilograms per capital while all
the other developing countries have consumption levels
well below 100 kilograms (40 kilograms on the average),
including some with very low levels (typically 5 kilo-
grams) in Sub-Saharan Africa itself, for example, Mali
and Somalia.
The high dependence of diets on these products in
some of the poorest countries with high incidence of
undernutrition makes this group of crops of prime impor-
tance in any assessment of future prospects in nutrition
and of policies to improve it. It is also a reminder that the
analysis of issues of food security, poverty, and under-
nutrition should not be unduly limited to cereals.
The prospects are that per capital food consumption of
starchy roots will continue to decline, partly as a result of
continuing urbanization (though its effect will be mainly
limited to cassava and yams). However, because of the
little growth in per capital incomes and the difficulties in
increasing supplies of cereals, no radical structural change
of diets away from the high dependence on these products
is to be expected in the Sub-Saharan African countries. In
the end, countries accounting for more than 50 percent of

Sub-Saharan Africa's population would still depend on
starchy foods for some 36 percent of their total food
supplies (calories). This assumes that production can
grow at rates somewhat above those of the last 20 years,
an outcome that is considered feasible. Overall, however,
the production of this group of crops in the developing
countries as a whole is expected to grow at a rate below
that of the population, a pattern well established in the
historical period. This reflects essentially the change in
diets resulting from urbanization, which tends to favor
foods (for examples, cereals) that are often cheaper and
more convenient for preparation and consumption in ur-
ban environments compared with the rural ones. Given
the potential for increasing production, policies to stimu-
late consumption could be important for improving nutri-
tion. Such policies may include research to develop new
root-based food products, such as composite flours, noo-
dles, chips, and dehydrated products.16

Pulses and Oilcrops
Pulses, often included in the statistics together with cere-
als in the foodgrains group, include a variety of products
(such as beans, peas, chickpeas, and lentils; soybeans and
groundnuts are included in the oilseeds, not in pulses) that
form an important component of the diet, particularly
those of the low-income population groups, in many
developing countries (FAO 1981). Average per capital
apparent food consumption of pulses in the developing
countries as a whole is some 7.5 kilograms and has been
falling (it was 12 kilograms in the early 1960s). There is,
however, still a large number of developing countries with
relatively high levels (10 to 20 kilograms), such as Burk-
ina Faso, Uganda, India, Nicaragua, and Brazil; and a few
with very high ones, such as Rwanda and Burundi. With

Chapter 3: The Outlook for World Food and Agriculture to Year 2010 41

Table 3.8-Roots, tubers, and plantains (starchy foods) in the developing countries
Growth with HighlMedium
Consumption of Starchy Foods
Other 93
Africa Latin America Developing Developing
Higha Mediumb Mediumb Countries Countries

Per capital supplies, all food
(calories per day)
1988-90 2,160 2,190 2,280 2,500 2,470
Percent from starchy foods 40 21 13 4 6
Percent from cereals 34 49 36 63 61
2010 2,230 2,240 2,540 2,790 2,730
Percent from starchy foods 36 20 11 3 6
Percent from cereals 35 47 38 58 56
Production growth rates, starchy
foods (percent per year)
1970-90 2.4 1.8 1.7 1.0 1.4
1988-90 to 2010 2.8 3.3 1.9 0.9 1.6
aMore than 200 kilograms per capital in 1988-90 (fresh product weight): Benin, Ghana, C6te d'lvoire, Nigeria, Togo, Angola, Central
African Republic, Congo, Gabon, Zaire, Burundi, Mozambique, Rwanda, Tanzania, and Uganda.
b100-200 kilograms per capital: Africa (Guinea, Liberia, Cameroon, Madagascar, Namibia); Latin America and the Caribbean (Dominican
Republic, Haiti, Bolivia, Colombia, Ecuador, Paraguay, and Peru).

a few exceptions, such as Brazil and Mexico, these coun-
tries have low per capital total food supplies, particularly
of livestock products. Pulses are, therefore, an important
source of protein, particularly in countries with low live-
stock and fish consumption levels. They provide 45 per-
cent of total protein availabilities in Rwanda and Burundi,
25 percent in Uganda, 20 percent in Haiti, and 14 percent
in India.
The trend toward decline in per capital consumption of
pulses in the developing countries as a whole was halted
in the 1980s, particularly if China is excluded from the
total. For the future, it is estimated that per capital food
consumption may remain at about the present levels (9.0
to 9.5 kilograms on the average, or 7.5 to 8.0 kilograms if
China is included). In practice, this is saying that the
experience of the 1980s rather than that of the longer
historical period may be representative of likely develop-
ments in the next 20 years, as shown in Table 3.9.
In 1988-90 the world produced oilcrops that corre-
sponded to some 71 million tons of oil equivalent. Actual
production of vegetable oil was, however, smaller (60
million tons), because some oilseeds are also used for
purposes other than oil extraction, such as for direct food,
feed, and seed. The production of the developing countries
was 45 million tons and their net exports were 3.7 million
tons. Their per capital direct food supplies of oilcrops (oil
equivalent) were 8.2 kilograms in 1988-90 compared
with 4.7 kilograms in 1969-71; this may grow to some 11
kilograms by 2010.
In the last 20 years, the sector experienced above-
average growth in production (4.0 percent per year) as
well as radical structural change in the shares of total
production of the individual oil crops and regions. The
shares ofoilpalm products, soybeans, sunflower seed, and
rapeseed rose rapidly in relation to other oilcrops; and the

share of traditional oilcrops-such as groundnuts, coco-
nuts, sesame, and cottonseed-declined.
The demand of the developing countries for oilseed
proteins, thejoint product of oilcrops, is expected to grow
faster than that for vegetable oils, reflecting the rapid
growth of the livestock sector, as discussed above. These
developments are likely to lead to a situation whereby the
developing countries will continue to have large and
growing net exports of vegetable oils (oils plus the oil
equivalent of net oilseed exports), but a declining net
surplus of oilseed proteins (from oilmeals and oilmeal
equivalent of oilseeds).

Possible Developments in the
Agricultural Trade Balance of the
Developing Countries
Length limitations preclude presentation of the projec-
tions and discussion of important policy issues pertaining
to other major agricultural products of the developing
countries, particularly the export ones (sugar, tropical
beverages, cotton, tobacco, and rubber). However, it is
relevant to the theme of this volume to highlight the
prospect that world market developments point in the
direction of the developing countries turning from net
agricultural exporters to net importers. Historical devel-
opments in the aggregate agricultural trade values of the
developing countries are shown in Table 3.10. The quan-
tity indices of total imports and exports indicate that,
between 1961-63 and 1988-90, imports in real terms
increased by 284 percent and exports by only 146 percent.
The net export balance in real terms was, in 1988-90, less
than one-third of its level 30 years ago. These trends
reversed in the 1980s because of the abrupt slowdown in
the growth of the agricultural imports in the crisis decade
of the 1980s.

42 Population and Food in the Early Twenty-First Century

Table 3.9-Growth rate of prod
demand for pulses an
the developing countries
1970- 1970- 1
80 90
Demand (all uses)
Developing countries 0.4
Developing countries, 0.4
excluding China
Oilcrops (oil equivalent)
Developing countries 5.3
Developing countries 0.2
Developing countries, 0.2
excluding China
Oilcrops (oil equivalent)
Developing countries 4.8

The import requirements and export availabilities of
the major crop and livestock products projected in the
FAO (1993a) study indicate that it may not be long before
the developing countries as a whole turn from net agricul-
tural exporters to net importers. This would happen if
values of exports and imports of the individual commodi-
ties were to change pari passu with the changes in vol-
umes. Quantitative and qualitative elements indicate that
the net exports of the agricultural commodities for which
the developing countries have positive net trade balances
($32 billion17 in 1988-90) are likely to grow at slower
rates than the net imports of commodities for which they
have negative balances ($27 billion in 1988-90), thus
leading to a turnaround from positive to negative of the
countries' aggregate trade balance on agricultural ac-
Some of the increases in net imports of the developing
countries, particularly of raw materials, are likely to be
more than compensated for by growth in the positive net
balance of trade in manufactures based on these products.
In addition, part of the increased net imports of cereals and
livestock products reflect developments in those develop-
ing countries that could finance these imports with export
earnings from other sectors. However, these prospective
developments are likely to be a heavy burden for those
countries that must continue to finance growing net food
imports from export earnings that are unlikely to be forth-
coming at the required rate from other export sectors.

Land-Yield Combinations Underlying
the Production Projections
It is a common perception that there is little scope for
increasing production through expansion of cultivated
land. This is true for most developing countries, but less
so for those countries that still have large tracts of land

auction and with crop production potential. The developing countries
J oilcrops in as a whole, excluding China,18 are estimated to have in
s crop production approximately 760 million hectares (of
980- 1988-90 which some 120 million hectares are irrigated) out of their
90 to 2010 total land, with some 2.5 billion hectares having rainfed
nt) crop production potential of varying qualities. This leaves
a balance of such land not yet used for growing crops of
1.8 billion hectares. However, most of this balance (more
1.8 2.1 than 90 percent) is in the two land-abundant regions of
Sub-Saharan Africa, and Latin America and the Carib-
bean, and is very unevenly distributed among countries,
3.4 for example, 27 percent in Brazil, 9 percent in Zaire, and
36 percent in 13 other countries. Naturally, not all of this
unused land is, or should be considered to be, a reserve
1.6 2.0 for agricultural expansion for well-known reasons (for
example, environmental, infrastructural, economic,
health). However, these numbers serve to put in proper
3.2 perspective the notion of increasing land scarcities as
constraints to agricultural growth.

Table 3.10-Value of agricultural trade, all devel-
oping countries, 1961-63 to 1988-40
1961-63 1969-71 1979-81 1988-90
(billion dollars)
Current prices
Exports 13.8 18.7 69.8 87.5
Imports 7.2 10.7 66.0 82.5
Net balance 6.6 8.0 3.8 5.0
Implied values at
1988-90 prices
Exports 35.6 44.9 64.4 87.5
Imports 21.5 27.4 63.0 82.5
Net balance 14.1 17.5 1.4 5.0

Note: These data include all crop and livestock products, includ-
ing both primary and most processed products, but not
manufactures based on agricultural raw materials, such as
textiles or leather goods. Commodities with net export
surplus in 1988-90 include coffee, tea, cocoa, oils and
oilseeds, sugar, spices, citrus, bananas, other fruits and
vegetables, cassava and roots, vegetable fibers (other than
cotton), tobacco, rubber; net importables were cereals,
meat, eggs, dairy products, animal fats, pulses, beverages
(mostly alcoholic), hider and skins, wool, and cotton.

Just as it happened in the past, land expansion can
reasonably be expected to continue to play a role in
agricultural growth in the countries in which a combina-
tion of potential and need so dictate. Thus, land in crop
production use may expand by some 90 million hectares
(12 percent) between 1988-90 and 2010, of which 70
million hectares would be in the land-abundant regions.
Because of the limited space in this volume, the relevant
data and projections are shown in Table 3.11 without
further comment for lack of space (full explanations are
given in Alexandratos 1995, Chapter 4). The projected
rates of expansion are below those recorded in the past,
particularly for irrigation.

Chapter 3: The Outlook for World Food andAgriculture to Year 2010 43

Table 3.11-Land and irrigation in crop production use, developing countries
1988-90 2010 2010
Total Cereal Total Cereals Aable Land Balancea
(million hectares)
92 countries, excluding China
Harvested 598 331 723 389 n.a.
Irrigated 136 96 180 117 n.a.
Rainfed 462 235 543 272 n.a.
Arable 757 n.a. 850 n.a. 1,725
Irrigated 123 n.a. 146 n.a. n.a.
Rainfed 634 n.a. 704 n.a. n.a.
Land-abundant regions
Harvested 233 107 303 145 n.a.
Irrigated 18 9 24 12 n.a.
Rainfed 215 98 279 133 n.a.
Arable 402 n.a. 471 n.a. 1,596
Irrigated 20 n.a. 25 n.a. n.a.
Rainfed 382 n.a. 446 n.a. n.a.
Land-scarce regions
Harvested 365 224 420 224 n.a.
Irrigated 118 87 156 105 n.a.
Rainfed 247 137 264 139 n.a.
Arable 355 n.a. 379 n.a. 129
Irrigated 103 n.a. 121 n.a. n.a.
Rainfed 252 n.a. 258 n.a. n.a.
Memo Item: China
Harvested 141.0 91.0 n.a. n.a. n.a.
Arable 95.7 n.a. n.a. n.a. n.a.
Irrigated 45.6 n.a. n.a. n.a. n.a.
Notes: n.a. means not available.
aLand with rainfed crop production potential of varying qualities that is not in crop production use.
bSub-Saharan Africa, and Latin America and the Caribbean.
CAsia and the Near East/North Africa.

As noted in the methodology, the projections of both
areas and yields were carried out for each crop, agro-
ecological land class (five rainfed and one irrigated), and
country. The yield projections for major cereals and land
classes of the developing countries, excluding China, are
shown in Table 3.12. As shown in the table, the growth
rates of average yields can be expected to be below those
of the historical period, particularly for the two major
cereals, wheat and rice. The reasons for the projected
slowdown in yield growth are well known: there is less
scope than in the past to achieve upwards shifts in yield
ceilings from research breakthroughs in modern varieties
and from rapid expansion of area, particularly of irrigated
land under such varieties. The result is that increases in
average yields would come less from quantum jumps in
the yields of the countries and areas with already high
yields today and more from raising those of the countries,
particularly the large producers, at the middle and lower
ranges in the yield distribution. This is why the yield
projections of Table 3.12 imply some narrowing of the
intercountry yield gap for each land class. However, this
does not imply that yield gaps can be eliminated, nor even
narrowed more than marginally.

Does this pattern conform to the historical experience?
This issue cannot be investigated for individual land
classes because no relevant historical data exist. Such data
exist only for average yields (over all land classes) in each
country. They show that the gap between the countries
with the highest and the lowest yields (simple averages of
the top and bottom deciles of the countries, ranked by
yield level) had widened between 1969-71 and 1988-90.
This occurred mainly through a process whereby the
yields of the countries in the top decile in 1969-71 rose
by more than those of the countries in the bottom decile.
The projections for average yields (over all land classes)
imply that the future may be unlike the past and the yield
gap may become somewhat narrower because the scope
of yield growth in the top decile countries of 1988-90 is
more limited than it was 20 years ago.
This pattern is illustrated in Table 3.13 with data of
individual countries for wheat and rice. For wheat, in
1988-90 the countries in the top decile of the distribution
had yields that were nearly twice as high as the countries
in the top decile of 1969-71. In contrast, yields grew much
less in the countries at the bottom decile. These develop-
ments are even more pronounced for rice.

44 Population and Food in the Early Twenty-First Century

The dependence of the future growth of aggregate
production of the developing countries on a modest nar-
rowing of the intercountry yield gap (as measured here,
the difference in the simple average of yields in the top
and bottom deciles of countries) should not, however, be
exaggerated. This is because the countries at the two ends
of the distribution (top and bottom deciles) account for a
relatively small part of the total production of the crop
examined. This is true even when, as has been done for
Tables 3.12 and 3.13, countries with less than 50,000
hectares under the crop (and, for Table 3.12, also under
the given land class) are excluded from the analysis. In
practice, the validity of the projections of total production
of the developing countries depends crucially on such
exclusions in order to ensure correct projections for yield
growth in the countries that account for the bulk of the
area under each crop.
For this purpose, Table 3.13 also shows the relevant
historical data and projections of the 10 percent of coun-
tries with the largest areas (top decile of countries ranked
by area under the crop). As shown, (1) these countries
have yields that are less than one-half those in the coun-
tries with the highest yields; (2) for wheat, their (simple)
average yield is projected to grow by 43 percent, which is
below the 62 percent increase of the last 20 years; (3) for
rice, the corresponding percentages are 47 percent and 50
percent; and (4) even with these increases these countries,
whose performance carries a large weight in the total,

would still have in 2010 (simple) average yields around
one-halfthose projected for the countries in the top decile.
Thus, although the gap may narrow, particularly for rice,
this would be the result of the more limited scope for yield
growth of the countries in the top decile, not because the
large countries with middle yields are projected to have
higher growth than in the past.
The preceding, rather lengthy, discussion was consid-
ered necessary to provide the reader with sufficient mate-
rial for thinking about the issue of the potential for further
growth in yields to underpin the growth of production.19
No attempt is made here to translate these projected yields
into concrete proposals for agricultural research (magni-
tude, modalities, priorities). No doubt, further growth in
yields, even at the more modest rates projected for the
future compared with the past, will not come about unless
the research effort continues unabated. The effects of
research on production growth may manifest themselves
in different ways: more impact through the results of
adaptive and maintenance research, leading to evolution-
ary rather than revolutionary growth, and less through
achievement of quantum jumps in yield ceilings.

In this chapter, an attempt was made to present a picture
of selected aspects of the world food and agriculture
situation as it may develop to the year 2010, not as it ought
to be from a normative standpoint. Inevitably, the conclu-

Table 3.12-Cereal yields in major agro-ecological land classes and intercountry differences, for
developing countries, excluding China
of Yields Growth Rates of
from Given Weighted Average Country Rangeb Average Yields
Land Class, 1988-90
1988-90 1988-90 2010 1969-71 1988-90 2010 1970-90 to 2010
(tons per hectare) (percent per year)
Rice (paddy), all land classes 100 2.8 3.8 0.9 -4.6 0.9 6.6 1.5 -7.2 2.3 1.5
Irrigated 58 3.7 5.2 1.7 7.2 3.4 8.0
Fluvisols and gleysols 24 2.4 3.1 1.0 3.6 1.4 4.0
Wheat, all land classes 100 1.9 2.7 0.5 2.7 0.8 5.1 1.2 6.4 2.8 1.6
Irrigated 60 2.4 3.3 1.1 -5.4 1.9 6.7
Rainfed, subhumid 21 1.7 2.1 0.9 2.9 1.2-4.1
Maize, all land classes 100 1.8 2.5 0.6-3.1 0.6-4.9 1.1 -6.0 1.8 1.5
Irrigated 24 3.8 4.6 1,6 -7.9 2.2 -8.4
Rainfed, subhumid 49 1.8 2.6 0.6 3.7 1.2 4.1
Rainfed, humid 17 1.3 1.7 0.4 -2.8 0.8 -3.6
Millet, all land classes 100 0.7 0.8 0.4-1.3 0.3 1.4 0.6-1.7 1.0 1.0
Rainfed, dry semi-arid 18 0.4 0.5 0.1-0.6 0.3-0.8
Rainfed, subhumid 27 0.9 1.1 0.6-1.8 0.7-2.2
Sorghum, all land classes 100 1.0 1.2 0.3-2.8 0.4-3.4 0.6-3.7 1.5 1.1
Rainfed, dry semi-arid 11 0.5 0.6 0.3 -1.0 0.4 -1.2
Rainfed, subhumid 32 1.3 1.7 0.6 -3.5 0.9 -3.9
yields of countries with at least 50,000 hectares in the crop, land class, and year shown.
bSimple averages of the yields of the bottom 10 percent and top 10 percent of the countries ranked by yield level. The same countries
are not always in the top or bottom deciles in each year.

Chapter 3: The Outlookfor World Food and Agriculture to Year 2010

sion is that progress will continue to be made, but that it
will be slow and that food security problems will persist.
Some countries and whole regions will likely fail to make
progress, and there is a risk of further deterioration in the
already precarious food security situation of some of
them. To have concluded on a more pessimistic tone
would have been equivalent to ignoring the rapid eco-
nomic progress taking place in Asia and the rise in living
standards of a significant proportion of world population.
To have adopted a more optimistic tone would have meant
ignoring the persistence of economic difficulties in many
countries, particularly in Sub-Saharan Africa.
The progressive slowdown in growth and the prospect
that it will continue is a complex phenomenon that cannot
be wholly, or even mainly, ascribed to the onset of pro-
duction "fatigue" on the side of global natural resources
and technology, making it difficult for world production
growth to keep up with that of effective demand. The
slowdown likely reflects a combination of positive and
negative developments. The positive ones include the
slowdown in the growth of world population and in the
demand of that part of the world population with largely
satisfied food consumption needs. The negative ones in-
clude the absence of poverty-reducing economic growth,
partly related to agricultural development failures, in
many low-income countries with precarious food security
situations. It is in these countries that agricultural devel-
opment failures imply grave threats to their food security.
To the extent that such failures reflect worsening situ-
ations of natural resources, environmental degradation,
and inadequate attention to technology development and
diffusion in low-income countries with high dependence
on agriculture, it is entirely proper to speak of local
resources and technology factors becoming increasingly
binding constraints to improving food security. This
would still be the case even if world production growth
were much less, or not at all, subject to such constraints.
It is in these situations that efforts to relax these con-
straints and improve the policy environment for agricul-
ture can have a high payoff in terms of food security and
can lead to a better future outcome than sketched out in
this chapter. Such an approach would recognize the cru-
cial role of local agriculture: generating employment and
income for the rural people, stimulating the nonfarm rural
sector and the overall economy, and increasing food

1. This chapter is a revised and shortened version of a
paper presented at the IFPRI roundtable from which
this volume was developed. It summarizes the rele-
vant parts of the FAO study Agriculture: Towards
2010, which was prepared in the period 1992 to
mid-1993 by a team led by the author, with all the
technical units of FAO and a number of external

collaborators contributing in varying degrees. The
projections of the study present the possible future
evolution as it may develop rather than as it ought to
be from a normative standpoint. The study is an
official FAO document that was on the agenda of the
17th Session of the FAO Conference, 6-25 Novem-
ber 1993. Therefore, this chapter represents FAO
views rather than those of the author in his personal
capacity. The source for all data in this chapter is the
FAO Agrostat database unless otherwise indicated.
Dates given in this chapter are in the following for-
mat: 1988-90 denotes the annual average of the three
years 1988, 1989, 1990; 1988-90 to 2010 denotes the
period covered in this study.
2. The FAO study covers 127 countries individually-
34 developed and 93 developing. The latter account
for 98.5 percent of the population of the developing
countries. The country list and a more complete
explanation of the methodology are given in Alexan-
dratos 1995.
3. "Tons" throughout this volume refers to metric tons.
4. The World Bank states that "Africa must set its target
for long-term agricultural growth no lower than 4
percent per year" (World Bank 1989).
5. The use of an exogenous GDP growth rate to study
and project agricultural variables in countries with
high dependence on agriculture violates the widely
held notion that in such countries the performance of
the overall economy and of agriculture are closely
interlinked. More research to fill this gap is sorely
needed, although making total GDP endogenous in a
study covering a large number of countries individu-
ally is a daunting task.
6. This statement is based on Mitchell and Ingco (1993),
who forecast further declines in world cereal prices
as compatible with global projections for the cereals
sector that are broadly similar to those of the FAO
study. The most recent round of FAO's medium-
term commodity projections (made with the FAO
World Food Model and used as an input to this study)
indicate for 2000 some weakening in wheat prices
and some firming of prices for rice and coarse grains
(FAO 1994, 34).
7. The projections were carried out for the individual
coarse grains because the production prospects could
only be evaluated for each one of these commodities
in each country separately, not for an aggregate group
of coarse grains.
8. Holding the standard deviation constant assumes
some reduction in relative inequality of distribution
(the coefficient of variation) when the mean rises.
This is justified by physiological considerations (that
is, a person needs a minimum food intake for survival
and there is a upper physiological limit to how much
food a person may consume) that dictate that the

46 Population and Food in the Early Twenty-First Century

Table 3.13-Intercountry gaps in average yields for wheat and rice, developing countries, excluding
Country 1969-71 Country 1988-90 2010
(tons per hectare)
Number of countries 32 33 34
Top decile Mexico 2.92 Zimbabwe 5.73
Egypt 2.74 Egypt 5.00
South Korea 2.31 Saudi Arabia 4.65

Average 2.65 Average 5.12 6.37
Bottom decile Algeria 0.61 Algeria 0.68
Myanmar 0.55 Bolivia 0.70
Libya 0.26 Libya 0.90

Average 0.47 Average 0.76 1.16
Decile of largest producers, Turkey 1.32 Turkey 2.02
by area India 1.23 India 2.12
Pakistan 1.11 Pakistan 1.81

Average 1.22 Average 1.98 2.84
Yield of top decile = 100 100.00 100.00 100.00
Bottom decile 18.00 15.00 18.00
Largest producers 46.00 39.00 45.00
Simple average, all countries 43.00 53.00 57.00
Note: Data and projections for countries with more than 50,000 hectares under wheat or rice in the year shown. Average yields are

scope for distributional inequalities is more limited
in the case of food intakes than in other "unbounded"
variables, such as income. Consequently, it may be
hypothesized that very low or relatively high levels
of average per capital food supplies will be associated
with more equal distribution of food intakes com-
pared with those that could prevail when the overall
average is the middle range. This is a useful hypothe-
sis for looking at the distributional issue when noth-
ing is known about prospective developments in the
other determining variables of the distribution of
food intakes. Thus, it can be expected that the distri-
bution of food intakes will tend to become more equal
in those developing countries in which the average
per capital food supplies will continue to edge upward
toward the 3,000-calorie level.
9. "Near East" is defined here as the region encompass-
ing Afghanistan, Iran, Iraq, Jordan, Lebanon, Saudi
Arabia, Syria, Turkey, and Yemen, which are in-
cluded in the group of 93 countries studied, as well
as Kuwait and the United Arab Emirates. North Af-
rica is defined as the region encompassing Algeria,
Egypt, Libya, Morocco, and Tunisia.
10. A number of studies consider that the former Soviet
Union may become a significant net exporter of
cereals (Mitchell and Ingco 1993; Johnson 1993).
11. These are net imports of all the developing countries,
after deduction of projected net exports of the export-

ing developing countries of some 30 million tons (up
from 17 million tons in 1988-90 and 14 million tons
in 1969-71). The net imports of the importing devel-
oping countries are projected at some 190 million
tons by the year 2010 (up from 106 million in 1988-
90 and 34 million tons in 1969-71).
12. Concern is often expressed at the burgeoning food
deficits of "Africa." In practice most of the increases
in net cereal imports originated in North Africa (Mo-
rocco, Algeria, Tunisia, Libya, Egypt) rather than in
the Sub-Saharan region. Thus, in 1988-90, net im-
ports in North Africa were 19.4 million tons (up from
2.7 million tons in 1969-71); in Sub-Saharan Africa,
8.1 million tons (2.7 million tons in 1969-71). The
estimated import requirements in no way refer to any
notion of Africa's "food deficit," a term often used
in a normative sense and variously estimated as the
difference between domestic production and total
consumption requirements to meet some normative
target of per capital consumption.
13. It is probable that the prereform data of the former
CPEs overstated the apparent per capital consumption
of meat.
14. There are a number of reasons for the projected
slowdown in China. In the first place, rapid growth
in per capital meat consumption in the historical pe-
riod started from very low levels of 30 years ago (4.5
kilograms in 1961-63) and received new impetus

Chapter 3: The Outlookfor World Food and Agriculture to Year 2010 47

Country 1969-71 Country 1988-90 2010
(tons per hectare)
44 47 50
Egypt 5.27 Egypt 6.65
South Korea 4.63 South Korea 6.41
North Korea 4.25 North Korea 8.11
Peru 4.14 Peru 5.16
Average 4.57 Average 6.58 7.25
Ghana 1.00 Liberia 1.14
Tanzania 1.00 Mozambique 0.87
Guinea 0.89 Guinea 0.83
Zaire 0.76 Zaire 0.91
Average 0.91 Average 0.94 1.55
Indonesia 2.35 Indonesia 4.22
Thailand 1.93 Thailand 2.00
Bangladesh 1.68 Bangladesh 2.57
India 1.67 India 2.63
Average 1.91 Average 2.86 4.20
100.00 100.00 100.00
20.00 14.00 21.00
42.00 43.00 58.00
47.00 45.00 53.00
simple country averages, not weighted by area.

after the reforms of the late 1970s, to reach 23.5
kilograms in 1988-90. The growth rate of per capital
consumption was 5.7 percent in 1970-90. If growth
were to continue as per trend, it would reach 75
kilograms in the year 2010. This is nearly the Euro-
pean level and unlikely for a country at the level of
development that China may reach in the next 20
years. Moreover, a continuation of trends of meat
production at the rate of the last 20 years would
probably put an intolerable strain on the cereals and
oilseeds sectors, with feed demand translating into
large import requirements. Feed now takes 55 to 60
million tons of cereals in China, or 17 to 18 percent
of total use. The per capital meat consumption pro-
jected in this study would still be a respectable 49
kilograms per capital in year 2010. It would translate
into a production growth rate of 4.6 percent per year.
15. In particular, the inclusion of plantains in this group
creates problems. It is meant to capture the similari-
ties with roots and tubers in food consumption and
nutrition in several countries in the humid tropics.
However, distinguishing between the role of plan-
tains in food consumption and nutrition from that of
bananas properly consumed as fruit is not always
easy. For the purpose of food consumption analysis,
plantains may be considered to be bananas that are
picked green and cooked before eating (FAO 1979
and 1990). Several countries, particularly in Central

America, are reported as having fairly high levels of
per capital consumption of bananas (40 to 50 kilo-
grams). The picture is further complicated by the
consumption of both bananas and plantains after
fermentation in the form of alcoholic beverages.
16. No discussion of the starchy-roots sector of the de-
veloping countries would be complete without a brief
examination of the future of cassava exports, mainly
to Europe for animal feed. Such exports, almost
entirely from Thailand and to a lesser extent Indone-
sia, experienced phenomenal growth in the past, from
5 million tons in 1969-71 to 24 million tons in
1988-90 in net terms (fresh product equivalent of
dried cassava). Imported cassava and oilmeals cap-
tured an increasing share of the European feed mar-
ket, substituting for cereals that were
uncompetitively priced by the domestic support poli-
cies. The situation may change in the future follow-
ing policy reforms in Europe, making further growth
difficult or even reducing exports from the develop-
ing countries.
17. All dollars in this volume are U.S. dollars.
18. The rest of this section refers to the developing coun-
tries, excluding China because in that country data
for present cropping patterns and yields by agro-
ecological zone are not available, nor are those for
still unused land with agricultural potential. More-
over, there is a strong possibility that the existing data

48 Population and Food in the Early Twenty-First Century

understate cultivated area and overstate yields (China
1990, 315). If so, the scope for production increases
through yield growth may be higher than commonly
thought (Johnson 1993). This is one of the reasons
why this study does not project large net food import
requirements for this country.
19. A comparison of the FAO projections made in 1985-
86 for the period 1982-84 to 2000 (Alexandratos
1988) with actual developments to 1992 of cereal
areas and yields in the developing countries shows
that the latter have been tracking fairly closely the
projected paths (for more details, see Alexandratos
1995, Chapter 4).

Alexandratos, N., ed. 1988. World agriculture: Toward
2000, an FAO study. New York: Belhaven Press and
New York University Press.
1995. World agriculture: Toward 2010, an
FAO study. Chichester, UK: John Wiley and Sons.
Brown, L. 1994. Facing food insecurity. In State of the
World 1994, ed. Lester Brown, A Worldwatch Institute
Report. New York: W. W. Norton and Co.
China, State Statistical Bureau of the People's Republic
of China. 1990. China Statistical Yearbook 1990. Bei-
jing: China Statistical Information and Consultancy
Service Center.
FAO (Food and Agriculture Organization of the United
Nations). 1979. Human nutrition in tropical Africa.
1981. Legumes in human nutrition. Rome.
1990. Roots, tubers, plantains, and bananas
in human nutrition. Rome: FAO.
1992. Worldfoodsupplies andprevalence of
chronic undernutrition in developing regions as as-
sessed in 1992. Document ESS/MISC/1/92. Rome.

1993a. Agriculture: Towards 2010. Docu-
ment C93/24. Rome.
1993b. World food model: Model specifica-
tion. Document No. ESC/M/93/1. FAO, Rome.
1994. Medium-term prospects for agricul-
tural commodities: Projections to year 2000. Rome.
FAO/WHO (Food and Agriculture Organization of the
United Nations and World Health Organization). 1992.
Nutrition and development: A global assessment.
Rome: FAO.
1994. Does China have a grain problem?
China Economic Review 4 (1): 1-14.
Johnson, D. Gale. 1993. Trade effects of dismantling the
socialized agriculture of the former Soviet Union.
Comparative Economic Studies 35 (4): 21-31.
Mitchell, D., and M. Ingco. 1993. The worldfood outlook.
Washington, D.C.: World Bank.
Platteau, J. P. 1990. The food crisis in Africa: A compara-
tive structural analysis. In The Political Economy of
Hunger, vol. 2, ed. Jen Dreze and Amartya Sen, 279-
387. Oxford, U.K.: Clarendon.
Shend, J. 1993. Agricultural statistics ofthe former USSR
republics and the Baltic States. Statistical Bulletin 863.
Washington, D.C.: USDA/ERS.
Tyers, R., and K. Anderson. 1992. Disarray in worldfood
markets: A quantitative assessment. Cambridge, U.K.:
Cambridge University Press.
United Nations. 1993. World population prospects: The
1992 revision. New York.
World Bank. 1989. Sub-Saharan Africa, from crisis to
sustainable growth: A long-term perspective study.
Washington, D.C.
1993. Global economic prospects and the de-
veloping countries, 1993. Washington, D.C.

Chapter 4

By Donald O. Mitchell and Merlinda D. Ingco

Meeting the world's food requirements during the twenty-
first century should be increasingly easy if past trends in
production and consumption continue. World food pro-
duction has more than kept pace with population growth,
and rates of growth of food production show few signs of
slowing. Grain yields continue to increase along a trend
that extends back to the 1960s. During the 1980s, world
grain production increased by 2.1 percent per year, while
population grew by 1.7 percent. The World Bank's index
of food commodity prices fell by 78 percent from 1950 to
1992 (World Bank 1992). The decline in food prices,
along with rising real incomes, has brought a better-fed
world. The relevant question is, Will these trends con-
The task of this chapter is to present a forecast of
regional and global grain demand and supply under the
assumption that past trends will continue. The projections
are based on a recently completed report, The WorldFood
Outlook (Mitchell and Ingco 1993). This report examines
historical trends in food production and consumption by
major countries and world regions. Specific forecasts of
wheat, rice, and coarse-grain production, consumption,
trade, and prices are made to 2010 on the basis of a global
econometric model.1 However, before these forecasts are
presented, the specific question raised by the organizers
of the IFPRI roundtable on population and food are briefly
addressed. The question is whether the 1980s represent a
period of significant change in historical trends in produc-
tion, and in particular whether 1984 represents a turning
point of significance in the growth of grain production or
The 1980s do represent a significant turning point, but
not in the growth of grain yields or production. The 1980s
were unique for different reasons. First, the rate of growth
of food consumption slowed from that of the 1960s and
1970s. Secondly, the collapse of the former CPEs should
have lasting consequences for the world food markets.
These changes seem to overshadow any changes in the
growth of world grain yields or production that occurred.
The slowing of the growth of food consumption is
reflected by the fact that the world's average per capital

grain consumption for all uses has not increased apprecia-
bly since 1978. In developing countries, the average has
not increased since 1984. Part of the slower growth during
the 1980s can be credited to slower economic growth in
many developing countries; however, consumption
growth also slowed in many Asian countries that did not
have slower income growth during that period. In China,
for example, real GDP grew substantially faster during the
1980s than during the 1970s, yet consumption of grain
grew by an average of 2.3 percent per year during the
1980s, compared with 5.2 percent per year during the
1970s. If this trend continues, then it marks the end of the
period of rapid growth of per capital grain consumption. It
also marks a shift away from a focus on increasing quan-
tities of food consumed to that of improving diet quality.
The second change of significance during the 1980s
was in the former CPEs. These countries were major grain
importers during the 1970s and 1980s, and their imports
contributed directly to the world food crisis of the 1970s.
While the political and economic transition in these coun-
tries is still unfolding, major changes in food consumption
clearly can be expected. These changes will almost cer-
tainly cause consumption and imports to decline. Produc-
tion will also likely begin to recover from recent declines
as privatization and modernization occur. Foreign ex-
change constraints will also encourage policy changes in
agriculture as imports are curtailed and exports encour-
Food consumption levels in the former CPEs have been
very high relative to other countries with similar income
levels and are expected to decline as pricing reforms and
economic adjustments are undertaken. Per capital grain
consumption for all uses in these countries was nearly
double that of Western European consumers-partly be-
cause of inefficient use and waste. For example, in 1988,
the level of grain consumption reached 783 kilograms per
capital in the former Soviet Union and 834 kilograms per
capital in Eastern Europe compared to 426 kilograms per
capital in the EC-10 countries. With consumption declines
and production reforms, the region is expected to eventu-
ally become a net exporter.

50 Population and Food in the Early Twenty-First Century

The production potential of this region is considered
by most experts to be great. Political and economic uncer-
tainty has caused grain production to stagnate in the region
since about 1980, while agricultural productivity has in-
creased steadily in other regions. However, reforms that
allow greater private ownership of land, smaller produc-
tion units, and importation of the best technologies would
likely stimulate production sharply. Allowing farmers to
receive world market prices for output and giving farmers
the opportunity to export would also stimulate greater
efficiency at all levels of production.
World grain yields continue to grow along the same
trend that has prevailed since about 1960. World grain
area harvested has declined since the peak reached in
1981. However, this seems to have had little effect on the
rate of growth of production. The decline in area harvested
occurred in two groups of countries. The major grain
exporting countries reduced area harvested by about 10
million hectares, and the former CPEs reduced area har-
vested by about 19 million hectares. Together these two
groups account for all of the decline, but the reasons for
the decline differ in each case. The major exporting coun-
tries reduced production because of declining prices, stag-
nant exports, and accumulating stocks, which led to
government policies that diverted land from production.
The former CPEs reduced area planted primarily because
of changes to their economic and political systems that
reduced the emphasis on extending grain production into
marginal producing areas. Area planted to grain in all
developing countries remained constant over this period,
with increases in some regions and declines in others.

Forecast to 2010
The main task of this chapter is to present a forecast of the
grain supply-and-demand balances for major countries,
regions, and the world to 2010. The forecast is taken from
The World Food Outlook (Mitchell and Ingco 1993) and,
as mentioned above, is based on an econometric model of
the world grain markets. Assumptions for the period to
2010 are made for population growth, GDP growth, con-
sumer prices, and exchange rates for each country or
region. Population assumptions are based on the United
Nations medium-variant population projection of 1990
(United Nations 1991). GDP assumptions are based on a
variety of sources and generally project moderate growth
in the developed countries (2.5-3.0 percent per year), slow
growth in Eastern Europe and the former Soviet Union
(1-2 percent per year), and rapid growth for the develop-
ing countries (4-6 percent per year). Assumptions for the
levels of world petroleum and fertilizer prices are also made
to 2010 on the basis of World Bank estimates (World
Bank 1992). Crop yields are endogenously projected by
the model; however, the projections closely approximate
the forecasts based on linear trends and should be viewed
as yield assumptions rather than yield forecasts.

The simulation results suggest that the outlook for the
world food situation is good, despite regional problems.
Most consumers can expect to have increased food sup-
plies and a greater variety of food at lower prices. The
prices of basic staples, such as grains, are projected to
continue to decline relative to other consumer prices and
relative to incomes. These changes suggest further im-
provements in diets of poor people, as has been the case
during the past 30 years. Such gains are based on higher
crop yields and modest increases in cropland area. Major
increases in resources devoted to crop production should
not be required unless crop yields succumb to yet-unfore-
seen problems.
The simulation results suggest only gradual increases
in world grain consumption, production, and trade. In the
absence of a major shock, such increases are all that is
required to continue the gains of recent years. The large
increases in consumption are probably past for most con-
sumers, despite large differences that remain in the levels
of consumption. For example, consumers in the highest-
income Asian countries-such as Japan, the Republic of
Korea, and Singapore-consume only about one-half as
much grain per capital as consumers in North America, but
it does not appear that this gap will be reduced quickly.
The most dramatic change now envisaged is for the
former CPEs to shift from grain importers to exporters.
These countries imported about 22 percent of world grain
trade in 1980, but imports are likely to decline sharply for
three reasons. First, in the short run, lower per capital
incomes, shortages of foreign exchange, and limits on
credit will depress imports. Second, in the longer run,
consumption levels will decline because of the reduction
of subsidies as food prices are brought up to international
levels. Third, in the opinion of most experts, the longer-
term restructuring of the agricultural sector will lead to
increased production. Estimates of future levels of con-
sumption, production, and trade cannot be very certain,
but the direction of change is clear.
The rate of growth of world grain consumption was
about halved during the past 20 years, from 3.1 percent
during the 1960s to 1.7 percent during the 1980s. On the
basis of the grain model simulations, world grain consump-
tion is projected to grow by about 1.4 percent per year from
1990 to 2010 compared with 1.7 percent per year during the
1980s. The slower growth will be due in large part to
declining consumption in the former CPEs as they restruc-
ture. Consumption in the developing countries is projected
to increase by 2.2 percent per year during 1990-2010
compared with 2.4 percent per year from 1980 to 1990
because of a slowing of the rate of population growth. Thus,
per capital consumption in the developing countries is pro-
jected to increase by about the same rate as during the 1980s.
The developed countries are expected to continue the slow
growth of consumption of the 1980s, increasing consump-
tion by 0.6 percent per year during 1990-2010.

Chapter 4: Global and Regional Food Demand and Supply Prospects 51

Regional problems will remain. Africa is projected to
face severe food problems because of rapid population
growth and slow income growth. Under such conditions,
farmers are unlikely to have adequate incentives to in-
crease production at even the rate of population growth.
The abundance of grain at low prices in the world market
will further weaken farm prices and Africa's ability to
produce. The situation could become critical, with mas-
sive food aid required. Some South Asian countries may
also face food problems, but these should be of much
smaller magnitude than in Africa. The high population
growth projected for Pakistan (2.9 percent per year) dur-
ing the 1990s, for example, will probably exceed the rate
of growth of grain production. However, rapid income
growth should allow imports to avert the kind of food
crises that Africa may face.
The former CPEs are projected to reduce sharply their
grain imports from the levels of the 1980s, and by the end
of the forecast period become significant exporters. Pro-
duction in the former CPEs is assumed to regain its former
growth path following the stagnation of the 1980s. This
growth will result from the restructuring of farms, mod-
ernization of farming methods, and improved technology.
The developed countries have historically exported
larger and larger quantities of grain to the other regions.
This trend was interrupted during the 1980s, with exports
declining and stocks increasing. The trend is expected to
resume during the 1990s, but the loss of the former CPEs
as a major export market will dampen growth. Production
in the developed countries should grow at slower rates
than during the 1960-80 period, while still meeting the
growing import demand of the developing countries and
maintaining moderate to high stock levels. Despite rapid
growth in exports to the developing countries, production
is projected to grow by only 1 percent per year, and even
this modest growth will lead to increases in stocks. The
area under grain production in the developed countries
will decline by an estimated 5.1 percent from 1990 to 2010
and by 10.8 percent relative to 1980.
Land devoted to world grain production is projected to
rise by 4.8 percent from 1990 to 2010. This increase is
expected to occur in the developing countries, while de-
creases will occur in cropland under grains in the devel-
oped countries and in the former CPEs. The largest
increase in land use is projected to occur in the Central
Africa region as very rapid population growth there puts
pressure on food demand.
Developing-country grain imports are projected to
continue to increase, as evidenced by the widening gap
between production and consumption. This gap began to
increase in the mid-1970s as incomes rose. Imports in-
creased even while production grew rapidly, as consumers
diversified their diets to include more wheat and meat.
Real grain prices are projected to decline throughout
the period as productivity increases exceed the demand

growth that would occur at constant prices. By 2010, real
wheat prices are projected to decline by 33 percent from
1992 levels, compared with a decline of 42 percent during
the 1980s. Rice and maize prices are projected to decline
by 31 percent and 21 percent, respectively, compared with
declines of 48 percent and 37 percent during the 1980s.
World grain production increases will need to slow
further if huge stock accumulations are to be avoided. The
adjustment will likely come in the developed countries.
World production will only need to grow by an estimated
1.4 percent per year during the period 1990-2010 to keep
pace with world demand. This would still allow consump-
tion in developing countries to increase by 2.2 percent per
year during 1990-2010 compared with 2.4 percent per
year during the 1980s.
The rapid growth of consumption for many developing
countries during the 1960s and 1970s seems to reflect a
catching-up period in which rising incomes allowed con-
sumers to increase grain consumption significantly. In
China, for example, consumption increased by 4.3 percent
per year during the 1960s, by 5.2 percent during the 1970s,
and by 2.3 percent during the 1980s (Table 4.1). The same
pattern is seen in Brazil, Indonesia, and North Africa. In
North Africa, consumption grew by 5.0 percent per year
during the 1970s, but by only 2.9 percent per year during
the 1980s. The slower growth during the 1980s reflected not
only the slower income growth in some countries but also
the attainment of an adequate diet by many consumers.
Japan experienced a similar pattern at an earlier period, as
grain consumption grew by 3.1 percent per year during the
1960s, by 2.4 percent per year during the 1970s, and by 0.6
percent per year during the 1980s. Some countries, such as
India and Pakistan, are currently going through a period of
rapid increases; however, most Asian, Latin American, and
North African countries are probably already past their
period of most rapid growth. If so, further declines in the
rate of consumption growth can be expected from them and
for the developing countries in aggregate.
A similar consumption pattern can be seen in the
former CPEs. Rapid growth occurred in Eastern Europe
during the 1960s and in the former Soviet Union during
the 1970s, followed by much slower growth during the
1980s. The slowdown in growth has been accelerated by
the economic collapse in these countries in recent years.
The high levels of per capital grain consumption reached
in these countries as compared with other countries with
similar cultural heritage and incomes makes increases
unlikely. For example, Eastern Europeans consumed 834
kilograms of grain per capital in 1988 compared with 427
kilograms in Western Europe. These high consumption
levels probably were due partly to high wastage rates as
well as to inefficient practices resulting from the highly
subsidized consumer prices.
Most developed countries have increased grain con-
sumption at slightly more than the rate of population

52 Population and Food in the Early Twenty-First Century

Table 4.1-Historical and projected grain production and consumption growth rates, 1960-2010
Production Consumption
1990- 2000- 1990- 2000-
1960-70 1970-80 1980-90 2000 2010 1960-70 1970-80 1980-90 2000 2010
(percent per year)
World 2.7 2.8 2.1 1.1 1.4 3.1 2.7 1.7 1.4 1.4
Developed countries 1.0 3.4 1.5 0.8 1.2 2.1 1.0 1.0 0.8 0.5
Australia 2.3 2.0 3.2 0.2 1.6 2.5 2.9 2.8 0.9 1.9
Canada 0.7 3.8 3.6 -0.3 1.3 3.0 0.7 1.5 0.4 0.6
EC-10 2.4 3.1 1.1 1.4 0.8 2.1 0.6 -0.5 0.9 0.3
Japan -2.2 -2.5 1.0 0.0 0.8 3.1 2.4 0.6 1.0 0.4
United States 0.3 3.7 1.5 0.6 1.4 1.7 0.4 2.5 0.4 0.5
Other 2.3 3.6 0.2 1.5 0.6 2.8 2.9 -1.0 1.8 0.6
Former CPEs 3.2 1.2 1.7 -0.5 1.0 3.7 2.5 0.8 -1.2 0.4
Eastern Europe 1.3 3.9 0.7 0.0 0.5 1.8 3.8 -0.3 -0.6 0.3
Former Soviet Union 3.8 0.2 2.1 -0.7 1.1 4.5 1.9 1.3 -1.4 0.4
Developing countries 4.1 3.1 2.7 1.9 1.7 3.6 3.9 2.4 2.4 2.1
Argentina 5.2 3.4 -2.8 2.0 1.3 0.8 0.4 -0.5 3.2 1.2
Brazil 4.2 4.8 0.5 3.3 1.9 3.9 5.1 1.3 2.5 2.2
Central Africa 3.8 2.5 1.5 4.2 2.5 4.1 3.2 2.3 3.4 2.7
China 5.9 3.7 4.1 1.6 1.6 4.3 5.2 2.3 2.4 1.8
East Asia 2.6 2.2 2.5 2.0 1.8 4.3 3.1 2.9 2.6 1.9
Egypt 2.9 1.1 4.5 1.9 1.1 4.4 4.4 3.5 2.0 2.0
India 3.1 2.1 3.3 1.4 1.8 2.8 1.9 3.1 1.8 2.1
Indonesia 2.4 4.2 4.1 1.5 1.7 2.1 4.7 3.4 2.3 2.0
Latin America 4.5 2.4 2.6 2.3 1.6 5.8 3.8 1.7 2.8 2.1
Mexico 6.8 2.4 2.5 1.7 1.8 5.9 5.5 2.0 2.5 3.1
Nigeria 1.3 0.5 -2.6 4.7 1.5 1.6 2.0 -3.5 5.2 2.2
North Africa 2.0 3.8 2.0 2.1 2.0 3.1 5.0 2.9 2.7 2.4
Pakistan 6.2 3.5 2.2 2.2 2.0 4.6 2.2 3.0 2.8 2.7
South Africa 5.1 4.6 -5.3 5.8 1.5 4.4 3.7 1.5 2.4 2.9
South Asia 1.3 2.5 2.1 0.2 1.4 1.2 2.6 2.4 2.0 1.9
Thailand 4.9 3.2 0.3 2.7 1.6 5.3 1.6 2.3 0.9 1.2

Source: Historical growth rates from 1960 to 1990 are based on USDA data, and projected growth rates from 1991 to 2010 are from
the simulation model.

growth during the past 20 year-about 1 percent per year
since 1970--and this trend is expected to continue.
Slower overall growth in grain consumption is projected
as population growth rates are projected to decline from
0.6 percent per year during the 1980s to 0.4 percent per
year during the 1990s, and to 0.3 percent per year during
the 2000-2010 period.
Trade has provided an increasingly large share of grain
consumption for the developing countries since the 1970s.
On balance, the developing countries imported about 3
percent of total grain consumed during the 1960-75 pe-
riod. This share steadily increased to 9 percent in 1990. It
is projected to continue increasing, but not at a faster rate.
By 2010, the developing countries are projected to import
14.8 percent of the total grain consumed.
The developed countries are expected to provide the
bulk of world grain exports, as has been the case during
recent decades. Projected net import levels and growth
rates are shown in Table 4.2. Net exports from the devel-
oped countries (shown as negative net imports) increase
from about 117 million tons in 1990 to 194 million tons
by 2010. This implies an average growth rate of 2.6
percent per year. Net imports by the developing countries
are projected to increase from 87 million tons in 1990 to

210 million tons by 2010. The former CPEs are projected
to shift from net grain imports of 27 million tons in 1990
to 16 million tons of net exports by 2010.
Large increases in grain net imports are projected to
occur in North Africa, Mexico, Indonesia, India, and
China, among other developing countries. The increases
primarily reflect rapid real GDP growth. Imports will be
mainly in the form of wheat and coarse grains, while rice
imports should remain low, as consumption is satisfied
largely by domestic production.
World average per capital grain consumption grew by
about 20 percent during the 18 years from 1960 to 1978,
but it has not increased appreciably since 1978. During
the 1960s, per capital consumption grew by 1.1 percent per
year, followed by 0.8 percent per year growth during the
1970s. Per capital consumption actually declined from
1980 to 1990 because of the increasing share of world
population in developing countries, which lowered the
The 1990s will most likely see declining world per
capital grain-consumption levels. In developing countries,
per capital grain consumption is projected to grow by 0.4
percent per year during the 1990-2010 period, compared
with 0.3 percent per year during the 1980s. Per capital

Chapter 4: Global and Regional Food Demand and Supply Prospects

Table 4.2-Historical and projected grain net imports, 1960-2010
Net Imports Growth Rates
1990- 2000-
1960 1970 1980 1990 2000 2010 1960-70 1970-80 1980-90 2000 2010
(million tons) (percent per year)
Developed countries -17.1 -24.8 -117.9 -117.4 -142.1 -194.4 3.7 16.9 0.0 1.9 3.2
Australia -7.9 -12.2 -12.4 -14.9 -13.6 -15.4 4.5 0.1 1.8 -0.9 1.3
Canada -10.1 -15.8 -19.5 -26.4 -28.5 -34.8 4.6 2.1 3.1 0.8 2.0
EC-10 21.3 22.3 -3.6 -23.9 -35.7 -45.3 0.5 n.a. 20.8 4.1 2.4
Japan 4.8 14.4 23.7 26.7 30.9 31.5 11.6 5.1 1.2 1.5 0.2
United States -29.3 -39.8 -114.5 -79.9 -96.9 -132.1 3.1 11.1 -3.5 2.0 3.1
Other 4.0 6.4 8.4 1.1 1.8 1.7 4.7 2.8 -18.5 5.1 -0.6
Former CPEs -0.7 1.8 48.4 26.5 2.3 -15.6 n.a. 39.0 -5.8 -21.5 n.a.
Eastern Europe 5.5 9.1 13.7 0.6 -1.6 -3.6 5.1 4.2 -27.3 n.a. 8.3
Former Soviet Union -6.2 -7.3 34.7 26.0 3.9 -12.1 1.6 n.a. -2.9 -17.2 n.a.
Developing countries 13.4 17.7 60.4 87.0 139.8 210.0 2.8 13.1 3.7 4.9 4.2
Argentina -3.6 -10.1 -18.3 -11.0 -12.9 -14.6 10.9 6.1 -4.9 1.5 1.3
Brazil 1.8 0.7 4.0 5.9 6.4 9.3 -9.3 19.4 4.1 0.7 3.8
Central Africa 0.4 1.5 5.2 7.0 7.3 10.6 14.0 13.0 3.0 0.4 3.9
China 2.2 2.3 14.0 3.7 11.3 21.6 0.7 19.5 -12.5 11.8 6.7
East Asia 1.0 7.7 15.3 21.9 31.4 39.0 23.1 7.1 3.7 3.7 2.2
Egypt 0.8 2.4 6.3 8.0 9.9 13.7 11.0 10.1 2.4 2.2 3.2
India 4.9 3.2 -0.8 -1.1 6.9 14.1 -4.2 n.a. 2.5 n.a. 7.1
Indonesia 1.1 1.1 3.5 2.1 5.7 7.6 0.7 12.1 5.2 9.3 4.3
Latin America 2.1 5.0 9.8 10.2 14.1 18.5 9.2 7.0 0.5 3.3 2.8
Mexico 0.1 -0.3 8.5 5.4 8.3 15.3 n.a. n.a. -4.5 4.4 6.4
Nigeria 0.1 0.4 2.1 0.6 1.9 3.3 13.8 18.1 -11.7 12.1 5.7
North Africa 4.1 7.6 19.6 30.8 43.6 59.0 6.2 10.0 4.6 3.6 3.1
Pakistan 1.0 0.9 -0.8 -0.3 2.1 4.5 -1.0 n.a. -11.1 n.a. 8.4
South Africa -1.6 -2.6 -4.6 1.0 -2.5 -0.4 5.1 5.6 n.a. n.a. -16.2
South Asia 1.1 1.3 2.1 2.9 9.2 12.8 0.9 5.2 3.3 12.3 3.4
Thailand -2.1 -3.3 -5.3 -5.1 -7.2 -9.1 4.7 5.0 -0.3 3.5 2.3
Sources: Historical data and growth rates from 1960 to 1990 are based on USDA data, and projected data and growth rates from 1991
to 2010 are from the simulation model.
Notes: n.a. is not available.
aThe sum of net imports for the developing countries and regions does not equal the total for the region because of imports that go to
unknown destinations. These imports are estimated and included in the regional totals, but are not included in the imports of a specific
country or region.

consumption levels are projected to decline in the former
CPEs for the reasons already noted, and this will largely
account for the decline in world per capital consumption.
Per capital consumption in the developed countries is
projected to increase by 0.2 percent to 0.3 percent per year,
consistent with trends of the past 20 years.
Regional differences in per capital grain consumption
reflect many factors, including economic, demographic,
and cultural differences; and these differences are largely
expected to remain. Asian consumers are expected to
maintain lower per capital consumption levels than con-
sumers in the Americas or Europe despite Asia's more
rapid growth in income. This reflects historical dietary
preferences as well as economic factors. Japan, for exam-
ple, has the highest per capital grain-consumption level
among Asian consumers but the lowest among developed
countries. Its level was only 58 percent of the average for
all developed countries in 1990. Further, the level of
consumption had not grown appreciably since 1980 de-
spite rapid income growth and rising consumer spending.
The differences between many Asian countries, including
Japan, is now small.

In the Asia region, Japan, which has the highest per
capital grain-consumption level among the Asian coun-
tries examined, consumed 306 kilograms in 1990 (Table
4.3). Countries that had much lower incomes, such as
China, consumed 298 kilograms per capital; and countries
in the East Asia region consumed 262 kilograms. Indone-
sia and Thailand consumed 204 and 211 kilograms per
capital, respectively, in 1990. These levels are projected to
increase gradually, with Japan increasing per capital con-
sumption to 331 kilograms per capital by 2010, China
increasing to 371 kilograms per capital to surpass Japan,
and the East Asia region increasing average consumption
to 297 kilograms per capital. Indonesia is projected to
increase per capital consumption to 231 kilograms, and
Thailand is projected to decrease consumption as it has
done since the 1970s, as consumers' diets in these coun-
tries diversify and grains become a smaller share of con-
India, Pakistan, and other South Asian countries have
low per capital grain-consumption levels compared with
East Asian consumers with similar incomes. This may be
due to the largely vegetarian diets of these countries

54 Population and Food in the Early Twenty-First Century

Table 4.3-Historical and projected per capital grain-consumption levels, 1960-2010
Amount of Consumption Growth Rate of Consumption
1990- 2000-
1960 1970 1980 1990 2000 2010 1960-70 1970-80 1980-90 2000 2010
(kilograms per capital) (percent per year)
World 275 307 332 331 322 325 1.1 0.8 0.0 -0.3 0.1
Developed countries 484 539 549 574 593 604 1.1 0.2 0.5 0.3 0.2
Australia 385 406 457 526 513 565 0.5 1.2 1.4 -0.2 1.0
Canada 912 1,024 972 1,028 991 997 1.2 -0.5 0.6 -0.4 0.1
EC-10 385 439 446 415 445 460 1.3 0.2 -0.7 0.7 0.3
Japan 219 268 303 306 325 331 2.1 1.2 0.1 0.6 0.2
United States 767 800 751 879 859 856 0.4 -0.6 1.6 -0.2 0.0
Other 379 456 563 484 551 561 1.9 2.1 -1.5 1.3 0.2
Former CPEs 541 700 822 829 695 686 2.6 1.6 0.1 -1.7 -0.1
Eastern Europe 555 622 849 788 716 720 1.1 3.2 -0.7 -1.0 0.1
Former Soviet Union 535 734 810 845 687 673 3.2 1.0 0.4 -2.0 -0.2
Developing countries 171 193 225 233 244 255 1.2 1.6 0.3 0.4 0.4
Argentina 453 423 373 311 380 388 -0.7 -1.2 -1.8 2.0 0.2
Brazil 204 227 294 272 292 313 1.1 2.6 -0.8 0.7 0.7
Central Africa 108 127 131 121 124 119 1.6 0.3 -0.8 0.2 -0.4
China 157 195 270 298 333 371 2.2 3.3 1.0 1.1 1.1
East Asia 197 228 245 262 283 297 1.5 0.7 0.7 0.8 0.5
Egypt 224 271 332 367 374 386 1.9 2.0 1.0 0.2 0.3
India 168 176 169 185 180 188 0.5 -0.4 0.9 -0.2 0.5
Indonesia 146 143 180 204 213 231 -0.2 2.3 1.3 0.4 0.8
Latin America 108 145 168 162 178 187 3.1 1.4 -0.3 0.9 0.5
Mexico 195 248 319 311 324 375 2.5 2.5 -0.2 0.4 1.5
Nigeria 151 139 132 69 87 81 -0.9 -0.5 -6.3 2.4 -0.8
North Africa 275 281 342 346 353 367 0.2 2.0 0.1 0.2 0.4
Pakistan 164 195 178 176 174 179 1.7 -0.9 -0.1 -0.1 0.3
South Africa 257 306 351 325 335 370 1.8 1.4 -0.8 0.3 1.0
South Asia 214 189 189 187 183 184 -1.3 0.0 -0.1 -0.2 0.1
Thailand 180 223 200 211 202 202 2.2 -1.1 0.5 -0.4 0.0

Sources: Historical data and growth rates from 1960 to 1990 are based on USDA data, and projected data and growth rates from 1991
to 2010 are from the simulation model.

compared with an increasing level of meat consumption are projected to decline further. Nigeria consumed only
in the East Asian countries. The growth of per capital grain 69 kilograms of grain per capital in 1990 compared with
consumption in the South Asian countries has been slow 151 kilograms in 1960. Further decline is predicted
and is projected to remain slow in the future, largely because population growth, which currently ex-
Comparisons of per capital grain consumption across ceeds 3 percent per year, is projected to remain high until
other countries and regions show some interesting general 2010. The other countries of Sub-Saharan Africa, which
trends. Per capital consumption in Latin American coun- are included in the Central Africa region in the model,
tries shows the same clustering characteristic as in the have fared somewhat better, with per capital consumption
Asian countries. Argentina, Brazil, and Mexico had simi- of 121 kilograms per year. However, with the population
lar consumption levels in 1990-311 kilograms per capital in this region also growing at near 3 percent per year,
in Argentina, 272 kilograms per capital in Brazil, and 311 keeping per capital consumption from declining will be
kilograms per capital in Mexico. However, the other coun- difficult; and the simulation projects a decline to 119
tries of Latin America, at an average of 162 kilograms per kilograms per capital by 2010.
capital, had much lower consumption. This large gap The level of per capital grain consumption depends
suggests that the region should expect significant in- largely on incomes and prices; however, the preference of
creases in consumption to increase significantly as in- consumers for wheat, rice, or coarse grains depends on the
comes increase. cultural background and availability of specific grains as
The Sub-Saharan African countries have the lowest well as on economic factors. Asia, for example, is tradi-
levels of per capital grain consumption, and these levels tionally a rice-consuming region, while Latin American

Chapter 4: Global and Regional Food Demand and Supply Prospects

and African countries depend more on coarse grains-
such as maize, sorghum, and millet-as staples. Pakistan,
Egypt, and North Africa are traditionally wheat-consum-
ing countries. Table 4.4, shows the historical and pro-
jected shares of wheat, coarse grains, and rice of total grain
consumption for major regions and selected countries.
The share of coarse grains in total world grain con-
sumption has been declining since 1960, while wheat and
rice have been increasing. This trend is most pronounced
in developing countries, where grains are primarily used
for direct human consumption and where coarse grains are
an inferior good to rice and wheat in this use. In India, for
example, the share of coarse grains has declined from 32.0
percent in 1960 to 22.3 percent in 1990. Other countries,
such as Thailand, have increased their consumption of
coarse grains for livestock and poultry feed, but overall
the world's share of coarse grains continues to decline. At
some stage, this trend will be reversed as the increase in
coarse grains for feed increases fast enough to offset the
decline in direct human consumption. However, such a
trend is not yet clear in the world figures nor for develop-
ing countries as a group. In this study, therefore, coarse-
grain consumption is projected to continue to decline as a
share of total grains during the forecast period to 2010.
The share of coarse grains in the developed countries and
in the former CPEs has remained relatively constant since
1960, at about 72 percent in the developed countries and
52 percent in the former CPEs. This high share in total
consumption reflects the use of coarse grains for feed.
The share of wheat in total grain consumption has
increased since 1960 and it is projected to continue in-
creasing throughout the forecast period because of in-
creases in the developing countries. Wheat is displacing
coarse grains in direct human consumption in nearly all
regions, and it is also displacing rice in many Asian
countries. In China, wheat has increased from 22.9 per-
cent of total grains in 1960 to 32.0 percent in 1990, while
coarse grains and rice have both declined. A similar trend
is seen in many other Asian countries-in India, the wheat
share increased from 19.4 percent to 31.3 percent over the
same period and, in Indonesia, the share increased from
1.2 percent to 5.2 percent. In Africa and Latin America,
the wheat share is increasing, while the coarse grains share
is decreasing. The changes in Latin America are occurring
more slowly than in Asia, with Brazil, for example, in-
creasing its wheat consumption share from 15.1 percent
in 1960 to 18.4 percent in 1990. In Nigeria, the wheat
share increased from nil to 12.8 percent of total grains by
1980 but has since declined to 5.1 percent as economic
problems have affected consumption.
Rice consumption as a share of total consumption has
changed significantly since 1960 in many countries and,
for the most part, these changes are projected to continue.
The share of rice has declined substantially in many Asian
countries, while it has increased in many other countries

as consumers' demand increased variety in their diets. In
Asia, many countries had diets centered around rice as the
basic grain in the 1960s. Thailand, for example, consumed
99.1 percent of its grains as rice in 1960. This is slowly
changing, however. In 1990, rice accounted for 70.5 per-
cent of Thailand's total grain consumption. If these trends
continue as projected, the share of rice will decline further
and by 2010 it will account for about 57 percent of total
grains in Thailand. Japan underwent a similar change
beginning in the 1950s-the share of rice was 61 percent
in 1951, and by 1990 rice only accounted for 25 percent
of total grains consumption, while coarse grains ac-
counted for 58 percent, primarily as feed. In India and
Pakistan, the rice share is declining, while wheat is in-
creasing. Coarse-grain consumption is still declining in
these countries, as the feed use of coarse grains is only
about 10 percent of total consumption.
Rice consumption is increasing in most developed
countries and in the former CPEs, but the share is small,
except for Japan. In the United States, the share of rice
doubled from 0.7 percent to 1.4 percent from 1960 to 1990
as the popularity of Chinese food increased. Other devel-
oped countries show similar trends as diets diversify.
However, it is unlikely that rice will ever be a major staple
in these countries.
The trends in grain shares at the global, regional, and
country level show a consistent pattern of change as diets
diversify and consumers shift among grains for direct
consumption and to increased coarse-grain use for feed-
ing. These trends are carried forward in the forecast.
Several trends are important. First, the share of wheat
continues to increase as consumers in developing coun-
tries continue to consume a larger share of their total
grains as wheat. Secondly, the overall trends do not indi-
cate an increase in the share of coarse grains in the
developing countries, although the decline is projected to
end. This projection suggests that the increased livestock
feeding that is evident in many countries, such as Thai-
land, will be offset by declines in human consumption of
coarse grains. Overall, the world demand for grains will
continue to favor wheat, while the coarse-grain share of
total grains demand continues to decline. Per capital rice
consumption is projected to decline in the developing
countries during the forecast period, while per capital
wheat and coarse-grain consumption will continue to
increase. This suggests that rice has become an inferior
good in the developing countries.

The simulation results present a picture of slow growth in
world grain consumption and surplus production capac-
ity. This leads to declining real grain prices. Despite this
global picture, regional problems are expected to remain
in Sub-Saharan Africa and to a lesser degree in South

56 Population and Food in the Early Twenty-First Century

Table 4.4-Share of wheat, coarse grains, and rice of total grain consumption, 1960-2010
1960 1970 1980 1990 2000 2010

World 28.2 29.9 30.7 32.7 32.5 33.1
Developed countries 25.4 23.9 22.8 27.1 25.0 25.2
Australia 50.1 51.7 52.1 46.4 41.3 40.8
Canada 26.1 21.3 22.4 24.6 23.4 23.2
EC-10 40.6 37.3 36.3 47.1 46.6 46.0
Japan 20.3 18.5 17.2 16.2 16.4 16.1
United States 11.6 12.8 12.5 17.1 13.9 15.3
Other 40.2 32.3 25.4 30.9 26.1 25.0
Former CPEs 45.1 51.9 46.9 46.6 43.3 42.3
Eastern Europe 32.1 38.8 34.5 41.4 39.4 39.3
Former Soviet Union 51.4 56.7 52.4 48.6 44.8 42.2
Developing countries 22.5 23.6 28.4 30.4 33.0 34.2
Argentina 35.3 40.0 37.5 47.8 40.8 37.9
Brazil 15.1 16.9 18.5 18.4 19.7 20.6
Central Africa 8.2 11.0 10.6 14.5 12.2 12.6
China 22.9 20.3 28.7 32.0 34.5 35.8
East Asia 7.5 12.1 10.9 14.3 13.8 13.9
Egypt 42.8 48.5 54.2 55.5 56.0 56.3
India 19.4 22.8 29.6 31.3 38.4 41.2
Indonesia 1.2 2.4 5.2 5.2 8.1 10.3
Latin America 29.1 35.4 32.7 32.1 32.8 32.3
Mexico 16.9 16.0 15.6 16.2 19.7 20.7
Nigeria 0.1 4.4 12.8 5.1 10.6 14.5
North Africa 58.4 60.4 60.1 57.9 55.4 55.3
Pakistan 72.3 70.8 76.2 80.2 84.5 84.6
South Africa 18.8 18.6 20.1 20.0 24.0 22.9
South Asia 16.0 15.8 23.9 21.1 36.5 40.2
Thailand 0.1 0.8 2.0 3.4 3.9 4.9

World grain consumption is projected to grow by only
1.4 percent per year during the period to 2010. This
appears to present a relatively modest challenge for world
production, which grew by 2.1 percent per year during the
1980s and by about 2.7 percent per year during the 1960s
and 1970s. If yields were to grow at 2 percent per year
during the period to 2010, which is not unreasonable, an
additional 11 percent of world cropland would need to be
removed from production. Yields may actually grow
more rapidly than in the recent past because of the genetic
improvements that are becoming available through
biotechnology and genetic engineering. If this occurs,
pressure to reduce cropland used for grain production
would intensify and this pressure would most likely fall
on the exporting countries.
An aspect of the simulation results that is critical to this
conclusion is that world grain consumption will grow
slowly during the period to 2010. This is consistent with
the past 15 years and with the assumption of slower
population growth. Most of the developing countries went
through their period of rapid increase in per capital grain
consumption during the 1960s and 1970s. Since about
1980, their per capital grain consumption has been nearly
constant. The simulation results are for only modest
growth, not for a resumption of the growth rates of the
1960s or 1970s.

Further support to the emerging picture of surplus
capacity is the prospect for grain surpluses in Eastern
Europe and the former Soviet Union. If production in-
creases, as projected, because of improved price incen-
tives, imported technology, better machinery, and
improved storage and transportation facilities, this region
will become an exporter. This would place further pres-
sure on grain prices and cause an even larger reduction in
production capacity in the exporting countries.
The two most important issues for the world grain
market outlook are developments in the former CPEs and
in the developing countries. The former CPEs of Eastern
Europe and the former Soviet Union are expected to
become grain exporters of significance over the next two
decades because of a combination of lower consumption
and higher production. Their per capital grain-consump-
tion levels are nearly 50 percent higher than in the devel-
oped countries and will most likely decline as subsidized
domestic prices rise to international levels. Production has
been stagnant since 1980 and has great potential to in-
crease with the adoption of market incentives and modern
Developing-country imports are expected to more than
offset declines in imports by Eastern Europe and the
former Soviet Union. By 2010, developing-country grain
imports are projected to increase to 210 million tons,

Chapter 4: Global and Regional Food Demand and Supply Prospects 57

Coarse Grains Rice
1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010
53.0 51.4 50.4 47.3 47.0 46.0 18.8 18.7 18.8 20.0 20.5 20.9
70.0 72.2 73.9 69.7 71.9 71.4 4.6 3.8 3.3 3.2 3.1 3.4
49.0 46.8 46.9 51.7 56.7 57.2 0.8 1.4 1.0 1.9 2.0 2.0
73.6 78.5 77.1 74.8 76.0 76.2 0.3 0.3 0.4 0.5 0.6 0.6
58.6 62.0 62.9 51.9 52.5 53.0 0.8 0.7 0.7 1.0 0.9 1.0
21.9 39.7 54.3 58.4 60.0 59.3 57.8 41.8 28.5 25.4 23.5 24.6
87.7 86.4 86.3 81.5 84.5 82.7 0.7 0.8 1.2 1.4 1.6 1.6
58.6 66.6 73.6 67.8 72.7 73.8 1.2 1.1 1.0 1.3 1.1 1.2
54.6 47.5 52.0 52.7 55.7 57.5 0.3 0.6 1.1 0.6 1.0 1.1
67.2 60.6 65.1 58.2 60.0 60.1 0.7 0.6 0.4 0.4 0.5 0.6
48.5 42.7 46.3 50.7 54.0 56.5 0.0 0.6 1.4 0.7 1.2 1.3
37.4 37.8 36.6 34.3 34.4 34.5 40.0 38.6 35.0 35.2 32.6 31.3
63.7 58.5 61.6 50.7 58.0 60.9 1.0 1.5 0.9 1.5 1.2 1.2
61.2 62.2 64.0 63.2 60.0 58.3 23.8 20.9 17.5 18.3 20.3 21.1
78.9 75.7 75.6 71.3 75.4 75.2 12.9 13.3 13.8 14.2 12.4 12.3
32.8 34.4 32.9 29.5 29.2 29.3 44.4 45.3 38.4 38.5 36.3 36.3
17.8 17.9 27.8 30.9 33.4 34.2 74.7 70.0 61.3 54.8 52.8 51.9
43.5 37.7 35.2 35.9 35.4 35.2 13.6 13.7 10.7 8.6 8.5 8.6
32.0 34.2 24.5 22.3 17.5 15.6 48.5 43.0 45.9 46.5 44.1 43.1
18.0 15.1 15.0 14.5 15.4 15.4 80.8 82.5 79.7 80.3 76.5 74.3
54.2 50.5 49.4 51.1 51.8 53.0 16.7 14.1 17.9 16.9 15.5 14.7
80.1 82.0 82.8 82.9 78.6 77.7 2.9 2.0 1.6 1.7 1.7 1.6
95.9 92.5 79.6 84.6 79.4 74.4 3.1 3.1 7.6 10.3 10.0 11.1
38.0 35.5 35.5 37.4 40.4 40.7 3.6 4.1 4.4 4.7 4.1 4.0
14.7 12.1 10.3 9.1 7.1 5.6 13.0 17.1 13.5 10.6 8.4 9.8
80.0 80.6 78.6 75.0 73.5 74.5 1.1 0.8 1.3 3.0 2.5 2.6
11.8 10.5 8.2 8.1 5.9 5.3 72.2 73.7 67.9 70.9 57.6 54.5
0.0 2.9 12.8 26.1 29.4 37.6 99.1 96.3 85.1 70.5 66.7 57.4

compared with the current level of 87 million tons. The
increase will primarily go to countries with rapidly grow-
ing incomes and rising levels of grain consumption for
both direct human consumption and as livestock and
poultry feed. A combination of slower population growth
and more rapid income growth in the developing countries
would allow per capital consumption levels to increase,
sourced partly from larger grain imports and partly from
larger domestic production-in contrast to the 1960s,
which saw grain imports by the developing countries rise
because of food shortages and inadequate diets.
The most important conclusion to come from our
analysis is that the world food system has many options
to meet future demand. If demand grows more rapidly
than projected in this study, increased production can
come from the developed countries, which have surplus
capacity, or from the former CPEs, which have great
potential to increase yields. Production can be increased
by higher yields or expanded cropland-both of which are
possible. If the grain markets are in surplus at current low
world prices, higher prices would almost certainly stimu-
late production.
Not all countries will share in the surplus. Many coun-
tries in Sub-Saharan Africa will most likely face severe
food problems in the future. Rapid population growth,
often exceeding 3 percent per year, combined with poor

incentives for farmers has led to declining per capital
grain-consumption levels. Incomes are not rising fast
enough to allow market imports to offset the decline,
which means that food aid must be offered as a temporary
solution. The problems of Africa are much greater than
just food production, and the solutions require much more
than higher food production. The long-term solution can
only come from better domestic policies that result in
higher food production, higher economic growth, and
lower population growth.

1. The appendix to this chapter describes this model and
defines the regions discussed in this chapter.

Mitchell, D. 0., and M. D. Ingco. 1993. The world food
outlook. Washington, D.C.: World Bank.
United Nations. 1991. World population prospects 1990.
New York.
USDA (U.S. Department of Agriculture). 1992. World
grain situation and outlook. Foreign Agricultural
Service Report FG 12-92. Washington, D.C.
World Bank. 1992. Price prospects for major primary
commodities. Washington, D.C.

58 Population and Food in the Early Twenty-First Century

Appendix: The Econometric
Simulation Model
The World Grains Model used for the simulation analysis
in Chapter 4 is a nonspatial, partial-equilibrium, net-trade
model. The model was originally estimated during the
early 1980s for use in forecasting and policy analysis by
the Commodities Division of the World Bank. It has been
revised and updated several times--most recently in
1990, using data through 1988. The model is global in
scope, with 15 countries modeled individually and the
remaining countries grouped into nine regions. The com-
modities included in the model are wheat, rice, and coarse
grains (maize, oats, barley, sorghum, rye, millet, and
mixed grains). Individual models are estimated for each
commodity and country or region, with cross linkages
between commodities. Each country model takes the price
from the world linkage and returns the level of net trade.
Table 4.5 contains the list of exogenous and endogenous
variables used in the model, along with their sources and
definitions. The model has been estimated primarily with
ordinary least squares, using annual data from 1960 to
1988. The entire model contains approximately 500 equa-
tions, which are linear, and is solved using the Gauss-
Siedel iterative procedure. The countries and regions
defined in the model are shown in Table 4.6.

The supply model consists of three separate components
corresponding to the decision sequence of producers.
First, a total cropland equation is estimated for each
country or region to capture the quantity of cropland
harvested for the three model commodities; second, the
allocation of this land among the commodities is esti-
mated; and third, a yield equation is estimated for each
commodity. Production for each commodity is the prod-
uct of the area harvested and yield equations.

The cropland equation is
TCHAt = the harvested area in year t of the
model commodities;
DTCRVt-I = the weighted revenue per hectare
based on world prices of the crops
expressed in constant local currency
in year t-l;
TGESt-i = the sum of wheat, coarse grains, and
rice ending stocks in year t-1; and
TIME = a linear trend with 1960 = 1, 1962=
2, etc.
The harvested area of each crop is estimated as a
function of total cropland harvested and relative commod-
ity revenues. This specification treats the determination of
harvested area as a short-run allocation decision given that
the decision of how much cropland to plant has already
been made. The specified equation is
HAi,t = f(TCHAt, HAi,t-1, RVi,t-1, RVj,it-, TIME),
HAj,t = the harvested area of commodity i in year t,
TCHAt = the total cropland harvested in year t of
wheat, coarse grains, rice;
R Vi, = the revenue of commodity i in year t;
TIME = a linear trend with 1960 = 1, 1961 = 2,
Crop yields are assumed to be influenced by seed
quality and by inputs, such as fertilizers, land quality, and
weather. The estimated yield model necessarily simplifies
these factors into variables that can be used to represent
the various factors. The model is
YD, t = f(TIME, RPFi,t-I, HAit, HYVi,t),

Table 4.5-Agricultural model variables
Variable Source Definition
Exogenous variables
Population WB, FAO Million people per calendar year
Income-GDP WB, IFS Billion local currency per calendar year
Exchange rate to dollars WB, IFS Local currency to U.S. dollars per calendar year
Consumer price index WB, IFS Index, 1975 = 100, per calendar year
Endogenous variables
World crop prices USDA Dollars per ton, f.o.b. gulf, simple monthly average, per crop year
Production USDA Thousand tons per crop year
Harvested area USDA Thousand hectares per crop year
Crop yields USDA Tons per hectare per crop year
Consumption USDA Thousand tons per crop year
Net trade USDA Thousand tons per crop year
Beginning stocks USDA Thousand tons per crop year
Notes: Data sources are World Bank (WB); FAO; International Financial Statistics (IFS), published by the International Monetary Fund;
and the USDA. The model is estimated using annual data from 1960 to 1988.

Chapter 4: Global and Regional Food Demand and Supply Prospects

Table 4.6-Model Regions

Countries included

Developed countries

United States
Other developed countries

Former CPEs
Eastern Europe
Former Soviet Union
Developing countries
Central Africa

East Asia

Latin America and the Caribbean

Middle East/North Africa

South Asia

Belgium, France, Luxembourg, Netherlands, West Germany, Ireland, United Kingdom,
Denmark, Greece
United States
Austria, Finland, Iceland, Malta, Norway, Portugal, Spain, Sweden, Switzerland,
New Zealand

Albania, Bulgaria, Czechoslovakia, Hungary, Poland, Romania, Yugoslavia
Former Soviet Union

Angola, Benin, Botswana, Burundi, Cameroon, Central African Republic, Chad, Congo,
Djibouti, Ethiopia, Gabon, The Gambia, Ghana, Guinea, Equatorial Guinea, Guinea-Bissau,
Ivory Coast, Kenya, Lesotho, Liberia, Malagasy Republic, Malawi, Mali, Mauritania, Mauritius,
Mozambique, Namibia, Niger, Reunion, Rwanda, Senegal, Sierra Leone, Somalia, Sudan,
Swaziland, Tanzania, Togo, Uganda, Upper Volta, Zaire, Zambia, Zimbabwe
Brunei, Burma, Fiji Islands, Hong Kong, Kampuchea, Laos, Malaysia, Mongolia, North Korea,
Pacific Islands, Papua New Guinea, Philippines, Singapore, South Korea, Vietnam
Bahamas, Barbados, Belize, Bermuda, Bolivia, Other Caribbean Islands, Chile, Colombia,
Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador, French Guiana,
Guatemala, Guyana, Haiti, Honduras, Jamaica, Nicaragua, Panama, Paraguay, Peru,
Surinam, Trinidad and Tobago, Uruguay, Venezuela
Algeria, Bahrain, Cyprus, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman,
Qatar, Saudi Arabia, Syria, Tunisia, Turkey, United Arab Emirates, Republic of Yemen and
Democratic Republic of Yemen
Afghanistan, Bangladesh, Bhutan, Nepal, Sri Lanka

Note: Region and country names are as they were in 1990.

YDi,t = the yield per hectare in year t of com-
modity i;
TIME = a linear trend with 1960 = 1, 1961 = 2,
RPF,,t-I = the lagged ratio of the price of the com-
modity i to the price of fertilizer;
HAi,t = the area harvested of the commodity i
in year t; and
HYVi,t = the percent of total area planted to the
high-yielding varieties.

Import Demand
The demand for imports can be viewed in general as a
function of income, prices, and relevant demand shifters:
Mi,t = f(Yt, Pi,t, Pjt, Zt),
Mi, = per capital imports of commodity i,

Yt = real per capital income,
Pi,t = the real import price of commodity i,
Pj,t = the real price of a related commodity, and
Zt = a set of relevant demand shifters.
The demand for grain imports is treated as the inde-
pendent demand for each of the primary grain catego-
ries-wheat, rice, and coarse grains.

Export Supply
The export supply curves for the exporting countries is
Xi,t = DSt DDi,t STKi,t+i,
Xi,t = the exports of commodity i in year t,
DSi,t = the domestic supply (production plus
beginning of stocks) of commodity i in
year t, and
STKi,t+I = the beginning stocks of commodity i in
year t.


Renion Counties Include

60 Population and Food in the Early Twenty-First Century

In the importing countries and regions, consumption is
calculated as the residual with production, net imports,
and stocks being estimated. For the exporting countries,
consumption is estimated directly. In these countries,
consumption is disaggregated into two final uses: feed and
all other. Per capital consumption is then estimated as a
function of income and relevant prices:
CFi,t = f(Yt, Pit, Pj,t), and
COi,t = f(Yt, P,t, Pj,t),
CFi,t and COit = per capital food and all other
consumption of commodity, i

P,,t and Pj,t

= real per capital income; and
= the real prices of the commodity
i and a related commodity.

To close each country model, the final component that
must be estimated is the level of ending stocks (next year's
beginning stocks). The ratio of ending stocks to domestic
consumption in the previous year is estimated. The equa-
tion is

STKit = the level of ending stocks of commodity
i in year t, and
Pit = the real price of commodity i in year t.
Experience has shown that this type of model will
perform reasonably well but will not capture totally the
volatility that is frequently present in stock holdings.

A single world price, the U.S. export price in dollars, is
assumed to exist for each of the commodities. Each coun-
try's border price is then equal to the world price ex-
pressed in constant units of local currency:
PWi, = U.S. export price of commodity i in year t,
Pki, = border price of country k for commodity
i in year t, and
Pkit = PWi,t*XRk,tCPIk t,
XRk,t = the exchange rate of country i relative to
the U.S. dollar, and
CPkt = the consumer price index of country k in
year t.
The model is solved for the world price, which equates
net imports with net exports.

DD,=-f (Pu

Chapter 5

By Mercedita Agcaoili and Mark W. Rosegrant

The next two decades will pose complex problems for
national policymakers and international donors involved
in agricultural development. Future directions in food
supply and demand will be determined not only by the
relative behavior of prices, incomes, and other market
forces but also by the decisions and policies of farmers,
national governments, and international donors as they
induce and react to the possibly conflicting structural
changes in food production and consumption trends. On
the one hand, there is cause for concern over future food
availability-growth rates in yield per hectare for cereals
in developing countries and for the world as a whole have
declined since the early 1980s. On the other hand, daily
food energy per capital in the developing countries in
aggregate increased by more than 7 percent during the
1980s (Pinstrup-Andersen 1993). Real world food prices
declined sharply during the same period, indicating ample
total food supplies at effective levels of demand. Most
developing countries, with the notable exceptions of Sub-
Saharan Africa and parts of South Asia, have steadily
reduced the rate of population growth over the past two
decades. In Asia, which accounts for a large share of total
world food demand, the rate of growth in consumption of
food staples due to increasing incomes is also declining,
as diets diversify with rising per capital income and the
shift of population to urban areas (Bouis 1989, 1991).
What will be the outcome in terms of food availability
and prices over the next two decades resulting from these
seemingly contradictory trends-some indicating poten-
tial shortages as food production slows, others pointing to
continued decline in demand pressure for food and in-
creasing availability of food? What policies will be nec-
essary to ensure continued increases in food availability?
The main focus of this chapter is to provide an over-
view of the likely world food supply-and-demand situ-
ation to the year 2010, as well as to provide insights on
the likely effects on supply, demand, net trade, and prices
of developments in the social and economic structure of
countries (for example, population changes and income
growth rates). In this chapter, recent trends that will
influence future supply and demand for food are

described. These trends provide input for the growth rate
assumptions for key parameters employed in the simula-
tion procedure. The International Food Policy and Trade
Simulation (IFPTSIM) model, which is a modified and
extended version of a trade model developed by Keiji Oga
(Agcaoili, Oga, and Rosegrant 1993; Oga and Gehlar
1993), was used to generate the simulation results pre-
sented in this chapter. On the basis of these results, policy
issues are raised and possible strategies that would pro-
mote better food availability are presented.

Major Trends and Key Factors
Shaping the Future Food Situation

Several important changes have taken place in the past
that set the stage for discussing possible future develop-
ments in world food and agriculture. This section exam-
ines the growth patterns in production, consumption, and
trade in basic food staples and animal products.1
Production Trends in Basic
Food Commodities
The production of basic staples in the world during the period
1966-90 followed an upward trend, with an average annual
growth rate of 1.9 percent. The time profile of growth rates
for area, yield, and output, however, shows that, after the
initial acceleration of growth, a gradual deceleration began in
the mid-1970s that became quite pronounced in the 1980s
(Table 5.1). This was brought about by the reduction in yield
growth rates, primarily that of wheat and rice, which have
been the main sources of foodgrain production increases. It
can be noted that yield was the most important component of
growth throughout the period, accounting for 83 percent of
production in 1966-74, 93 percent in 1974-82, and 90 per-
cent in 1982-90. Yield is the main source of growth in the
final period in all regions except Sub-Saharan Africa, where
production growth was sustained by area expansion. The
strong area effect in Sub-Saharan Africa came from the large
portion of foodgrain production in other cereals, and roots
and tubers-which grow relatively well in more adverse
rainfed areas, where opportunities for expansion still exist.

62 Population and Food in the Early Twenty-First Century

Table 5.1-Average annual growth rates of food production, by region
Region 1966-90 1966-74 1974-82 1982-90


Latin America

Sub-Saharan Africa

Developing countries

Developed countries




























Source: FAO/IFPRI database.
Notes: n.a. means not available. Regional definitions follow those of FAO, except Japan is excluded from the developed countries and
is included in Asia. Developing countries include primarily the Middle East/North Africa.

The general slowdown in productivity growth from the
early 1980s resulted from a number of factors. One factor
is declining world prices, particularly of wheat and rice,
which caused a direct shift of land to more profitable
cropping alternatives. The shift into more diversified
cropping, while an appropriate farmer response to chang-
ing incentives, puts greater pressure on productivity
growth in existing areas. Probably more important in the
long run, declining world prices have caused a slowdown
in input use, investment in agricultural research, and
irrigation infrastructure, all of which affected further yield
improvements. The continuing intensification of rice
farming has led to a diminished gap between potential
yields and actual farmer yields, stagnant or declining
yields on experiment stations, and increased input require-

ments for sustaining current yield gains (Rosegrant and
Pingali 1994).
Changes in policies-particularly those directly target-
ing the agricultural sector, such as reduction in fertilizer
and other subsidies to agriculture-have also contributed
to reduced output growth. The process of production
intensification has caused land degradation in some areas,
resulting in the declines in marginal productivity. Broader
economic and structural reforms-such as the institu-
tional reforms in China, Eastern Europe, and the former
Soviet Union-altered effective relative factor prices and
incentive structures of the sector.
As with crops, the general picture in meat production
was an upward trend, but growth rates slowed starting in
the mid-1970s.2 This slowdown was brought about by two

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010 63

related factors: slower growth in livestock numbers and 1.7 percent per year in meat produced. Similarly, growth
the rapid slowdown in production of ruminants (cattle, rates differed among regions, although relatively higher
sheep, goats) in developing countries, which was only rates of growth were obtained from the developing re-
partly offset by the structural change in meat production gions than from the developed ones.
and consumption towards nonruminants (pigs and poul-
try). The rapid dwindling of fertile grazing lands has Demand Trends in Food Crops and
limited the capacity of some areas to support larger num- Livestock Products
ber of animal stocks, particularly ruminants. In many Population growth has been the primary factor accounting
countries, livestock numbers already exceed the carrying for food demand increases. However, other significant
capacity of unimproved natural grasslands. Thus, much of developments contributed a good deal not only to overall
the increases in meat output came from nonruminants, demand growth but also to improvements in per capital
which are primarily raised on intensive or semi-intensive food intake. Rapid growth of incomes and foreign ex-
commercial farms that use feedgrains. Again, the growth change availability of the oil-exporting and some other
pattern greatly varied by species. Growth rates were high- middle-income countries, and their consequent urbaniza-
est for poultry, averaging more than 3.0 percent per year tion, resulted in the rapid growth in demand for both food
in the number of poultry and 4.0 percent per year in the staples and livestock products-the latter being reflected
amount of meat production. Production rates in cattle and in the demand for feedgrains. Total domestic use of food
buffalo slowed down-from 1.5 percent to 0.6 percent per staples increased by about 1.9 percent per year between
year in the number of livestock, and from 2.7 percent to 1966 and 1989 (Table 5.2).3 Amounts used directly for

Table 5.2-Average annual growth rates in total domestic use of food crops and livestock products,
All Paddy Other Roots and
Crops Wheat Rice Maize Sorghum Cassava Cereals Tubers Pulses Meata Eggsa
Total domestic use 2.58 4.89 2.62 4.64 -0.41 2.20 -0.62 0.81 0.41 5.83 8.59
Food 2.35 5.34 2.90 3.02 -0.46 1.99 -0.61 -0.63 0.34
Feed 4.02 3.10 1.06 6.02 0.35 5.96 -1.12 4.34 1.81
Other uses 1.90 3.33 1.54 3.99 -1.04 2.21 -0.83 1.64 -0.32
Latin America
Total domestic use 2.16 3.21 3.14 3.27 6.09 -0.52 1.94 1.09 0.81 2.43 4.02
Food 2.15 3.62 3.38 2.90 -0.03 -0.26 1.22 1.50 0.85
Feed 2.67 0.72 2.82 3.55 6.39 -0.28 1.07 -2.23 -4.24
Other uses 1.09 2.78 2.20 3.13 4.30 -1.25 1.38 0.64 1.26
Sub-Sahara Africa
Total domestic use 2.83 7.53 4.31 3.33 1.87 2.86 2.01 2.20 1.89 2.05 1.80
Food 3.22 7.72 5.13 3.65 2.26 3.16 2.64 2.49 2.00
Feed 2.05 5.29 -0.20 3.81 -0.43 0.22 3.14 -2.06 7.17
Other uses 1.72 6.48 1.96 1.24 0.58 2.01 0.03 1.82 1.34
Middle East/North Africa
Total domestic use 3.77 3.68 3.47 5.23 1.52 -1.01 3.24 5.92 2.74 1.85 3.54
Food 3.62 3.93 4.18 2.87 1.12 -1.20 -0.91 6.39 2.86
Feed 4.68 3.95 0.14 8.24 3.49 -0.74 4.20 9.47 1.63
Other uses 2.95 2.62 0.96 3.56 2.33 -0.19 3.74 3.87 3.26
Developing countries
Total domestic use 2.91 4.81 3.04 4.47 1.64 2.18 1.32 1.58 0.94 4.17 6.60
Food 2.57 4.90 3.04 2.97 0.28 2.30 -0.06 0.64 0.87
Feed 5.13 4.06 1.10 6.45 6.56 1.38 4.20 4.47 1.78
Other uses 2.58 3.55 2.54 3.98 0.80 2.29 2.09 2.14 0.84
Developed countries
Total domestic use 1.22 1.35 3.53 2.24 -0.18 10.18 1.10 -0.44 1.69 1.63 0.61
Food 0.80 1.15 2.44 2.74 2.23 -5.51 -1.04 0.46 -0.35
Feed 1.28 1.49 3.12 1.77 -0.27 11.13 1.24 -1.75 2.57
Other uses 1.52 1.30 7.07 5.09 3.82 3.94 1.60 -0.33 1.14
Total domestic use 1.91 2.80 2.71 2.97 0.95 2.28 1.19 0.07 1.70 2.59 3.05
Food 2.10 3.43 3.16 3.05 0.49 2.14 -0.05 0.02 1.27
Feed 1.85 1.88 0.05 2.67 1.37 4.03 1.46 0.18 2.92

Other uses 1.58 2.09 1.79 4.46 0.34 0.95 1.53 0.09 1.32
Source: FAO/IFPRI Database.
aFor the 1982-89 period.

64 Population and Food in the Early Twenty-First Century

food, which represented about 38 percent of total domes-
tic use, expanded at the rate of 2.1 percent per year. Feed
use, at the world level, increased at a slower rate because
of the slow growth in the developed countries, which
accounts for the biggest share in total feed use (71 per-
cent). This trend has been partly due to declines in con-
sumption of livestock products in the former CPEs as
market reforms proceeded to eliminate subsidies.4
The shares of feed use in the consumption of food crops
increased over time, while those of food use declined
(Table 5.3). Share of feed use to total domestic use in
Sub-Saharan Africa is lowest, at 6 percent; in other devel-
oping regions, the share ranged from 19 percent (Asia) to
32 percent (Latin America). Growth of feed use was
particularly rapid in the developing countries. FAO
(1993) reports that the rapid income growth in the region
had induced shifts in consumption patterns toward more
livestock products, the demand of which increased at the
rate of 5.3 percent per year between 1970 and 1990.5 Feed
use of food crops in Asia also expanded substantially,
increasing by more than 4.0 percent per year. This raised
the shares from 17 percent in 1966-74 to 21 percent in
1982-89. Despite this rate of growth, however, consump-
tion data in the region show that food use still accounts
for the bulk of domestic use of basic food staples (67
percent). Unlike other developing regions, the share of
food use increased in Sub-Saharan Africa, where direct
human consumption of total food staples represents about
three-quarters of total domestic food use.
Among the food staples, wheat's rate of growth in food
use was 3.4 percent per year, higher than that of rice (3.2
percent per year). The difference is more striking in the
developing countries, with wheat's share for food use in
total domestic use having gradually increased by as much
as 5 percent over time. The share of rice, on the other hand,
remained at almost the same percentage level (Agcaoili
and Rosegrant 1992). Shifts in preferences for higher-
quality food-primarily towards livestock products, in-
cluding milk and eggs-has been taking place as a result
of rising per capital income and a faster rate of urbaniza-
tion.6 Countries in East Asia provide examples of rapidly
increased meat consumption accompanying fast eco-
nomic growth. Countries with lower incomes, on the other
hand, reduce meat consumption or resort to substituting
lower-priced meats. For instance, in Latin America, total
meat consumption declined as real incomes fell during the
1980s and consumption shifted from beef to lower-priced

Trends in Food Trade
Since the mid-1960s, the trade position of developing
countries has improved, particularly in foodgrains. Trends
in foodgrain imports, exports, and net trade as a proportion
of their domestic production are shown in Table 5.4. For
Asia as a whole, the export of total foodgrains increased

from 3.5 percent to 8.0 percent of domestic production
between 1966-74 and 1982-89, while imports slightly
decreased from 8.9 percent to 8.1 percent of domestic
production. In this regard, Asia has performed much better
than the other developing regions in improving net trade
position in foodgrains. Latin America, for example,
shifted from a net exporter to a net importer of foodgrains
in 1982-89, while Sub-Saharan Africa and the developing
countries continued to import more than they exported of
the commodities.
The volume of meat trade remains a very small propor-
tion of world production, partly because of trade restric-
tions (World Bank 1992). In 1991, only 10 percent of beef
output, 3 percent of pork output, and 7.6 percent of poultry
was exported. The United States remains the world's
largest beef importer. However, the importance of Asian
countries, primarily Japan, has increased significantly
since the mid-1980s. The European Union is the largest
beef exporter, accounting for 26 percent of world exports
in 1991, followed by Australia with 21 percent. Beef
exports by some developing countries have also increased
since the mid-1980s, with major gains reported in Argen-
tina and China. In the poultry market, the leading export-
ers are the European Union, the United States, Brazil,
Hungary, and Thailand. The United States remains the
largest exporter of poultry meat. Thailand's poultry ex-
ports are primarily fueled by strong demand in Japan and
other Asian countries as well as by success in the Euro-
pean markets.

Price Trends
Real prices of major grains generally declined in the past.
For wheat, the decline averaged 2.9 percent per year-2.7
percent per year for rice, from 2.9 percent to 3.4 percent
per year for coarse grains, and 1.3 percent per year for beef
(World Bank 1992). It is only during the mid-1970s that
real prices climbed to levels that were even higher than
those attained in the early 1950s. This rise came from
increases in oil prices, which put greater pressure on the
balance of payments of many developing countries and
thus forcing many of them to reduce imports of production
requisites, subsidies on inputs and food consumption, and
expenditures on infrastructure for the agriculture sector.
Agricultural performance was deeply affected, cutting
production tremendously, so that some countries were on
the verge of a food crisis. The African drought com-
pounded the economic problems of the region, culminat-
ing in a full-scale food crisis that engulfed the majority of
the African countries.
From the end of the 1970s until the early 1980s, a
different economic scenario developed, with the industri-
alized countries slumping into their deepest postwar re-
cession at the end of the 1970s. Their growth averaged just
over 1 percent per year between 1980 and 1984. Conse-
quent sluggish demand in these major markets, together

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

with increasing supplies (as a result of the rapid rise in
prices in the mid-1970s), led to falls in many international
commodity prices.
Meat prices have fluctuated much more than grain
prices. This is due to great supply variability amid rising
demand for meat products. The uptrend in beef prices
since the mid-1980s was due to relatively tight beef sup-
plies at a time when the Asian countries continued their
strong demand for beef imports. This is in contrast to the
continuing decline in the price of pork and poultry at a
time when supplies remain in abundance because of more
rapid production from improved technology and from the
shifting of consumption to higher-priced meats as income

The Simulation Model
The IFPTSIM model has been developed at IFPRI to serve
as a research tool in analyzing policy issues, particularly
those related to the relative performance of commodities
in the international market.7 The IFPTSIM model is built
upon existing global trade models, such as the USDA
Static World Policy Simulation (SWOPSIM) model, the
OECD Ministerial Trade Mandate (OECD/MTM) model,
and the FAO World Food Model, with extensions to
incorporate the more desirable features of each of the
models.8 In its present structure, the IFPTSIM model is
(1) partial equilibrium, with its focus on the agriculture
sector; (2) global, covering 33 countries or regions and 14
commodities; (3) dynamic, in that it allows partial adjust-
ment to exogenous shocks; (4) nonspatial (and, hence,
cannot be used to analyze trade patterns); and (5) syn-
thetic, because of its use of elasticities derived from other
The IFPTSIM model is a representation of a competi-
tive agricultural market for crops and livestock. It is
specified as a set of country or regional models, each with
a particular structure within which supply, demand, and
prices for some categories of agricultural commodities are
determined. Some important staple crops-such as roots
and tubers, pulses-are not covered by the model because
of data limitations. The 33 country and regional agricul-
tural submodels are linked through trade, a specification
that highlights the interdependence of countries and com-
modities in the world agricultural economy. The model
uses a system of supply and demand elasticities, incorpo-
rated into a series of linear equations, to approximate the
underlying production and demand functions. Table 5.5
lists the countries, regions, and commodities covered in
the model. The appendix to this chapter presents the
typical set of equations for a given country i and commod-
In the crop models, domestic production is determined
by the area harvested and yield per hectare, as shown in
equation 4. Harvested area is specified in terms of a partial
adjustment function of the crop's own price, cross-price

effects with other crops, and the past year's harvested area
(equation 1 in the appendix). The yield function in the
model is typically a function of the past yield level and a
trend factor that is used to reflect technology develop-
ments (equation 2). For the major foodgrains, the yield
function is expanded to include not only the trend factor,
which reflects technology development, but also the ef-
fects of output and input prices, including fertilizer and
labor (equation 3). Livestock production is modeled simi-
larly (equations 5-7) to the food crops.
Total domestic demand for a commodity is the sum of
demand for food, feed, and other industrial uses. Total
food demand and demand for other industrial uses (equa-
tion 9) are per capital demand for a commodity multiplied
by population. Per capital demand (equation 8) is a func-
tion of real per capital income, the real price of the com-
modity, real prices of competing commodities, and a time
trend representing changes in tastes and preferences. Feed
demand, on the other hand, is a derived demand relating

changes in the volume of output of livestock products to
changes in the composition of feed rations through the
feed demand elasticities (equation 10). Feed demand elas-
ticities determine the rate with which one feed can be
substituted for another in response to a movement of
relative prices.
Prices are endogenous in the system and are generally
specified as the sum of world prices (PW), in domestic
currency terms; the assistance to the commodity measured
by the producer subsidy equivalents and consumer sub-
sidy equivalents (PSEs/CSEs); and a margin (MI) allow-
ing for factors such as transport costs or quality
differences. The PSEs/CSEs make up the standard policy
structure in the IFPTSIM model. However, for some
countries and commodities, policy instruments that ex-
plicitly shift demand and supply relationships are also
used. The U.S. acreage reduction program, for example,
is reflected in the model through a production control
Trade is the difference between domestic supply and
total demand (equations 15 and 16) and, as such, does not
permit separate identification of imports and exports in
cases where a country is an exporter and importer of the
same commodity. Stocks are not explicitly modeled be-
cause markets are assumed to be in equilibrium in the
intermediate run. For some commodities, however, stocks
are presumed proportional to trade prices (PT). The im-
port quota is included only in the stock equation of rice in
Japan. This also has been relaxed slightly in the model
because of Japan's recent move to partially open its mar-
ket to rice imports.
Finally, the market clearing condition is defined by
equating the sum of net trade balances to zero at the world
level. When an exogenous variable is shocked, world
reference prices are assumed to adjust to a new equilib-
rium where each adjustment is passed back to the effective

66 Population and Food in the Early Twenty-First Century

Table 5.3-Average share of food, feed, and other uses to total domestic use of food crops, 1966-89
1966-74 1974-82
Food Feed Food Feed
to Total to Total to Total to Total
Domestic Domestic Other Domestic Domestic Other
Region/Countrya Commodities Use Use Uses Use Use Uses
(percent per year)

Developed countries

Latin America

Sub-Saharan Africa

Developing countries


Paddy rice
Paddy rice
Paddy rice
Paddy rice
Paddy rice
Paddy rice




Source: FAO/IFPRI database.
Notes: OCER is other cereals; and ORT is other roots and tubers.
aRegional definitions follow those of FAO, except Japan is excluded from the developed countries and is included in Asia. Developing

Chapter 5: Global and Regional Food Supply, Demand and Trade Prospects to 2010

1982-89 1966-89
Food Feed Food Feed
to Total to Total to Total to Total
Domestic Domestic Other Domestic Domestic Other
Use Use Uses Use Use Uses
(percent per year)







countries are primarily in the Middle East/North Africa.

68 Population and Food in the Early Twenty-First Century

Table 5.4--Ratio of foodgrain exports and imports to production, 1966-89
1966-74 1974-82 1982-89
Net Net Net
Region Commodities Imports Exports Trade Imports Exports Trade Imports Exports Trade

Developed countries

Latin America

Sub-Saharan Africa

Paddy rice
Paddy rice
Paddy rice
Paddy rice

Developing countries Wheat
Paddy rice


Paddy rice

33.96 0.66 -33.30
2.38 2.40 0.02
15.31 4.01 -11.30
9.23 0.63 -8.60
1.39 18.02 16.63
6.90 0.45 -6.45
0.15 0.23 0.07
1.98 1.67 -0.31
8.91 3.51 ..
12.28 25.02 12.74
9.27 15.16 5.89
16.52 17.62 1.10
25.35 25.96 0.61
0.00 0.00 0.00
6.91 8.68 1.77
1.84 1.76 -0.08
11.20 8.30 -2.90
11.91 14.64
53.44 28.31 -25.13
4.36 3.96 -0.40
3.21 15.66 12.45
4.97 22.83 17.86
0.01 0.46 0.45
20.39 20.45 0.07
1.40 0.51 -0.90
4.69 3.18 -1.51
11.56 11.92


9.46 -114.40
1.58 -13.23
5.89 1.98
0.06 -0.74
0.75 0.75
0.56 -2.98
0.09 -0.05
4.96 3.69
2.92 ..
1.65 -24.04
13.81 0.72
0.29 -8.60
2.87 2.51
0.00 -0.05
2.08 -2.15
8.22 2.56
21.08 14.76
6.25 ..
18.40 0.67
3.47 0.10
12.49 0.46
12.40 1.05
5.01 0.77
5.97 -0.30
0.96 -0.15
4.50 -0.12
7.90 ..


1.00 -25.50
2.53 0.91
4.06 -18.86
1.08 -2.35
35.25 33.76
1.12 -10.89
0.23 0.04
1.71 -0.55
5.87 ...
29.68 15.85
20.30 6.77
27.86 6.56
35.63 -2.33
0.00 0.00
10.76 2.63
2.60 0.10
11.39 -0.67
19.74 ...
30.36 -31.02
4.71 -0.89
14.18 3.35
30.45 14.51
0.20 0.19
19.64 -19.01
0.79 -0.66
7.03 -2.89
13.42 .
17.47 -212.06
0.54 -30.49
3.67 -4.35
0.15 -1.02
0.17 0.17
0.50 -4.48
0.05 -0.05
3.65 2.18
3.27 ...
2.03 -38.91
3.87 -31.16
0.81 -42.40
6.04 2.98
0.01 -0.29
2.30 -12.28
5.66 -1.08

12.75 20.14 7.39
19.58 5.11 .
19.96 20.52 0.56
3.45 3.68 0.23
17.55 17.82 0.27
16.84 18.11 1.27
11.33 13.50 2.17
8.19 7.91 -0.28
1.35 1.20 -0.15
5.69 5.52 -0.17
10.54 11.03 .

19.32 0.87 -18.45
0.87 2.69 1.82
20.98 5.77 -15.21
4.45 2.90 -1.55
4.14 45.87 41.74
10.24 1.82 -8.42
0.34 0.42 0.08
4.72 3.64 -1.08
8.13 8.00 ...
15.38 33.18 17.81
20.49 26.38 5.89
17.47 26.20 8.73
30.18 33.59 3.42
0.00 0.00 0.0
7.59 12.24 4.64
4.54 4.32 -0.23
14.45 16.24 1.79
15.73 21.74 ...
47.88 31.75 -16.12
7.43 3.92 -3.51
11.33 11.44 0.11
23.47 23.30 -0.17
0.05 0.31 0.26
48.31 17.66 -30.65
1.38 0.72 -0.66
11.39 6.81 -4.58
18.90 11.99


3.43 -158.41
0.20 -40.28
3.29 -6.66
0.24 -1.70
0.06 0.04
0.48 -6.80
0.07 -0.21
2.56 -0.42
4.10 -48.75
2.84 -54.80
0.28 -72.70
6.76 3.12
0.54 -45.04
2.62 -39.49
4.44 -1.18
21.18 9.56
5.34 ...
21.46 0.56
3.63 0.38
15.84 0.12
16.63 0.54
18.84 1.62
10.04 0.26
1.84 -0.25
9.03 0.53
12.16 ...

Note: An ellipsis (. ..) means not applicable. OCER is other cereals; and ORT is other roots and tubers.

Chapter 5: Global and Regional Food Supply, Demand and Trade Prospects to 2010 69

Table 5.5-Countries, regions, and commodities covered by the IFPTSIM model

Developed countries

Developing countries


Latin America and the Caribbean

Sub-Saharan Africa

Middle East/North Africa

Asian developing countries


Animal products

United States
European Union
Other Western European countries
Former Soviet Union
Eastern Europe
New Zealand
Other developed countries
Other Latin America and Caribbean countries
Other Sub-Saharan Africa countries
Other Middle East/North Africa countries
South Korea
Other Southeast Asia countries
Other South Asia countries
Other East Asia countries
Other developing countries
Rest of the world
Wheat, rice, maize, other coarse grains, soybeans
Beef, pork, poultry, mutton and lamb, fluid milk, eggs

producer and consumer prices through the price transmis-
sion equations. Changes in the domestic prices sub-
sequently affect the supply and demand of commodities,
necessitating their iterative adjustment until the world
supply-and-demand balance and world net trade are again
equal to zero.
The base year data are three-year averages centered in
1988. The price and income elasticities, as well as data for
the policy variables, are taken from OECD (1991); FAO
(1986); and Roningen, Sullivan, and Dixit (1991). For
many Asian countries, however, the price and income
elasticity parameters are updated using results from more
recent studies.9 Growth rates in yield, population, and per
capital income (GDP) are, likewise, adjusted to reflect
recent trends and developments, especially the effect of
structural and economic reforms now being initiated in
many developing countries.

Simulation Assumptions
Three major factors will continue to influence global and
regional food markets in the next 20 years: population
growth and urbanization, improvement in incomes, and
the rate of increase in food production through further

yield increases. Projections of future food supply and
demand are based mainly on the projected growth rates of
these key parameters.

Population Growth and Urbanization
Population growth and urbanization will be important
determinants of food supply and demand imbalances in
the future. The growth rate of world population seems to
have peaked at 2.1 percent per year in 1965-70 and then
declined progressively to the current level of 1.7 percent
per year (United Nations 1992). Further declines in the
global growth rates are foreseen, with declining birth rates
and the fall in mortality rates leveling off. Reduction in
regional population growth rates are expected to be as
much as 50 percent of their 1970-80 level. Only Sub-Sa-
haran Africa is projected to have a higher growth rate,
from 2.9 percent per year in 1970-80 to slightly more than
3.0 percent per year in the 1990s and beyond.
The rate of urbanization has also become more rapid
with the relatively higher population growth in the urban
areas. Approximately 44 percent of the population of
developing countries is projected to reside in the urban
areas by 2000, up from 22 percent in 1960 and 30 percent

70 Population and Food in the Early Twenty-First Century

in 1980 (Pinstrup-Anderson 1992, Table 15). Urbaniza-
tion will accelerate dietary transition, as has been noted in
the past, from the basic staples-such as sorghum, millet,
maize, and root crops-to cereals that require less prepa-
ration (such as wheat and rice), fruits, livestock products.
and processed foods. In the model, the effect of urbaniza-
tion is reflected in the income and price elasticities.

Improvements in Income
The growth rate in real per capital income during the 1980s
was disappointing for the developing countries as a whole
(IMF 1993). The low average rate of growth covers large
variation among regions. High growth rates were obtained
in Asia, particularly East Asia, while Sub-Saharan Africa,
the Middle East/North Africa,10 and Latin America expe-
rienced significant reductions in per capital real incomes.
The high rates of economic growth in Asia are expected
to continue through the next 20 years. Most other regions
are, likewise, projected to show significant income
growth, with the negative growth in per capital income of
Sub-Saharan Africa being replaced by a very low but
positive growth rate.
The continued improvement of economic conditions in
the various regions rests on the successful completion of
the structural adjustments and economic reforms in Latin
America; Sub-Saharan Africa; the former CPEs, includ-
ing the former Soviet Union; and some countries in Asia
and the Middle East. The success of these reforms (and
current downward shocks in consumption) may enable
countries, especially the former CPE countries, to signifi-
cantly reduce their imports and in the aggregate shift their
position to net exporters. These developments will have a
significant effect on the global market because of the share
of these countries in the world trade of food crops and
livestock products.
However, despite the general rise in income, large
concentrations of the poor still exist in many countries.
About 1 billion people are categorized as living in pov-
erty, most of them in Asia and Sub-Saharan Africa. The
implication of this uneven distribution of income for
future food demands is that the commodity composition
of such demand will greatly vary between regions. In
Africa, for example, demand expansion will occur primar-
ily in staple foods, while demand for more expensive
foods (such as livestock products, sugar, fruits, and proc-
essed foods) will increase sharply in East Asia and the
faster-developing countries of Southeast Asia.

Yield Increases from Technical
Change and Investment

Greater food demand resulting from population and in-
come growth must be met by food production increases.
As indicated earlier, opportunities for food production
increases through expansion into lands that are not cur-

rently under cultivation is very limited. Other resources,
such as water, that facilitated food production increases in
the past are also becoming scarce. This implies that future
output increases must come primarily from higher yields.
Thus, the biggest challenge in the next two or three
decades is the development of more efficient production
technologies. Efforts are now being undertaken, particu-
larly in the developing countries, to promote further in-
creases in agricultural production. Prospects for growth
are being initiated from cost-reducing technological
change that is fully integrated into the larger economy
through increased investment in infrastructure and in
services that permit farmers' adoption of the technology.
Solutions to these problems are complex. They involve
the foundation of strongly linked national and interna-
tional institutions in the areas of production, infrastructure
building, and financing.
Yield growth rates achieved in the past decade through
technical change and investment can be maintained as
long as the pace of policy reform continues and govern-
ments continue to invest in cost-effective agricultural
research, extension, irrigation and water development,
human capital, and rural infrastructure at the levels at-
tained in the late 1980s and early 1990s. Further reduc-
tions from these investment levels (which are already
significantly lower than the levels of the 1970s and early
1980s) will slow production growth rates. The potential
impact of reductions from current real investment levels
(or acceleration of resource degradation) could be serious,
as shown in one of the simulation scenarios below.
What are the policy assumptions underlying the pro-
jected yield growth rates for the major crops in developing
countries? Future growth in rice and wheat productivity in
developing countries in Asia and Latin America will come
from renewal of improvement in yield potential of modern
varieties, through hybrid and new plant structures for rice,
and through continued generation and adoption of wheat
varieties that generate evolutionary increases in yield po-
tential and improved stability in many environments. Im-
provement in yield potential will also come, and, more
importantly, from improved management and efficiency of
use of the scarce resources used in rice production, in
contrast to the rapid dissemination of modem technology
that has been dominant in the past (Rosegrant and Pingali
1994; Byerlee 1993). Thus, fertilizer policy must shift from
having a sole focus on increasing the level of use of fertilizer
to also improving the efficiency of nutrient balance and the
timing and placement of fertilizers. In crop protection, the
focus on disseminating chemical pesticides should shift to
using integrated pest management techniques; and in irri-
gation, the emphasis should shift from investing in new
systems to improving water use efficiency and productivity
in existing systems. Solid yield growth in maize in Asia and
Latin America is also expected as the spread of well-adapted
modern varieties increases, with benefits from public- and

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010 71

private-sector research. The maize sector will also benefit
from reforms described above for wheat and rice.
The yield growth rates projected for Sub-Saharan
Africa assume considerable progress in removing the
constraints discussed above that have caused long-term
stagnation in much of the agriculture sector. While the
policy prescriptions of different observers vary, key ele-
ments of agricultural development in Sub-Saharan Africa
include (1) continued reform of the agricultural policy
environment, including price, exchange rate, marketing,
and input supply policies that will provide incentives for
private sector investment in farming, marketing, and proc-
essing; (2) improved security of land tenure to induce
conservation investments and improve access to credit;
(3) increased investment in research and extension, rural
infrastructure, and human capital development, including
education, health, and nutrition; and (4) improved man-
agement, priority setting, and agroclimatic-specific tar-
geting of research and extension efforts (Cleaver 1993;
Delgado and Pinstrup-Andersen 1993; Reardon et al.
1993; Harrison 1990; Von Braun and Paulino 1990).
Progress in several of these areas of policy reform will be
required to generate the yield growth projected here.
For livestock, given the limitations on the availability
of grazing areas, growth in beef production will primarily
depend on genetic development and improved feeding
technology. Similar technology improvements are being
undertaken in the pork and poultry subsector, where tech-
nical change will be more rapid than in beef. In developing
countries, in particular, the underlying growth rate in
technical change in pork and poultry production will rise
faster than beef, sheep, and goat meat production because
of the import and adaptation of technology from the
developed world. However, the continued investments in
technology improvements and production efficiency, par-

ticularly in the middle-income economies, could result in
declines in production costs.

Results of Baseline Projections

Projected World Prices

The most aggregate indicators of relative food supply and
demand yielded projected world prices of the major food
commodities. Projected price trends are cause for opti-
mism for future aggregate food supplies (an optimism that
will be tempered by emerging problems for some regions
of the world, which will be described below). As can be
seen in Table 5.6, real world food prices are generally
expected to be stable or declining for most commodities.
Wheat prices are projected to decline by 9 percent, maize
prices by 15 percent, and coarse-grain prices by 10 per-
cent. The smallest price declines are projected for rice (1
percent) and soybeans (4 percent). Real prices for all
livestock products are also projected to fall, some quite
significantly. A major structural change behind these de-
clining commodity prices is the shift of the former CPEs
from major cereal importers to significant cereal exporters
due to removal of consumption subsidies and other eco-
nomic reforms.
These trends indicate that, if commitment to agricul-
tural growth is sustained by governments and interna-
tional donors, pressure on aggregate world food supplies
from rising populations and incomes will not be over-
whelming. However, these aggregate price trends mask
emerging problems at the regional and country level: the
regional trends described below indicate a cause for con-
siderable concern for getting the food to where it is needed

Table 5.6-Projected world price levels under alternative scenarios, 1988-2010
Price Scenarios for 2010 Price Levels Scenarios for Growth Rates, 1988-2010
Commodities Level Baseline 1 2 3 4 Baseline 1 2 3 4
(dollars) (percent)
Wheat 130 118 152 148 211 107 -0.44 0.71 0.59 2.23 -0.88
Rice 210 207 244 226 288 226 -0.07 0.68 0.33 1.45 0.33
Maize 93 79 88 93 121 72 -0.74 -0.25 0.00 1.20 -1.16
Other Grains 87 78 92 95 124 84 -0.50 0.25 0.40 1.62 -0.16
Soybeans 211 203 207 219 235 204 -0.18 -0.09 0.17 0.49 -0.15
Beef 1,464 1,329 1,403 1,390 1,553 1,416 -0.44 -0.19 -0.24 0.27 -0.15
Pigmeat 1,597 1,479 1,586 1,621 1,881 1,543 -0.35 -0.03 0.07 0.75 -0.16
Sheepmeat 591 488 562 584 776 559 -0.87 -0.23 -0.05 1.25 -0.25
Poultry 815 700 736 756 868 735 -0.69 -0.46 -0.34 0.29 -0.47
Eggs 976 771 831 765 1,035 805 -1.07 -0.73 -1.10 0.27 -0.87
Source: 1988 data are from the FAO Agrostat database. All other data are results from the IFPTSIM model.
Notes: Scenario 1: 20 percent increase in population growth rate.
Scenario 2:15 percent increase in income growth rate.
Scenario 3: 25 percent reduction in yield growth rate.
Scenario 4: Trade liberalization in all countries.

72 Population and Food in the Early Twenty-First Century

Projected Production and
Consumption Trends

Cereal production in the future will grow at an average
rate of 1.6 percent per year (Table 5.7). Most of this
growth will still come from yield increases, although their
growth rates will be slower than those achieved in the
past.11 For the cereals as a whole, expansion of area will
not be a source of future growth. Similarly, projected
growth rates in consumption are also lower than those
obtained in the past (see Table 5.2). As in the past,
developing countries' consumption of food crops and
livestock products will grow much faster than the devel-
oped regions because of more rapid population growth.
Of greatest concern is the projected rapid growth in
food imports, particularly cereals, in Sub-Saharan Africa.
Imports of cereals in Sub-Saharan Africa are projected to
increase from 11 million tons in 1988 to 32 million tons
in 2010. These large increases in cereal imports are pro-
jected to occur in spite of relatively strong projected
growth in cereal production (2.4 percent per year during

the projection period). Wheat, rice, and other coarse-grain
production is projected to grow at 2.6 percent per year,
while maize is projected to grow at a slower rate of 1.9
percent per year. Growth in area will continue, but at a
slower pace than that attained in the 1980s. Yield growth
rates will also be slightly reduced. If production growth
rates in Sub-Saharan Africa decline to levels comparable
with the rest of the world (1.8-2.0 percent), projected
cereal imports would increase to 42 million tons. Even
rapid food production growth cannot close the food gap
in Sub-Saharan Africa at the projected level of annual
population growth of 3.1 percent. It is likely that financ-
ing even the more optimistic level of imports will be very
difficult to achieve from domestic Sub-Saharan Africa
sources. In the absence of a reduction in population
growth, the international community will need to devise
appropriate combinations of financing and food aid to
bridge these gaps in the foreseeable future. Per capital
consumption levels are therefore lowest in the region, and
only small increases that are far from adequate for nutri-
tional improvement are projected in the next decade.

Table 5.7-Baseline projections for production and consumption of various commodities, by country
1988 2000
Region Commodities Production Demand Net Supply Production Demand Net Supply
(thousand tons)
Developed, all Cereals 870,940 788,825 82,115 1,013,483 889,314 124,169
Soybean 54,631 54,388 243 67,453 64,495 2,958
Ruminants 38,618 38,600 18 44,197 43,037 1,160
Nonruminants 62,323 74,310 -11,987 70,501 70,237 264
Eggs 19,641 19,534 107 21,813 22,001 -188
Eastern Europe and the Cereals 280,029 310,461 -30,432 316,185 324,529 -8,344
former Soviet Union Soybeans 1,292 3,028 -1,736 1,620 3,343 -1,723
Ruminants 11,502 11,689 -187 12,328 12,567 -239
Nonruminants 16,352 15,986 366 17,874 17,247 627
Eggs 6,341 6,314 27 7,209 7,196 13
Developing, all Cereals 812,166 894,258 -82,092 1,014,445 1,138,614 -124,169
Soybeans 45,459 45,702 -243 59,480 62,426 -2,946
Ruminants 21,276 21,294 -18 26,801 27,961 -1,160
Nonruminants 39,370 39,671 -301 51,745 52,009 -264
Eggs 15,941 16,048 -107 21,528 21,340 188
Latin America Cereals 103,393 113,959 -10,566 125,425 138,517 -13,092
Soybean 30,061 27,609 2,452 39,333 36,510 2,823
Ruminants 9,954 9,051 903 12,087 11,088 999
Nonruminants 7,538 7,569 -31 9,607 9,613 -6
Eggs 3,642 3,657 -15 4,874 4,620 254
Sub-Saharan Africa Cereals 51,531 62,185 -10,654 68,710 90,045 -21,335
Soybean 250 250 0 270 358 -88
Ruminants 2,764 2,874 -110 3,423 4,132 -709
Nonruminants 1,179 1,369 -190 1,688 2,003 -315
Eggs 673 706 -33 1,053 1,025 28
Asia Cereals 582,758 603,702 -20,944 724,100 762,098 -37,998
Soybean 14,751 17,289 -2,538 19,362 24,827 -5,465
Ruminants 5,472 5,448 24 7,050 7,434 -384
Nonruminants 28,917 28,559 358 37,820 37,412 408
Eggs 10,113 10,162 -49 13,547 13,622 -75
Middle East/North Africa Cereals 74,484 114,412 -39,928 96,210 147,954 -51,744
Soybean 397 554 -157 515 731 -216
Ruminants 3,086 3,921 -835 4,241 5,307 -1,066
Nonruminants 1,736 2,174 -438 2,630 2,981 -351
Eggs 1,513 1,523 -10 2,054 2,073 -19
Source: 1988 data are from FAO Agrostat database; all other data are results from the IFPTSIM model.

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

Cereal imports are also expected to grow rapidly in the
Middle East/North Africa and East Asia. In the Middle
East/North Africa, cereal imports are projected to increase
by more than 60 percent, from about 40 million tons in
1988 to 65 million tons in 2010, with wheat accounting
for about half of cereal imports (Table 5.7). In developing
Asia, the small regional export surplus in rice will be
maintained, but wheat imports will nearly double because
of diversification of diets out of rice as a result of contin-
ued strong growth in incomes. In addition, imports of
maize and other coarse grains will increase rapidly, as
demand for feeds becomes stronger with the rapidly grow-
ing pig and poultry sectors. These rapidly growing cereal
imports in the Middle East/North Africa and in develop-
ing Asia are not in general a cause for concern, because
they are primarily the result of rapid economic growth,
which will also generate the income to finance the im-
However, although the regional trends for Asia may
not be a cause for general alarm, some emerging problems
again exist at the country level. (Note that country-level

details are not presented in tables but are taken from
detailed model output.) Indian wheat imports are pro-
jected to increase from 440,000 tons in 1988 to 1.3 million
tons in 2010. The projected improvement of incomes in
India resulting from market reforms that are currently
being undertaken will enable the country to finance the
growing wheat imports. Moreover, as income growth
continues, demand could shift from the more basic staples
to other products, enabling India to realize some grain
exports in the future.
In the rest of South Asia, wheat imports are projected
to grow from 3.2 million tons in 1988 to 8.6 million tons
in 2010. This rapidly increasing gap is most serious where
slow growth in wheat production (2.0 percent) will be
coupled with rapid population growth, particularly in
Pakistan, resulting in rapidly expanding imports. The
production projections reflect the relatively poor perfor-
mance of the agriculture sector in Pakistan during the past
decade, which is due to declining investment, government
policies taxing agriculture, and increasingly serious soil
salinity and other environmental degradation (Rosegrant

or region, 1988, 2000, and 2010
2010 Growth Rates, 1988-2010
Production Demand Net Supply Production Demand
(thousand tons) (percent per year)
1,173,946 1,013,289 160,657 1.366 1.145
78,906 72,999 5,907 1.685 1.347
49,525 47,003 2,522 1.137 0.899
83,326 81,194 2,132 1.329 0.404
25,039 23,930 1,109 1.110 0.927
388,982 381,314 7,668 1.505 0.939
1,950 3,567 -1,617 1.889 0.747
14,242 13,318 924 0.976 0.595
21,452 18,688 2,764 1.242 0.712
8,304 7,822 482 1.233 0.978
1,231,310 1,391,967 -160,657 1.909 2.032
73,940 79,878 -5,938 2.236 2.570
32,901 35,421 -2,520 2.001 2.340
65,365 67,497 -2,132 2.331 2.445
26,162 27,267 -1,105 2.277 2.439
151,733 165,379 -13,646 1.759 1.707
49,732 44,520 5,212 2.315 2.196
14,336 13,076 1,260 1.672 1.686
12,020 11,608 412 2.144 1.963
5,624 5,791 -167 1.995 2.111
86,085 118,481 -32,396 2.360 2.974
291 475 -184 0.693 2.960
4,257 5,536 -1,279 1.983 3.025
2,157 2,822 -665 2.784 3.343
1,269 1,382 -113 2.925 3.100
875,584 924,876 -49,292 1.868 1.958
23,262 33,981 -10,719 2.092 3.119
8,775 10,026 -1,251 2.170 2.811
47,711 49,121 -1,410 2.302 2.496
16,796 17,455 -659 2.333 2.489
117,908 183,231 -65,323 2.110 2.164
655 902 -247 2.302 2.240
5,533 6,783 -1,250 2.689 2.523
3,477 3,946 -469 3.208 2.747
2,473 2,639 -166 2.258 2.530

74 Population and Food in the Early Twenty-First Century

and Evenson 1993). Closing the emerging wheat gap in
South Asia will require policy reforms and increased
investment in agricultural growth.
In Latin America, total cereal imports are projected to
grow from 10.6 million tons in 1988 to about 13.6 million
tons in 2010 as demand growth slightly outpaces produc-
tion growth. These gradual increases in cereal imports
should not be difficult to finance, given the solid growth
in income and nonagricultural exports resulting from
Latin American countries' continuing to reform macro-
economic and trade policies. Production growth in cereals
in the region is projected at 1.8 percent per year, while
projected demand growth is only 1.7 percent per year. If
these respective growth rates are sustained, Latin America
could eliminate imports and become a net exporter of
What will be the source for the increased net imports
in the aggregate for the developing countries? The United
States, European Union, and other OECD countries
(mainly Australia and Canada) will significantly boost
cereal exports over the next two decades, primarily be-
cause of slow growth in demand for food from the existing
high levels of consumption together with a slight recovery
in production growth compared with the late 1980s. De-
mand pressure on world markets from Eastern Europe and
the former Soviet Union will also be lessened, as gradual
recovery in production growth is expected to outstrip
demand growth. This will enable the region to shift from
a major net importer of cereals, with imports totaling
about 30.4 million tons in 1988, to a significant net
exporter of about 7.7 million tons in 2010.
Except for sheepmeat, which would have about the
same rate of production as in the past, projected produc-
tion growth rates of livestock products are estimated to be
lower than their respective rates in 1982-90. The beef
production growth rate would be 0.4 percent lower than
the growth rate in 1982-90 (that is, the projection estimate
is 1.4 percent versus the past growth rate of 1.8 percent
per year).Growth rates in pigmeat, poultry, and eggs
would be reduced by 50 percent or more from their
respective past growth rates of 3.6 percent, 4.2 percent,
and 3.2 percent. The slowdown in growth in production
for these meats is caused mainly by declining prices,
which are in turn the result of a slowdown in demand
growth, particularly in the former CPEs-which include
countries in Eastern Europe, the former Soviet Union, and
even China-from the phenomenal demand growth rates
in the 1980s. Until 1991, urban consumption of meat and
milk was heavily subsidized in the Soviet Union. Subsi-
dies, although generally of smaller magnitude, also ex-
isted in the countries of Eastern Europe in the late 1980s,
with the possible exception of Poland. Subsidies have
been eliminated, and consumption of livestock products
have already declined and will continue to decline. Sub-
sequently, the use of grains for feed will also be reduced.

The rapid historical rates of growth in meat demand from
very low base levels in China and the other rapidly grow-
ing countries of Asia will decline as population growth
rates decrease and as income elasticities of demand for
meat decline with increasing incomes.12
Growth in demand for livestock in the developing
countries will remain at about 2.3-2.4 percent per year,
which is lower than the 3.0-4.0 percent rates of growth in
the 1980s. Compared to production growth rates for
grains, the consumption growth rates for the livestock
products will still be higher. This will put great pressure
on the crop subsector in terms of increasing output levels
to meet the fast-growing demand for animal feeds.

Projected Levels of Per Capita
Food Demand

Average per capital demand levels of foodgrains and
livestock products are expected to increase, but at a lower

Table 5.8-Per capital demand of various corn-

Base Year
Region Commodities 1988

Developed, all Cereals 606.2
Soybeans 41.8
Ruminants 29.7
Nonruminants 57.1
Eggs 15.0
Eastern Europe and the Cereals 805.1
former Soviet Union Soybeans 7.9
Ruminants 30.3
Nonruminants 41.5
Eggs 16.4
Developing, all Cereals 230.0
Soybeans 11.8
Ruminants 5.5
Nonruminants 10.2
Eggs 4.1
Latin America Cereals 250.8
Soybeans 60.8
Ruminants 19.9
Nonruminants 16.7
Eggs 8.0
Sub-Saharan Africa Cereals 132.5
Soybeans 0.5
Ruminants 6.1
Nonruminants 2.9
Eggs 1.5
Asia Cereals 229.1
Soybeans 6.6
Ruminants 2.1
Nonruminants 10.8
Eggs 3.9
Middle East/North Africa Cereals 340.7
Soybeans 1.6
Ruminants 11.7
Nonruminants 6.5
Eggs 4.5
Source: 1988 data are from FAO Agrostat database; all other
Note: See Table 5.6 for the definition of the scenarios.

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

rate than in the past. In developing countries, the average
annual expansion rate of per capital consumption of cereals
(primarily rice, wheat, maize, and other coarse grains)
under the baseline scenario would only be 0.4 percent,
from 230 kilograms per capital per year in 1988 to 250
kilograms per capital per year in 2010 (Table 5.8). This
would be about the same growth rate achieved in 1980-
90, which was 0.5 percent per year.13 The major staple,
rice, is projected to grow more slowly because of the
diversification of diets out of rice in response to rapid
income growth and urbanization. For wheat and maize,
however, the more rapid per capital demand reflects large
growth rates in the derived demand for animal feeds. This
also holds true for soybeans, which have grown in impor-
tance as a major source of protein for human consumption
as well as animal feeds. Per capital demand growth for
cereals in the developed countries, which is projected at
0.6 percent per year, would be slightly higher than the rate
achieved in the previous decade (0.5 percent). This is
because of increased demand for coarse grains in the

European Union and other countries in Western Europe,
which will offset the expected demand slowdown in the
former CPEs.
For livestock products, the baseline projections ob-
tained for both the developed and developing countries
show significantly lower annual growth rates in per capital
demand than those in the past. For the developing coun-
tries, projected per capital demand would grow at 0.75
percent (baseline projection result) versus the 2.4 percent
rate of growth in 1980-90 (from the FAO Agrostat data-
base). The reduced per capital growth rate in the develop-
ing countries is mainly influenced by the prospective
demand slowdown in China, which accounts for 40 per-
cent of total meat consumption. The decline is also sig-
nificant in the developed region, where the projected
growth rate would be a mere 0.09 percent per year; the
growth rate was 0.8 percent in the previous decade. This
reduction in meat demand, after the rapid expansion in the
past due to growth in incomes, will come from the stabi-
lization (or for some meat products, slight declines) in

modities under alternative scenarios, by region, 1988-2010

Scenarios for Per Capita Demand Levels, 2010 Growth Rates, 1988-2010
Baseline 1 2 3 4 Baseline 1 2 3 4
(kilograms per capital per year) (percent)
700.3 689.8 717.6 686.4 696.7 0.7 0.6 0.8 0.6 0.6
50.5 49.7 51.5 49.1 50.6 0.9 0.8 1.0 0.7 0.9
32.5 32.0 32.7 31.5 32.3 0.4 0.3 0.4 0.3 0.4
56.1 54.8 56.2 52.6 57.7 -0.1 -0.2 -0.1 -0.4 0.0
16.5 16.2 16.5 15.7 16.6 0.4 0.3 0.4 0.2 0.5
862.5 848.2 874.5 838.4 870.1 0.3 0.2 0.4 0.2 0.4
8.1 8.0 8.0 7.8 8.1 0.1 0.1 0.1 -0.0 0.1
30.1 30.0 29.7 29.3 29.8 -0.0 -0.0 -0.1 -0.1 -0.1
42.3 41.8 42.1 41.3 42.1 0.1 0.0 0.1 -0.0 0.1
17.7 17.4 17.7 17.1 17.6 0.4 0.3 0.4 0.2 0.3
250.2 240.7 254.4 240.7 252.9 0.4 0.2 0.5 0.2 0.4
14.4 13.5 15.1 13.7 14.4 0.9 0.6 1.1 0.7 0.9
6.4 6.1 6.6 6.1 6.3 0.7 0.5 0.8 0.5 0.6
12.1 11.5 12.6 11.6 12.1 0.8 0.5 0.9 0.6 0.8
4.9 4.7 5.1 4.8 4.9 0.8 0.6 1.0 0.7 0.8
279.3 271.3 284.3 270.8 282.6 0.5 0.4 0.6 0.3 0.5
75.2 71.5 77.7 71.1 73.4 1.0 0.7 1.1 0.7 0.9
22.1 21.4 22.5 21.4 21.0 0.5 0.3 0.6 0.3 0.2
19.6 19.1 20.0 18.8 20.0 0.7 0.6 0.8 0.6 0.8
9.8 9.4 10.0 9.3 9.7 0.9 0.7 1.0 0.6 0.9
137.5 130.3 138.7 130.1 137.8 0.2 -0.1 0.2 -0.1 0.2
0.6 0.5 0.6 0.5 0.6 0.2 -0.2 0.4 0.1 0.1
6.4 6.2 6.4 6.0 6.3 0.2 0.0 0.2 -0.1 0.1
3.3 2.9 3.5 3.1 3.3 0.5 0.0 0.8 0.2 0.5
1.6 1.5 1.7 1.6 1.6 0.3 0.0 0.5 0.2 0.3
253.7 246.1 258.9 246.0 257.5 0.5 0.3 0.6 0.3 0.5
9.3 8.9 10.0 9.0 29.3 1.6 1.4 1.9 1.5 7.0
2.8 2.6 2.9 2.7 2.8 1.3 1.1 1.6 1.2 1.3
13.5 12.8 14.0 12.8 13.4 1.0 0.8 1.2 0.8 1.0
4.8 4.6 5.0 4.7 4.8 1.0 0.8 1.2 0.9 1.0
386.2 367.4 388.2 358.0 384.6 0.6 0.3 0.6 0.2 0.6
1.9 1.8 2.0 1.9 1.9 0.6 0.5 0.8 0.5 0.6
14.3 13.6 15.0 13.9 14.5 0.9 0.7 1.1 0.8 1.0
8.3 7.9 8.6 7.8 8.3 1.1 0.9 1.3 0.8 1.2
5.6 5.3 5.8 5.3 5.5 0.9 0.7 1.1 0.7 0.9
data are results from the IFPTSIM model.

76 Population and Food in the Early Twenty-First Century

consumption of the CPEs as relative domestic prices as
well as incomes change in response to the market reforms.
Sub-Saharan Africa shows the lowest per capital con-
sumption levels for most food categories. Furthermore,
only small increases are projected in per capital consump-
tion in the region. This would still be far from adequate
for nutritional improvement, because of rapid population
growth. Latin America's per capital consumption level is
projected to show a rising trend, after several years of zero
growth. Asia is expected to continue a modest and gradu-
ally decelerating growth in per capital demand, particu-
larly in wheat and rice, but with a shift to a stronger per
capital demand for meat products.

Alternative Simulation Results

Several alternative scenarios are considered in this section
to explore how world food trends would be influenced by
changes in the economic environment. Simulation of three
different scenarios are undertaken: (1) a more rapid popu-
lation growth rate; (2) faster economic development
through increases in income growth rates; (3) deteriora-
tion of the growth rate in crop yields, a scenario which
could arise with continued decline in investments in agri-
cultural development or accelerated degradation of the
resource base; and (4) a full trade liberalization scenario.
These scenarios provide deeper understanding of the fu-
ture direction of the world food market.
In all scenarios, real world prices in 2010 for some
commodities, particularly rice and wheat, will be higher
than their base year price levels (Table 5.6). A 20 percent
increase in the population growth rate, for example, would
cause rice and wheat price levels to increase by less than
1 percent per year over the projection period. Prices for
maize, soybeans, and livestock products would decline,
but at a much slower rate than those projected in the
baseline scenario. This is due to consumers shifting from
livestock (and the resultant drop in derived demand for
maize and soybeans) to cheaper food items with lower real
per capital income as prices rise and as population in-
creases. Other coarse-grain prices will rise as their de-
mand increases, especially in countries where they are
consumed as basic staples, such as those in Sub-Saharan
The largest price appreciation, for all products, would
come from the yield reduction scenario, indicating the
potentially serious problems that would arise from failure
to pursue policies to maintain yield growth. Increases in
real world prices would range from 0.5 percent per year
for soybeans to as high as 2.2 percent per year for wheat.
For the livestock products, price growths would range
from 0.30 percent per year for beef and eggs to 1.25
percent per year for sheepmeat. The effects of the relative
price movements become much clearer when the supply
and demand trends under each scenario are considered.

Scenario 1: 20 Percent Increase in
Population Growth Rate
The direct effect of population growth has a larger positive
impact on total consumption than the accompanying re-
duced per capital income. Thus, a higher population
growth rate is expected to result in larger demand in-
creases in both the crops and livestock products. The
increased total demand for food drives prices up, provid-
ing some boost in output, as shown in Table 5.9. As a
consequence, per capital food consumption falls signifi-
cantly as per capital income drops and prices increase. Net
food imports are expected to expand, especially in Asia
and Sub-Saharan Africa, as demand increases more rap-
idly. In Sub-Saharan Africa, demand for foodgrains is
projected to increase by more than 3.3 percent per year;
in the baseline scenario, the projected demand increase is
3.0 percent per year. Net imports will therefore be about
14 percent higher than the baseline net imports. Asia's net
imports would be 25 percent higher than that obtained in
the baseline scenario. As expected, projected per capital
demand of commodities would be smaller than those in
the baseline scenario.

Scenario 2: 15 Percent Increase in
Per Capita GDP Growth Rate
The assumption of a 15 percent increase in the per capital
income growth rate will also result in higher commodity
prices in 2010 relative to the prices projected in the
baseline. Price increments are, however, lower than those
projected in scenario 1 for rice and wheat, but higher for
maize, other coarse grains, and soybeans. Some livestock
products, especially the nonruminants, will also show a
significant price rise. This price behavior comes from
greater projected demand for livestock products in many
developing countries, which will consequently increase
net imports of the commodities. In contrast, projected
grain imports will rise only slightly as a result of the
combined effects of slightly higher production (induced
by relatively high prices) and lower demand (due to
substitution with other food commodities). In response to
this projected pattern of consumption growth in develop-
ing countries, the developed countries will have higher
levels of livestock product exports than the baseline. Their
grain exports are slightly higher also. In contrast to the
estimates in scenario 1, per capital demand of commodities
in this scenario expand slightly.

Scenario 3: 25 Percent Reduction in
Yield Growth Rate
This assumption would lead to a general contraction of
output, which would significantly raise prices, as indi-
cated earlier. This is particularly expected in developed
regions where output increases from area expansion in
response to price increases are very limited. In developing

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

countries, however, area expansion for some crops can be
expected, with additional land being brought into cultiva-
tion, especially in land-surplus regions (such as Eastern
Europe and the former Soviet Union, Latin America, and
Sub-Saharan Africa) in response to the price increases,
which would partially or wholly offset the initial yield
growth decline.
The projected total (as well as per capital) demand for
crops and livestock products would drop significantly
because of the resulting high commodity prices. This
would have severe adverse effects on the developing
countries, particularly in Sub-Saharan Africa and deficit
regions in South Asia. Sub-Saharan Africa's per capital
grain consumption would be almost 5.4 percent lower
than that projected in the baseline scenario. For the devel-
oping region as a whole, this would mean a drop of about
4.0 percent from the per capital demand estimates in the
baseline scenario. In the developed region, consumption
levels would slightly fall from those obtained under the
baseline scenario. This negligible change in consumption
levels implies smaller net balances for exports.

Scenario 4: Trade Liberalization
The final scenario simulates full trade liberalization for all
agricultural commodities. This scenario does not attempt
to simulate implementation of the General Agreement of
Tariffs and Trade (GATT) reforms but rather examines
complete removal of trade barriers. Real world prices of
most meat products under this scenario are also projected
to decline, but again at a slower rate, so that 2010 price
levels will be slightly higher than those projected in the
baseline simulation. In the case of ruminants (particularly
beef), the higher projected prices in 2010 under trade
liberalization ($1,416 per ton), compared with the base-
line projected prices ($1,329 per ton), will come from the
combined effect of reduced production14 and mainte-
nance of consumption levels as the import demand from
Japan and South Korea will increase. Price trends in
nonruminants (pigs and poultry) will be primarily influ-
enced by China's market reforms because of this coun-
try's continued dominance of the world markets
particularly of pigmeat.15 The removal of price subsidies
for feeds, including rice, will account for production
contraction, which will subsequently raise prices to levels
above those projected for the baseline.
Because ofthe reduction in livestock numbers, demand
for feedgrains, including wheat and meals, will be lower.
This explains the significant decline in prices of major
feedgrains, particularly maize. The decline in wheat
prices, on the other hand, will be driven primarily by
production increases in the United States that result from
the removal of the area production program. The large
excess supply from Eastern Europe and the former Soviet
Union will more than offset the fall in output and subsi-
dized exports from the European Union.

Trade liberalization will also open the highly protected
high-income rice markets of Japan and South Korea,
maintain access to important markets in the European
Union, and limit adoption or expansion oftrade-distorting
internal support policies in Asia and elsewhere. This will
substantially increase import demand for rice as well as
prices, particularly for thejaponica rice, which is gener-
ally preferred by East Asian countries. The implication of
this policy on Asia is very important, as rice is highly
protected in the region. The scenario projects rice exports
to decrease slightly to 6.3 million tons in 2010 (compared
with about 7.0 million tons in the baseline scenario). Note
that production will rise, but at the same time a larger
portion of production will be absorbed domestically as
domestic prices are adjusted to reflect true market condi-
tions. Per capital consumption of rice will increase in
almost all countries.
Trade liberalization would have an income effect that
is not captured in the model, because income growth is
determined exogenously. This could change the produc-
tion, consumption, and price behavior described above. A
more open market environment, which enhances trading
and investment opportunities, should lead to reallocation
of economic resources to more productive uses, such that
higher productivity should garner higher returns to those
resources. Incomes will then be increased in the process
and in turn will increase demand for most products and
affect food balances and prices. This would increase de-
mand for rice in the Middle East and Sub-Saharan Africa,
for example, where it is considered a superior food. In the
rice cultures of Asia, on the other hand, higher incomes
will reduce rice consumption in favor of wheat and live-
stock products. Higher incomes will always increase de-
mand for livestock products and necessary feeds. In
addition, in developing countries where use of coarse
grains for food is still prevalent, higher incomes would
result in a shift from coarse grains to wheat and rice. These
shifts are not captured in the projection results, as incomes
do not adjust simultaneously.


If governments and the international community maintain
(or renew) their commitment to agricultural growth
through policy reform and sustained, cost-effective in-
vestment in agricultural research, extension, irrigation
and water development, human capital, and rural infra-
structure, pressure on aggregate world food supplies from
rising populations and incomes will not be overwhelming.
Projected per capital availability of food will increase, and
real world food prices will be stable or declining for the
key food crops. However, these aggregate price trends
conceal emerging problems at the regional and country
level, showing that problems will continue in getting food
to those people who need it the most.

78 Population and Food in the Early Twenty-First Century

Table 5.9-Production and consumption of various commodities under alternative scenarios, by
Scenario I Scenario 2
Region Commodities Production Demand Net Supply Production Demand Net Supply
(thousand tons)
Developed, all Cereals 1,197,343 1,020,286 177,057 1,196,881 1,038,288 158,593
Soybeans 79,873 73,585 6,288 82,561 74,536 8,025
Ruminants 50,092 47,307 2,785 50,134 47,276 2,858
Nonruminants 83,764 81,033 2,731 84,862 81,304 3,558
Eggs 25,686 23,943 1,743 25,970 23,846 2,124
Eastern Europe and the Cereals 404,658 385,433 19,225 404,743 386,615 18,128
former Soviet Union Soybeans 1,938 3,627 -1,689 1,965 3,543 -1,578
Ruminants 14,326 13,631 695 14,265 13,131 1,134
Nonruminants 21,534 19,016 2,518 21,764 18,612 3,152
Eggs 8,315 7,924 391 8,359 7,828 531
Latin America Cereals 157,171 169,290 -12,119 156,945 168,348 -11,403
Soybeans 50,012 44,612 5,400 51,124 46,010 5,114
Ruminants 14,913 13,345 1,568 14,785 13,307 1,478
Nonruminants 12,337 11,921 416 12,520 11,816 704
Eggs 5,662 5,870 -208 5,682 5,939 -257
Sub-Saharan Africa Cereals 89,498 126,561 -37,063 89,921 119,508 -29,587
Soybeans 291 495 -204 293 500 -207
Ruminants 4,371 6,005 -1,634 4,359 5,499 -1,140
Nonruminants 2,151 2,839 -688 2,154 3,007 -853
Eggs 1,272 1,462 -190 1,274 1,436 -162
Asia Cereals 896,514 958,160 -61,646 889,992 943,721 -53,729
Soybeans 23,302 34,550 -11,248 23,651 36,395 -12,744
Ruminants 8,914 10,254 -1,340 8,944 10,577 -1,633
Nonruminants 48,096 49,942 -1,846 48,234 50,944 -2,710
Eggs 16,883 17,972 -1,089 16,921 18,340 -1,419
Middle East/North Africa Cereals 120,469 186,698 -66,229 120,336 184,210 -63,874
Soybeans 656 931 -275 664 933 -269
Ruminants 5,543 6,922 -1,379 5,548 7,111 -1,563
Nonruminants 3,380 3,993 -613 3,398 4,097 -699
Eggs 2,450 2,706 -256 2,453 2,739 -286
Developing, all Cereals 1,263,652 1,440,709 -177,057 1,257,194 1,415,787 -158,593
Soybeans 74,261 80,588 -6,327 75,732 83,838 -8,106
Ruminants 33,741 36,526 -2,785 33,636 36,494 -2,858
Nonruminants 65,964 68,695 -2,731 66,306 69,864 -3,558
Eggs 26,267 28,010 -1,743 26,330 28,454 -2,124
Note: See Table 5.6 for the definition of the scenarios.

The most serious continuing problem is the projected
rapid growth in food imports, particularly cereals, in
Sub-Saharan Africa. Imports of cereals in Sub-Saharan
Africa are projected to nearly double, from 11 million tons
in 1988 to 21 million tons in 2010, despite relatively
optimistic projected growth in cereal production in Sub-
Saharan Africa. If production growth rates in Sub-Saharan
Africa decline to levels comparable to the rest of the
world, projected cereal imports in this region would in-
crease to 42 million tons. These food gaps indicate the
likelihood of continued chronic malnutrition and the
threat of periodic famine for years to come. Even rela-
tively rapid food-production growth cannot close the food
gap in Sub-Saharan Africa at the projected level of annual
population growth of 3.1 percent. Domestic sources of
foreign exchange in Sub-Saharan Africa are not likely to
be able to finance even the more optimistic level of
imports. If population growth rates cannot be slowed, the
international community will need to implement appro-
priate combinations of financing and food aid to bridge
these gaps in Sub-Saharan Africa for the foreseeable future.

Projected rapidly growing food imports (particularly
wheat) in South Asia potentially pose another serious
problem. While projected increased imports are partly a
measure of success (for example, improved income
growth in India), they also indicate difficult problems in
the rest of South Asia, where slow growth in wheat
production will be coupled with rapid population growth
(particularly in Pakistan, with a population growth rate of
3.0 percent), resulting in rapid growth in wheat consump-
tion. The production projections reflect the relatively poor
performance of the agriculture sector in Pakistan during
the past decade due to declining investment, government
policies taxing agriculture, and increasingly serious soil
salinity and other environmental degradation. Closing the
emerging wheat gap in South Asia will require policy
reform and increased investment in resource conservation
and agricultural growth.
The scenario that explores the impact of a significant
decline in yield growth due to reduction in international
and national investment in agriculture and increased deg-
radation of the resource base is also cause for great con-

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

region, 2010
Scenario 3 Scenario 4
Production Demand Net Supply Production Demand Net Supply
(thousand tons)
1,145,297 993,202 152,095 1,207,938 1,008,023 199,915
78,666 71,011 7,655 81,191 73,138 8,053
47,989 45,578 2,411 46,486 46,751 -265
80,253 76,174 4,079 83,599 83,474 125
25,528 22,661 2,867 24,753 24,063 690
396,897 370,672 26,225 387,139 384,660 2,479
1,772 3,466 -1,694 1,940 3,564 -1,624
13,470 12,971 499 14,473 13,153 1,320
20,341 18,279 2,062 21,996 18,629 3,367
7,705 7,566 139 8,426 7,784 642
152,210 160,315 -8,105 126,358 167,353 -40,995
46,315 42,096 4,219 47,945 43,437 4,508
14,595 12,659 1,936 16,489 12,427 4,062
11,824 11,157 667 12,437 11,821 616
5,244 5,488 -244 5,716 5,749 -33
85,955 112,137 -26,182 85,308 118,724 -33,416
286 465 -179 291 475 -184
4,093 5,170 -1,077 4,361 5,388 -1,027
1,863 2,654 -791 2,181 2,806 -625
1,106 1,340 -234 1,286 1,377 -91
836,226 896,692 -60,466 878,697 938,683 -59,986
21,418 32,943 -11,525 23,037 35,206 -12,169
8,236 9,711 -1,475 9,097 10,086 -989
43,708 46,765 -3,057 49,160 48,707 453
15,199 17,233 -2,034 16,999 17,440 -441
112,542 169,884 -57,342 116,982 182,500 -65,518
588 882 -294 655 897 -242
4,781 6,578 -1,797 5,115 6,901 -1,786
2,793 3,691 -898 3,389 3,957 -568
2,173 2,533 -360 2,500 2,624 -124
1,186,933 1,339,028 -152,095 1,207,345 1,407,260 -199,915
68,607 76,386 -7,779 71,928 80,015 -8,087
31,705 34,118 -2,413 35,062 34,802 260
60,188 64,267 -4,079 67,167 67,291 -124
23,722 26,594 -2,872 26,501 27,190 -689

cern. Moderate declines in crop yield growth from those
projected in the base run result in very large increases in
food prices and declines in per capital consumption of
food. These results show the necessity of maintaining
international support for investment in agriculture, at a
time when donor interest is waning. Much of the excite-
ment of the early Green Revolution and construction of
huge irrigation projects is over, and the tasks facing na-
tional governments and the international community are
at once both more complex and less obviously rewarding.
Future gains in productivity in food crop production will
require the more efficient use of important resources, such
as water, rather than expansion in resource availability.
Such gains also will be more difficult to obtain as research
and extension investments generate incremental rather
than revolutionary gains. Nevertheless, the rates of return
to these investments remain very high throughout the
developing world (Rosegrant and Evenson 1993).
In Sub-Saharan Africa, efforts for reducing population
growth rates must continue--to improve the agricultural
policy environment, security of land tenure, and access to

credit; and to improve the management of and selectively
increase investment in research and extension, rural infra-
structure, and development of human capital. Great strides
have been made in improving the availability and af-
fordability of food on a broad basis. However, the chal-
lenges to maintain this momentum remain formidable.

1. Basic food staples here consist of rice, wheat, maize,
sorghum, and other cereals, roots and tubers, and
pulses, for both food and feed use. Food staples, food
crops, and foodgrains refer to the same group of
commodities and are therefore used interchangeably
in the paper, especially in sections preceding the
discussion of the simulation results.
2. Much of the discussion of livestock production and
consumption is taken from various published materi-
als, such as Agriculture Towards 2000 (FAO 1987)
and the updated version Agriculture Towards 2010
(FAO 1993), Crosson and Anderson (1992), and
other World Bank publications.

80 Population and Food in the Early Twenty-First Century

3. Available data for crop and livestock product use are
only up to 1990. The time series for the demand
analysis is only up to 1989.
4. Note that the former CPEs share about 40 percent of
the developed countries' total cereal consumption
between 1969-71 and 1988-90 (FAO 1993).
5. The slower rate of total domestic use for meat in the
1980s, as shown in Table 5.2, reflects stabilization of
demand for livestock products.
6. Per capital consumption of meat products increased
by about 2.1 percent per year between 1961 and 1990,
while that of cereals was less than 1 percent (esti-
mated from FAO Agrostat data).
7. The IFPTSIM model is a modified version of the
trade model developed by K. Oga using the Oga
model-building system (Oga and Gehlar 1993).
8. The IFPTSIM model can be constructed to replicate
some of the world trade models, particularly the
SWOPSIM and OECD/MTM models.
9. These studies are conducted in conjunction with
IFPRI's Projections and Policy Implications of
Medium- and Long-Term Rice Supply and Demand
10. The Middle East is defined here as the region encom-
passing Afghanistan, Bahrain, Cyprus, Gaza Strip,
Iran, Iraq, Jordan, Kuwait, Lebanon, Oman, Qatar,
South Arabia, Syria, Turkey, United Arab Emirates,
and Yemen.
11. Results by crop can be obtained from the authors
upon request.
12. The reduced per capital growth rate in developing
countries is mainly influenced by the prospect of a
demand slowdown in China, which accounts for 40
percent of total meat consumption. Reasons for the
projected slowdown of consumption in China are
discussed in more detail inAgriculture Towards 2010
(FAO 1993, Chapter 3). The rapid growth in per
capital meat consumption in the past is believed to
have started from very low levels (4.5 kilograms in
1961-63) to reach 23.5 kilograms in 1988-90. Such
rapid growth is unlikely to happen again considering
that income levels in China would not grow as fast.
13. Estimated from the FAO Agrostat database.
14. Increased supply from countries, such as Brazil and
Argentina, that currently tax their producers will
partly offset supply reduction in developed countries
when the latter's subsidies for producers are removed.
15. China presently accounts for one-third of the world
supply of pigmeat.

Agcaoili, M. C., K. Oga, and M. W. Rosegrant. 1993.
Structure and operation of the International Food Pol-
icy and Trade Simulation (IFPTSIM) model. Paper
presented in the Second Workshop of the Research

Project on Projections and Policy Implications of
Medium- and Long-Term Rice Supply and Demand,
IRRI, Los Baflos, the Philippines.
Agcaoili, M. C., and M. W. Rosegrant. 1992. Regional
trends and projections for food supply, demand, and
trade in Asia. Washington D.C.: International Food
Policy Research Institute.
Bouis, H. 1989. Prospects for rice supply/demand bal-
ances in Asia. Final report submitted to Rockefeller
Foundation. Washington, D.C.: International Food
Policy Research Institute.
1991. Rice in Asia: Is it becoming a commer-
cial good? Comment. American Journal of Agricul-
tural Economics 73 (2): 522-527.
Byerlee, D. 1993. Modem varieties, productivity, and
sustainability; recent experience and emerging chal-
lenges. Paper presented at the 1993 American Agricul-
tural Economics Association Pre-Conference
Workshop on Post-Green Revolution Agricultural De-
velopment Strategies in the Third World: What Next?
Orlando, Fla., U.S.A., August.
Cahill, C., and W. Legg. 1990. Estimation of agricultural
assistance using producer and consumer subsidy
equivalents: Theory and practice. OECD Economic
Studies, Paris (13): 13-44.
CIAT (International Center for Tropical Agriculture).
1992. Trends in CIAT commodities, 1992. CIAT
Working Document 3, June.
Cleaver, K. 1993. Green revolution directions: Summary
of workshop meetings. Summary discussion for the
1993 American Agricultural Economics Association
Pre-Conference Workshop on Post-Green Revolution
Agricultural Development Strategies in the Third
World: What Next? Orlando, Fla., U.S.A., August.
Comia, C.A., R. van der Hoeven, and T. Mkandawire, eds.
1992 Africa's recovery in the 1990 's: From Stagnation
and adjustment to human development. New York: St.
Martin's Press.
Crosson, P., and J. R. Anderson. 1992. Resources and
global food prospects: Supply and demandfor cereals
to 2030. World Bank Technical Paper 184. Washing-
ton D.C.: World Bank.
Delgado, C. L., and P. Pinstrup-Andersen. 1993. Agricul-
tural productivity in the Third World: Patterns and
strategic issues. A keynote address to the 1993 Ameri-
can Agricultural Economics Association Pre-Confer-
ence Workshop on Post-Green Revolution
Agricultural Development Strategies in the Third
World: What Next? Orlando, Fla., U.S.A., August.
FAO (Food and Agriculture Organization of the United
Nations). Various years. Production yearbook. Rome.
Various years. Trade yearbook. Rome.

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

1986. The FAO World Food Model: Model
specification, supplement to FAO agricultural com-
modity projections to 1990. Rome.
1987. Agriculture Towards 2000. Rome.
1988. Economic and social development: Po-
tentials for agricultural and rural development in
Latin America and the Caribbean. Rome.
1993. Agriculture Towards 2010, Rome.
Harrison, P. 1990. Sustainable growth in African agricul-
ture. In Economic and sectoral policy issues. The
Long-Term Perspective Study of Sub-Sahara Africa,
vol. 2. 48-73. Washington, D.C.: World Bank.
IMF (International Monetary Fund). 1993. World eco-
nomic outlook. Washington D.C.
Lele, U. 1989. Agricultural growth, domestic policies, the
external environment and assistance to Africa.
MADIA Discussion Paper 1. Washington, D.C.:
World Bank.
Mellor, J., C. Delgado, and M. Blackie. 1988. Accelerat-
ing food production in Sub-Sahara Africa. Baltimore,
Md., U.S.A.: John Hopkins University Press.
Mitchell, D., and M. Ingco. 1993. The world food outlook:
Malthus must wait. International Economics Depart-
ment, World Bank, Washington, D.C., July. Mimeo.
OECD (Organization of Economic Cooperation and De-
velopment). 1991. The OECD Ministerial Trade Man-
date Model for policy analysis: Documentation.
OECD/GD(91)159. Paris.
Oga, K., and C. Gehlar. 1993. The International Food
Policy Simulation (IFPSIM) model: A documentation
and applications. International Food Policy Research
Institute, Washington, D.C. Mimeo.
Paulino, L., and J. W. Mellor. 1984. The food situation in
developing countries: Two decades in review. Food
Policy 9 (4): 291-303.
Pinstrup-Andersen, P. 1992. Global perspectives for food
production and consumption. Tidsskriftfor Land Ok-
onomi (December): 145-169.
1993. World food trends and how they may be
modified. Paper presented during the CGIAR Center
Week. International Food Policy Research Institute,
Washington, D.C.
Pinstrup-Andersen, P., and P. B. Hazell. 1985. The im-
pact of the Green Revolution and prospects for the
future. Food Reviews International 1 (1): 1-25.
Puetz, D., J. Von Braun, R. Hopkins, D. Madani, and R.
Pandya-Lorch. 1992. Underrated agriculture: Increas-
ing need but declining aid for agriculture in low-
income countries. International Food Policy Research
Institute, Washington, D.C. Mimeo.
Reardon, T., E. Crawford, R. Bernsten, C. Eicher, J.
Oehmke, and K. Savadogo. 1993. Challenges for cre-

ating and sustaining a green revolution in Africa. Paper
presented at the 1993 American Agricultural Econom-
ics Association Pre-Conference Workshop on Post-
Green Revolution Agricultural Development
Strategies in the Third World: What Next? Orlando,
Fla., U.S.A., August.
Roningen, V., J. Sullivan, and P. Dixit. 1991. Documen-
tation of the Static World Policy Simulation
(SWOPSIM) model framework. Washington, D.C.:
U.S. Department of Agriculture.
Rosegrant, M. W., and R. E. Evenson. 1993. Determi-
nants of productivity growth in Asian agriculture: Past
and future. Paper presented at the 1993 American
Agricultural Economics Association Pre-Conference
Workshop on Post-Green Revolution Agricultural De-
velopment Strategies in the Third World: What Next?
Orlando, Fla., U.S.A., August.
Rosegrant, M. W., and F. Kasryno. 1993. Food crop
production growth in Indonesia: Supply response with
choice of technique. International Food Policy Re-
search Institute, Washington, D.C. Mimeo.
Rosegrant, M. W., and P. L. Pingali. 1994. Policy and
technology for rice productivity growth in Asia. Jour-
nal of International Development 6 (6): 665-688.
Stewart, F., L. Lall, and S. Wangwe, eds. 1992. Alterna-
tive Development Strategies in Sub-Saharan Africa.
New York: St. Martin's Press.
Tyers, R., and K. Anderson. 1986. Distortions in world
food markets: A quantitative assessment. Background
Paper 22. World Development Report. Washington,
D.C.: World Bank.
1988. Liberalizing OECD agricultural policies
in the Uruguay Round: Effects in trade and welfare.
Journal ofAgricultural Economics 39: 97-216.
United Nations. 1992. WorldPopulation Prospects, 1992.
New York.
Von Braun, J., and L. Paulino. 1990. Food in Sub-Sahara
Africa: Trends and policy challenges for the 1990s.
Food Policy 15 (6): 505-517.
World Bank. 1991(a). China: Managing an agricultural
transformation-Grain sector review. Report 8552-
CHA. Washington, D.C.
1991(b). World development report 1990.
New York: Oxford University Press for the World
1992. Priceprospectsfor majorprimary com-
modities. ITD/IED. Washington, D.C.
1993. Global economic prospects and the de-
veloping countries. Washington D.C.
Yudelman, M. 1993. Demand and supply of foodstuffs up
to 2050 with special reference to irrigation. World
Wildlife Fund, Washington, D.C. Mimeo.

82 Population and Food in the Early Twenty-First Century

Appendix: General Structure of the Trade
Simulation Model, IFPTSIM

Crop Production

Area: log ABt, y = a, ya + al, (1 + bt, y) log ABt-, y
+ a2, i log PSt, i
+ -c-j (a3, ic log PSt, ic) (1)

Yield: YHt, j = (1 + bt, i) YHt-, i

YHt, i = (1 + bt, ) YHt-1, i
+ al, y log PSt, y + Ej (al, PFt, y) (3)

Production: QSt, y = ABt, y x YHi, y

Livestock Production

Number of animals:
log ALt, y = ao, y + al, y (1 + bt, U) log ALt-1, y
+ al, y log PSt, y
+ Y-_j (a2, ic log PSt, ic)
+ Yj (al, y log Plt, i)

Yield: YLt, = (1 = + bt, i) YLt-1, y

Production: QSt, y = SLt, y x YLt, y


Per capital demand for food and other uses:
log PQFt, y = ao, i + al, y log PDt, y
+ XYcj (a3, ic log PDt, ic)
+ a4 log YYt, i

Total demand for food (QF) and other uses (QE):
QFt, ij (or QEt, i) = PQFt, y x NNt, i
x (1 + bt, in)

Total demand:
TQD = QDt, + QLt, (

log QLt, y = ao, y + ,j al, y log Plt, i
+ XFj (a2, ic log PIt, ic)
+ Yj d, ij log QSt, i (

Domestic Prices

Intermediate (feed) prices:
Pt, y = XR, i x P W, y + MIt, y + PSEIt, y (

Supply prices:
PSt, y = XRt, ix PWt, y + MIt, y + PSEt, (

Demand price:
PDt, i = XRt i x PW1, y + MIt, it + CSEt, y

Net Trade

Crop: NMt, y = QDt, y + QLt, + STt, y QSt, y

Livestock: NMI, y = QDt, y + STt, y QSI, y


STt, yi = STt-1, y x (PTt-1, i / PTt, y)

Market Clearing

,i NM, y = 0


AB is the crop area;
AL is the number of animal livestock;
SL is the number of animal slaughtered;
(5) YH is crop yield per hectare;
() YL is livestock yield per animal;
QS is crop/livestock production;
() PQF is per capital demand for food and other uses;
QD is total demand of food and other uses;
QL is total demand for feeds;
TQD is total demand for all uses;
YY is per capital income;
NN is total population;
PS and PD are effective prices at the producer
() and consumer level, respectively;
PF is the price of factor inputs (for example,
fertilizer, labor inputs);
PI is the effective intermediate (feed) price;
() PW is the commodity world price;
9) XR is the exchange rate in local currency per U.S.
S PSE is the producer subsidy equivalent reflecting
transfers and other support paid to producers
in implementing agricultural policies;
CSE is the consumer subsidy equivalent reflecting
transfers and other support paid to consumers in
) implementing agricultural policies;
PSM and CSM are the parts of the PSE that are
related to the market price;
PSO and PCO are the parts of the producer and
12) consumer subsidy equivalent, respectively,
other than market price support;
PSEI and CSEI are subsidy equivalents for inter-
13) mediate products;

Chapter 5: Global and Regional Food Supply, Demand, and Trade Prospects to 2010

MI is the margin between world and domestic
intermediate prices;
NM is net trade;
ST is stock level;
a is elasticity in the supply, demand, and feed-use

b is the growth rate;
d is the feed ratio;
i is the country index;
j is the index for the product considered;
c-j is the index for all products butj; and
t is the time index.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs