World agriculture: tow

ards 2015/2030Sum

mary report

FAO

towards 2015/2030World agriculture:

Summary report

towards 2015/2030World agriculture:

Summary report

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS

Rome, 2002

The designations employed and the presentation of material in thisinformation product do not imply the expression of any opinionwhatsoever on the part of the Food and Agriculture Organization ofthe United Nations concerning the legal status of any country,territory, city or area or of its authorities, or concerning thedelimitation of its frontiers or boundaries. In the presentation ofstatistical material, countries are, where appropriate, aggregatedin the following main economic groupings: "Developed countries"(including the developed market economies or industrial countriesand the transition countries) and "Developing countries". Thedesignations "developed" and "developing" are intended forstatistical convenience and do not necessarily express a judgementabout the stage of development reached by a particular country.

All rights reserved. Reproduction and dissemination of material in thisinformation product for educational or other non-commercial purposes areauthorized without any prior written permission from the copyright holdersprovided the source is fully acknowledged. Reproduction of material in thisinformation product for resale or other commercial purposes is prohibitedwithout written permission of the copyright holders. Applications for suchpermission should be addressed to the Chief, Publishing and MultimediaService, Information Division, FAO, Viale delle Terme di Caracalla, 00100Rome, Italy or by e-mail to copyright@fao.org

© FAO 2002

ISBN 92-5-104761-8

iii

Foreword

This report summarizes the main findings of the FAO study, World agriculture:towards 2015/30, which updates and extends the FAO global study, Worldagriculture: toward 2010, issued in 1995. It assesses the prospects, world-wide, for food and agriculture, including fisheries and forestry, over the yearsto 2015 and 2030. It presents the global long-term prospects for trade andsustainable development and discusses the issues at stake in these areasover the next 30 years.

In assessing the prospects for progress towards improved food securityand sustainability, it was necessary to analyse many contributory factors.These range from issues pertaining to the overall economic and internationaltrading conditions, and those affecting rural poverty, to issues concerning thestatus and future of agricultural resources and technology. Of the many issuesre-viewed, the report concludes that the development of local food productionin the low-income countries with high dependence on agriculture for employ-ment and income is the one factor that dominates all others in determiningprogress or failure in improving their food security.

The findings of the study aim to describe the future as it is likely to be, notas it ought to be. As such they should not be construed to represent goals ofan FAO strategy. But the findings can make a vital contribution to an increasedawareness of what needs to be done to cope with the problems likely to persistand to deal with new ones as they emerge. It can help to guide correctivepolicies at both national and international levels, and to set priorities for theyears ahead.

The world as a whole has been making progress towards improved foodsecurity and nutrition. This is clear from the substantial increases in per capitafood supplies achieved globally and for a large proportion of the population ofthe developing world. But, as the 1995 study had warned, progress was slowand uneven. Indeed, many countries and population groups failed to makesignificant progress and some of them even suffered setbacks in their alreadyfragile food security and nutrition situation. As noted in the latest (2001) issueof The State of Food Insecurity in the World, humanity is still faced with thestark reality of chronic undernourishment affecting over 800 million people:17 percent of the population of the developing countries, as many as 34 per-cent in sub-Saharan Africa and still more in some individual countries.

The present study predicts that this uneven path of progress is, unfortunate-ly, likely to extend well into this century. It indicates that in spite of somesignificant enhancements in food security and nutrition by the year 2015,mainly resulting from increased domestic production but also from additionalgrowth in food imports, the World Food Summit target of halving the numberof undernourished persons by no later than 2015 is far from being reached,and may not be accomplished even by 2030.

By the year 2015 per capita food supplies will have increased and theincidence of undernourishment will have been further reduced in mostdeveloping regions. However, parts of South Asia may still be in a difficultposition and much of sub-Saharan Africa will probably not be significantly

iv

better and may possibly be even worse off than at present in the absence ofconcerted action by all concerned. Therefore, the world must brace itself forcontinuing interventions to cope with the consequences of local food crisesand for action to remove permanently their root causes. Nothing short of asignificant upgrading of the overall development performance of the laggingcountries, with emphasis on hunger and poverty reduction, will free the worldof the most pressing food insecurity problems. Making progress towards thisgoal depends on many factors, not least among which the political will andadditional resource mobilization required. Past experience underlines thecrucial role of agriculture in the process of overall national development,particularly where a large part of the population depends on the sector foremployment and income.

The study also foresees that agricultural trade will play a larger role insecuring the food needs of developing countries as well as being a sourceof foreign exchange. Net cereal imports by developing countries will almosttriple over the next 30 years while their net meat imports might even increaseby a factor of almost five. For other products such as sugar, coffee, fruits andvegetables the study foresees further export potential. How much of thisexport potential will materialize depends on many factors, not least on howmuch progress will be made during the ongoing round of multilateral tradenegotiations. Developing countries’ farmers could gain a lot from lowertrade barriers in all areas, not only in agriculture. In many resource-rich butotherwise poor countries, a more export-oriented agriculture could providean effective means to fight rural poverty and thus become a catalyst foroverall growth. But the study also points at potentially large hardships forresource-poor countries which may face higher prices for large importvolumes without much capacity to step-up production.

Numerous studies that assessed the impacts of freer trade conclude thatlower trade barriers alone may not be sufficient for developing countries tobenefit. In many developing countries, agriculture has suffered not only fromtrade barriers and subsidies abroad but has also been neglected by domesticpolicies. Developing countries’ producers may therefore not benefit greatlyfrom freer trade unless they can operate in an economic environment thatenables them to respond to the incentives of higher and more stable inter-national prices. A number of companion policies implemented alongside themeasures to lower trade barriers can help. These include a removal ofthe domestic bias against agriculture; investment to lift product quality tothe standards demanded abroad; and efforts to improve productivity andcompetitiveness in all markets. Investments in transportation and communi-cations facilities, upgraded production infrastructure, improved marketing,storage and processing facilities as well as better food quality and safetyschemes could be particularly important, the latter not only for the benefit ofbetter access to export markets, but also for reducing food-borne diseasesaffecting the local population.

On the issue of sustainability, the study brings together the most recentevaluation of data on the developing countries’ agricultural resources, howthey are used now and what may be available for meeting future needs. Itdoes the same for the forestry and the fisheries sectors. The study providesan assessment of the possible extent and intensity of use of resources over

v

the years to 2030 and concludes that pressure on resources, including thosethat are associated with degradation, will continue to build up albeit at aslower rate than in the past.

The main pressures threatening sustainability are likely to be thoseemanating from rural poverty, as more and more people attempt to extracta living out of dwindling resources. When these processes occur in anenvironment of fragile and limited resources and when the circumstancesfor introducing sustainable technologies and practices are not propitious,the risk grows that a vicious circle of poverty and resource degradation willset in. The poverty-related component of environmental degradation is un-likely to be eased before poverty reduction has advanced to the level wherepeople and countries become significantly less dependent on the exploitationof agricultural resources. There is considerable scope for improvements inthis direction and the study explores a range of technological and other policyoptions. Provided such improvements in sustainability are put in place, theprospects point to an easing of pressures on world agricultural resources inthe longer term with minimal further buildup of pressures on the environ-ment caused by agricultural practices.

I conclude by reiterating the importance of developing sustainable localfood production and of rural development in the low-income countries. Mostof them depend highly on agriculture for employment and income as animportant and, often, the critical component of any strategy to improve theirlevels of food security and alleviate poverty. It is for this reason that sustain-able agricultural and rural development is given enhanced priority in TheStrategic Framework for FAO: 2000–2015.

Jacques DioufDirector-GeneralFood and Agriculture Organizationof the United Nations

Contents

Acknowledgements viiiAbout this report ixExecutive summary 1The projections at a glance 9

Long-term perspectives 11The outlook for agriculture 11Prospects for food and nutrition 14

Food and agriculture in national and international settings 21Poverty and agriculture 21International trade and globalization 24

Prospects by major sector 32Crop production 32Land, water and crop yields 38The role of technology 49Livestock: intensification and its risks 58Towards sustainable forestry 64World fisheries: a choice of futures 69

Prospects for the environment 75Agriculture and the environment 75Agriculture and climate change 78

Annex 1: Countries and commodities covered 83Annex 2: Statistical tables 86Sources 96Acronyms 97

viii

This summary document, which is derived from a full technical report underthe same title, was prepared mainly by Paul Harrison. The full report is theproduct of cooperative work by most technical units of FAO. It was preparedby a team led by Jelle Bruinsma under the general direction of Hartwig deHaen, Assistant Director-General, Economic and Social Department. Membersof the core team were Nikos Alexandratos, Josef Schmidhuber, GeroldBödeker and Maria-Grazia Ottaviani.

In addition to the contributions made by the members of the core team,the following FAO staff members and consultants (in alphabetical order)made technical contributions and drafted sections or chapters of the fullreport: Clare Bishop, Giacomo Branca, Robert Brinkman, Sumiter Broca,Concha Calpe, Lawrence Clarke, Jean-Marc Faurès, Günther Fischer, TheodorFriedrich, René Gommes, Ali Gürkan, David Hallam, Jippe Hoogeveen, SimonMack, Michael Martin, Jorge Mernies, Rebecca Metzner, Miles Mielke, NancyMorgan, Freddy Nachtergaele, Loganaden Naiken, CTS Nair, Nquu Nguyen,David Norse, Joachim Otte, Jan Poulisse, Terri Raney, Nadia Scialabba, KostasStamoulis, Henning Steinfeld, Peter Thoenes, Vivian Timon, Bruce Traill, DatTran, Jeff Tschirley, David Vanzetti, Ulf Wijkstrom and Alberto Zezza. Thenature of their contributions is specified in the Acknowledgements of the fullreport.

Members of the Working Group for the FAO Priority Area for InterdisciplinaryAction on Global Perspective Studies provided comments on the various drafts.

Green Ink Ltd, under the direction of Simon Chater, undertook the editing,layout and artwork. FAO’s Publishing and Multimedia Service was responsible for the final editing and printing.

Acknowledgements

ix

This report is a shorter account of the results of the FAO study, Agriculture:towards 2015/30. It presents the latest FAO assessment of long-termdevelopments in world food, nutrition and agriculture, including the forestryand fisheries sectors. It is the product of an interdisciplinary exercise, involvingmost of FAO’s technical units and disciplines, and continues the tradition ofFAO’s periodical perspective studies for global agriculture, the latest of whichwas published in 1995 (Alexandratos, 1995). Earlier editions were Alexandratos(1988), FAO (1981) and FAO (1970).

The projections were carried out in considerable detail, covering about 140countries and 32 crop and livestock commodities (see Annex 1). For nearly allthe developing countries, the main factors contributing to the growth ofagricultural production were identified and analysed separately. Sources ofproductivity growth, such as higher crop yields and livestock carcass weights,were distinguished from other growth resources, such as the area of cultivatedland and the sizes of livestock herds. Special attention was given to land,which was broken down into five classes for rainfed agriculture and a sixthfor irrigated agriculture. This level of detail proved both necessary andadvantageous in identifying the main issues likely to emerge for worldagriculture over the next 30 years. Specifically, it helped to spot localproduction and resource constraints, to gauge country-specific requirementsfor food imports and to assess progress and failure in the fight against hungerand undernourishment. The high degree of detail was also necessary forintegrating the expertise of FAO specialists from various disciplines, as theanalysis drew heavily on the judgement of in-house experts. Owing to spaceand other constraints, the results are, however, mainly presented at theaggregate regional and sectoral levels, which can mask diverging developmentsbetween individual countries and commodities. Likewise, space considerationsmilitated against the inclusion of references to the numerous sources drawnupon in this report. References have therefore been limited to statisticalsources and the sources of figures, tables and maps. These are given on p. 96.A complete list of references is provided in the main technical report.

Another important feature of this report is that its approach is “positive”rather than “normative”. This means that its assumptions and projectionsreflect the most likely future but not necessarily the most desirable one. Forexample, the report finds that the goal of the 1996 World Food Summit — tohalve the number of chronically undernourished people by no later than 2015— is unlikely to be accomplished, even though this would be highly desirable.Similarly, the report finds that agriculture will probably continue to expand intowetlands and rainforests, even though this is undoubtedly undesirable. Ingeneral, the projections presented are therefore not goals of an FAO strategybut rather a basis for action, to cope both with existing problems that are likelyto persist and with new ones that may emerge. It should also be stressed thatthese projections are certainly not trend extrapolations. Instead, they in-corporate a multitude of assumptions about the future and often representsignificant deviations from past trends.

About this report

x

A long-term assessment of world food, nutrition and agriculture could dealwith a great number of issues, the relevance of which depends on the reader’sinterest in a particular country, region or topic. As a global study, however,this report had to be selective in the issues it addresses. The main focus is onhow the world will feed itself in the future and what the need to produce morefood means for its natural resource base. The base year for the study is thethree-year average for 1997-99 and projections are made for the years 2015and 2030. The choice of 2015 allows assessment of whether or not the goal ofthe 1996 World Food Summit — to halve the number of chronically under-nourished people — is likely to be reached. Extending the horizon to 2030creates a sufficiently long period for the analysis of issues pertaining tothe world’s resource base — in other words, the world’s ability to cope withfurther degradation of agricultural land, desertification, deforestation, globalwarming and water scarcity, as well as increasing demographic pressure.Naturally, the degree of uncertainty increases as the time horizon is extended,so the results envisaged for 2030 should be interpreted more cautiously thanthose for 2015.

The analysis is, inter alia, based on the long-term developments expectedby other organizations. The population projections, for instance, reflect thelatest assessment (2000 Assessment, Medium Variant) available from theUnited Nations (UN, 2001), while those for incomes are largely based on thelatest projections of gross domestic product (GDP) from the World Bank. Mostof the agricultural data are from FAO’s database (FAOSTAT), as in July 2001.Because these assumptions critically shape the projected outcomes, it isimportant to note that they can change significantly, even over the short term.For example, the historical data and the projections for the growth ofpopulation and GDP used in the 1995 study have since been revised in manycountries, often to a significant extent. World population in the 1995 study,for instance, was projected at 7.2 billion for 2010, whereas the current UNprojections peg the figure at 6.8 billion. Similarly, it is now assumed thatsub-Saharan Africa’s population will reach 780 million by 2010, comparedwith 915 million in the 1995 study. The GDP projections for sub-SaharanAfrica also differ from those assumed in the 1995 study: per capita incomegrowth over the period 1997-99 to 2015 is now projected at 1.8 percent a year,compared with 0.7 percent in the 1995 study (over the period 1988-90 to 2010).Finally, FAO’s historical data for food production, demand and per capitaconsumption were often drastically revised for the entire time series asmore up-to-date information became available.

This report begins by presenting the expected developments in world agri-cultural demand, production and trade (both in total and by major commoditygroup), and the implications for food security and undernourishment. Itcontinues with a discussion of the main issues raised by these developments.These include the role of agriculture in rural development, poverty alleviationand overall economic growth, and the effects of globalization and freer trade.The report then discusses production and policy issues in the crop, livestock,forestry and fisheries sectors, including natural resource use and agriculturaltechnology issues. It concludes with an assessment of the environmentalimplications of agricultural production, including its interactions with climatechange.

1

Executive summary

In recent years the growth rates of world agricultural production and cropyields have slowed. This has raised fears that the world may not be able togrow enough food and other commodities to ensure that future populationsare adequately fed.

However, the slowdown has occurred not because of shortages of land orwater but rather because demand for agricultural products has also slowed.This is mainly because world population growth rates have been decliningsince the late 1960s, and fairly high levels of food consumption per person arenow being reached in many countries, beyond which further rises will belimited. But it is also the case that a stubbornly high share of the world’spopulation remains in absolute poverty and so lacks the necessary income totranslate its needs into effective demand.

As a result, the growth in world demand for agricultural products isexpected to fall from an average 2.2 percent a year over the past 30 years to1.5 percent a year for the next 30. In developing countries the slowdown willbe more dramatic, from 3.7 percent to 2 percent, partly as a result of Chinahaving passed the phase of rapid growth in its demand for food.

This study suggests that world agricultural production can grow in linewith demand, provided that the necessary national and international policiesto promote agriculture are put in place. Global shortages are unlikely, butserious problems already exist at national and local levels and may worsenunless focused efforts are made.

Food and nutritionMassive strides have been made in improving food security. The proportionof people living in developing countries with average food intakes below 2 200kcal per day fell from 57 percent in 1964-66 to just 10 percent in 1997-99. Yet776 million people in developing countries remain undernourished — aboutone person in six.

Global progress in nutrition is expected to continue, in parallel with areduction in poverty as projected by the World Bank. The incidence of under-nourishment should fall from 17 percent of the population of developingcountries at present to 11 percent in 2015 and just 6 percent in 2030. By 2030,three-quarters of the population of the developing world could be living incountries where less than 5 percent of people are undernourished. Less than8 percent live in such countries at present.

Despite impressive reductions in the proportion of undernourished,continuing population growth means that progress in reducing the total

2

number will be slower. The World Food Summit of 1996 set a target ofhalving the number of undernourished people to about 410 million by 2015.This study’s projections suggest that this may be difficult to achieve: some610 million people could still be undernourished in that year, and even by2030 about 440 million undernourished may remain. Priority for local foodproduction and reduced inequality of access to food could improve thisperformance. The problem of undernourishment will tend to become moretractable and easier to address through policy interventions, both nationaland international, as the number of countries with high incidence declines.

Agriculture, poverty and international tradeUndernourishment is a central manifestation of poverty. It also deepensother aspects of poverty, by reducing the capacity for work and resistanceto disease, and by affecting children’s mental development and educationalachievements.

Currently, one in four people in developing countries are living in extremepoverty, subsisting on less than US$1 a day. This proportion is down fromalmost one-third in 1990. But because of population growth the fall innumbers has been slower, from 1269 million to 1134 million. The latestWorld Bank assessment to 2015 suggests that such reductions in globalpoverty could continue. Sub-Saharan Africa is the exception, however. Herethe numbers of poor rose steeply during the 1990s and seem likely tocontinue to do so. Seven out of ten of the world’s poor still live in rural areas.Growth in the agricultural sector has a crucial role to play in improving theincomes of poor people, by providing farm jobs and stimulating off-farmemployment. Some direct nutritional interventions may also be needed —such as vitamin and mineral supplementation of basic foods — while health,water and sanitation measures to reduce the effects of illness on foodabsorption will also be important.

Trade has an important role to play in improving food security andfostering agriculture. Some estimates put the potential annual increasein global welfare from freer trade in agriculture as high as US$165 billion. Butthe progress made in the current round of trade negotiations has been limitedand the benefits so far remain modest. If future reforms focus too narrowly onthe removal of subsidies in the countries of the Organisation for EconomicCo-operation and Development (OECD), most of the gains will probably bereaped by consumers in developed countries. Developing countries shouldbenefit more from the removal of trade barriers for products in which theyhave a comparative advantage (such as sugar, fruits and vegetables), fromreduced tariffs for processed agricultural commodities, and from deeperpreferential access to markets for the least developed countries.

Internal reforms are also needed within developing countries if freetrade is to contribute to poverty reduction. Such reforms include: a reductionof the bias against agriculture in national policy making; the opening ofborders for long-term foreign investments; the introduction of schemes toimprove food quality and safety; investments in roads, irrigation, seeds andskills; improved quality standards; and safety nets for the poor who facehigher food prices.

3

Globalization in food and agriculture holds promise as well as presentingproblems. It has generally led to progress in reducing poverty in Asia. But ithas also led to the rise of multinational food companies with the potentialto disempower farmers in many countries. Developing countries need thelegal and administrative frameworks to ward off the threats while reapingthe benefits.

Crop productionThe annual growth rate of world demand for cereals has declined from2.5 percent a year in the 1970s and 1.9 percent a year in the 1980s to only1 percent a year in the 1990s. Annual cereal use per person (includinganimal feeds) peaked in the mid-1980s at 334 kg and has since fallen to317 kg.

The decline is not cause for alarm: it was above all the natural result ofslower population growth and shifts in human diets and animal feeds.However, in the 1990s it was accentuated by a number of temporary factors,including serious recessions in the transition countries and some East andSoutheast Asian countries.

The growth rate of demand for cereals is expected to rise again to1.4 percent a year to 2015, slowing to 1.2 percent per year thereafter. Indeveloping countries overall, cereal production is not expected to keep pacewith demand. The net cereal deficits of these countries, which amounted to103 million tonnes or 9 percent of consumption in 1997-99, could rise to265 million tonnes by 2030, when they will be 14 percent of consumption. Thisgap can be bridged by increased surpluses from traditional grain exporters,and by new exports from the transition countries, which are expected to shiftfrom being net importers to being net exporters.

Oilcrops have seen the fastest increase in area of any crop sector, ex-panding by 75 million ha from the mid-1970s until the end of the 1990s,while cereal area fell by 28 million ha over the same period. Future per capitaconsumption of oilcrops is expected to rise more rapidly than that of cereals.These crops will account for 45 out of every 100 extra kilocalories added toaverage diets in developing countries between now and 2030.

Sources of growth in crop productionThere are three main sources of growth in crop production: expanding the landarea, increasing the frequency with which it is cropped (often through irrigation),and boosting yields. It has been suggested that we may be approaching the ceilingof what is possible for all three sources.

A detailed examination of production potentials does not support this view atthe global level, although in some countries, and even in whole regions, seriousproblems already exist and could deepen.

Land. Less new agricultural land will be opened up than in the past. In thecoming 30 years, developing countries will need an extra 120 million ha forcrops, an overall increase of 12.5 percent. This is only half the rate of increaseobserved between 1961-63 and 1997-99.

4

At global level there is adequate unused potential farmland. A comparisonof soils, terrains and climates with the needs of major crops suggests that anextra 2.8 billion ha are suitable in varying degrees for the rainfed productionof arable and permanent crops. This is almost twice as much as is currentlyfarmed. However, only a fraction of this extra land is realistically availablefor agricultural expansion in the foreseeable future, as much is needed topreserve forest cover and to support infrastructural development. Access-ibility and other constraints also stand in the way of any substantial expansion.

More than half the land that could be opened up is in just seven countriesof tropical Latin America and sub-Saharan Africa, whereas other regionsand countries face a shortage of suitable land. In the Near East and NorthAfrica, 87 percent of suitable land was already being farmed in 1997-99,while in South Asia the figure is no less than 94 percent. In these regions,intensification through improved management and technologies will be themain, indeed virtually the only, source of production growth. In many placesland degradation threatens the productivity of existing farmland and pasture.

Water. Irrigation is crucial to the world’s food supplies. In 1997-99, irrigatedland made up only about one-fifth of the total arable area in developingcountries but produced two-fifths of all crops and close to three-fifths ofcereal production.

The role of irrigation is expected to increase still further. The developingcountries as a whole are likely to expand their irrigated area from 202 millionha in 1997-99 to 242 million ha by 2030. Most of this expansion will occur inland-scarce areas where irrigation is already crucial.

The net increase in irrigated land is predicted to be less than 40 percent ofthat achieved since the early 1960s. There appears to be enough unusedirrigable land to meet future needs: FAO studies suggest a total irrigationpotential of some 402 million ha in developing countries, of which only half iscurrently in use. However, water resources will be a major factor constrainingexpansion in South Asia, which will be using 41 percent of its renewablefreshwater resources by 2030, and in the Near East and North Africa, whichwill be using 58 percent. These regions will need to achieve greater efficiencyin water use.

Yields. In the past four decades, rising yields accounted for about 70 percentof the increase in crop production in the developing countries. The 1990s sawa slowdown in the growth of yields. Wheat yields, for example, grew at anaverage 3.8 percent a year between 1961 and 1989, but at only 2 percent ayear between 1989 and 1999. For rice the respective rates fell by more thanhalf, from 2.3 percent to 1.1 percent.

Yield growth will continue to be the dominant factor underlying increasesin crop production in the future. In developing countries, it will account forabout 70 percent of growth in crop production to 2030. To meet productionprojections, future yield growth will not need to be as rapid as in the past. Forwheat yields, an annual rise of only 1.2 percent a year is needed over the next30 years. The picture for other crops is similar. Growth in fertilizer use indeveloping countries is expected to slow to 1.1 percent per year over the nextthree decades, a continuation of the slowdown already under way.

5

Overall, it is estimated that some 80 percent of future increases in cropproduction in developing countries will have to come from intensification:higher yields, increased multiple cropping and shorter fallow periods.

Improved technologyNew technology is needed for areas with shortages of land or water, or withparticular problems of soil or climate. These are frequently areas with a highconcentration of poor people, where such technology could play a key role inimproving food security.

Agricultural production could probably meet expected demand over theperiod to 2030 even without major advances in modern biotechnology.However, the new techniques of molecular analysis could give a welcomeboost to productivity, particularly in areas with special difficulties, therebyimproving the incomes of the poor, just as the green revolution did in largeparts of Asia during the 1960s to 1980s.

Needed for the twenty-first century is a second, doubly green revolutionin agricultural technology. Productivity increases are still vital, but must becombined with environmental protection or restoration, while new technolo-gies must be both affordable by, and geared to the needs of, the poor andundernourished.

Biotechnology offers promise as a means of improving food security andreducing pressures on the environment, provided the perceived environ-mental threats from biotechnology itself are addressed. Genetically modifiedcrop varieties — resistant to drought, waterlogging, soil acidity, salinity andextreme temperatures — could help to sustain farming in marginal areas andto restore degraded lands to production. Pest-resistant varieties can reducethe need for pesticides.

However, the widespread use of genetically modified varieties will dependon whether or not food safety and environmental concerns can be adequatelyaddressed. Indeed, the spread of these varieties, in the developed countriesat least, has recently slowed somewhat in response to these concerns, whichmust be addressed through improved testing and safety protocols if progressis to resume.

Meanwhile, other promising technologies have emerged that combine in-creased production with improved environmental protection. These includeno-till or conservation agriculture, and the lower-input approaches of inte-grated pest or nutrient management and organic agriculture.

LivestockDiets in developing countries are changing as incomes rise. The share ofstaples, such as cereals, roots and tubers, is declining, while that of meat,dairy products and oilcrops is rising.

Between 1964-66 and 1997-99, per capita meat consumption in develop-ing countries rose by 150 percent, and that of milk and dairy products by60 percent. By 2030, per capita consumption of livestock products couldrise by a further 44 percent. As in the past, poultry consumption will growfastest.

6

Productivity improvements are likely to be a major source of growth.Milk yields should improve, while breeding and improved management willincrease average carcass weights and offtake rates. This will allow increasedproduction with lower growth in animal numbers, and a correspondingslowdown in the growth of environmental damage from grazing or wastes.

In developing countries, demand will grow faster than production,producing a growing trade deficit. In meat products this will rise steeply,from 1.2 million tonnes a year in 1997-99 to 5.9 million tonnes in 2030 (despitegrowing meat exports from Latin America), while in milk and dairy productsthe rise will be less steep but still considerable, from 20 million to 39 milliontonnes a year.

An increasing share of livestock production will probably come fromindustrial enterprises. In recent years production from this sector has growntwice as fast as that from more traditional mixed farming systems and morethan six times faster than from grazing systems.

ForestryDuring the 1990s the world’s total forest area shrank by 9.4 million ha —about three times the size of Belgium — each year. However, the rate ofdeforestation was slower in the 1990s than in the 1980s. Industrial andtransition countries expanded their forest areas, and many developingcountries — including Bangladesh, China, India, Turkey and Viet Nam —are now planting more forest area than they cut.

The crop projections suggest that cropland will need to expand by anextra 120 million ha by 2030, while urban land areas will continue to grow bya considerable amount. Much of this extra land will have to come from forestclearance. In addition, by 2030 annual world consumption of industrialroundwood is expected to rise by 60 percent over current levels, to around2 400 million m3.

Even so, deforestation is expected to slow further in the coming decadesand the world is unlikely to face a wood supply crisis. Production of wood-based materials is continually increasing in efficiency. The area of plantationsis also growing rapidly: production of industrial roundwood from plantations isexpected to double by 2030, from 400 million m3 today, to around 800 millionm3. In addition, a big increase in tree-growing outside forests and plantations— along roads, in towns, around homes and on farms — will boost the supplyof wood and other tree products.

The central challenges for the forestry sector are to find ways of managingnatural and cultivated tree resources so as to increase production, improvethe food security and energy supply of the poor, and safeguard the environ-mental services and biodiversity provided by forests.

FisheriesWorld fisheries production has kept ahead of population growth over the pastthree decades. Total fish production almost doubled, from 65 million tonnesin 1970 to 125 million tonnes in 1999, when world average intake of fish,crustaceans and molluscs reached 16.3 kg per person. By 2030, annual fish

7

consumption is likely to rise to some 150 to160 million tonnes, or between19 and 20 kg per person.

This amount is significantly lower than the potential demand, becauseenvironmental factors are expected to limit supply. By the turn of the century,three-quarters of ocean fish stocks were overfished, depleted or exploitedup to their maximum sustainable yield. Further growth in the marine catchcan be only modest. During the 1990s the marine catch levelled out at 80 to85 million tonnes a year, not far from its maximum sustainable yield.

Aquaculture compensated for this marine slowdown, doubling its shareof world fish production during the 1990s. It will continue to grow rapidly,at rates of 5 to 7 percent a year up to 2015. In all sectors of fishing it will beessential to pursue forms of management conducive to sustainableexploitation, especially for resources under common ownership or noownership.

Environment and climateOver the next 30 years, many of the environmental problems associated withagriculture will remain serious. Loss of biodiversity caused by the expansionand intensification of production often continues unabated even in thedeveloped countries, where nature is highly valued and, supposedly,protected.

Nitrogen fertilizers are a major source of water and air pollution. Thecrop projections imply slower growth in the use of these fertilizers than inthe past, but the increase could still be significant for pollution. Projectionsalso suggest a 60 percent increase in emissions of ammonia and methanefrom the livestock sector. Comprehensive measures will be needed to controland reduce air and water pollution from these sources.

Global warming is not expected to depress food availability at the globallevel, but at the regional and local levels there may be significant impacts.Current projections suggest that the potential for crop production willincrease in temperate and northerly latitudes, while in parts of the tropicsand subtropics it may decline. This may further deepen the dependence ofdeveloping countries on food imports, though at the same time it mayimprove the ability of temperate exporters to fill the gap. Rising sea levelswill threaten crop production and livelihoods in countries with large areasof low-lying land, such as Bangladesh and Egypt.

Food insecurity for some vulnerable rural groups in developing countriesmay well worsen. By 2030, climate change is projected to depress cerealproduction in Africa by 2 to 3 percent. Improved seeds and increased fertilizeruse should more than compensate, but this factor will still weigh heavily onefforts to make progress.

Forestry and agriculture both contribute to human impact on climate.The burning of biomass — in deforestation, savannah fires, the disposalof crop residues and cooking with firewood or dung — is a major source ofatmospheric carbon dioxide, while fertilizers and animal wastes create largeemissions of nitrous oxide and ammonia.

Forestry can help to soak up some of the carbon released by humanactivities. Between 1995 and 2050, slower deforestation, together with

8

regeneration and plantation development, could reduce carbon dioxideemissions by the equivalent of 12 to 15 percent of all fossil fuel emissions.

Farming also has a positive role to play. By 2030 the amount of carbonlocked up in cropland soils, as soil organic matter from crop residues andmanure, could rise by 50 percent if better management practices areintroduced.

9

Population 1979-81 1997-99 2015 2030 2050(millions)

World 4 430 5 900 7 207 8 270 9 322Developing countries 3 259 4 595 5 858 6 910 7 987Industrial countries 789 892 951 979 986Transition countries 382 413 398 381 349

Population growth 1979 to 1999 1989 to 1999 1997-99 to 2015 2015 to 2030 2030 to 2050(% per annum)

World 1.6 1.5 1.2 0.9 0.6Developing countries 1.9 1.7 1.4 1.1 0.7Industrial countries 0.7 0.7 0.4 0.2 0.0Transition countries 0.5 0.1 - 0.2 - 0.3 - 0.4

GDP growth 1997-99 to 2015 2015 to 2030 1997-99 to 2015 2015 to 2030(% per annum) total total per capita per capita

World 3.5 3.8 2.3 2.9Developing countries 5.1 5.5 3.7 4.4Industrial countries 3.0 3.0 2.6 2.8Transition countries 3.7 4.0 4.0 4.3

Growth in demand foragricultural products 1969 to 1999 1979 to 1999 1989 to 1999 1997-99 to 2015 2015 to 2030(% per annum)

World 2.2 2.1 2.0 1.6 1.4Developing countries 3.7 3.7 4.0 2.2 1.7Industrial countries 1.1 1.0 1.0 0.7 0.6Transition countries - 0.2 - 1.7 - 4.4 0.5 0.4

Growth in agriculturalproduction (% per annum) 1969 to 1999 1979 to 1999 1989 to 1999 1997-99 to 2015 2015 to 2030

World 2.2 2.1 2.0 1.6 1.3Developing countries 3.5 3.7 3.9 2.0 1.7Industrial countries 1.3 1.0 1.4 0.8 0.6Transition countries - 0.4 - 1.7 - 4.7 0.6 0.6

Calorie consumption 1961-63 1979-81 1997-99 2015 2030 (kcal/capita/day)

World 2 283 2 552 2 803 2 940 3 050Developing countries 1 960 2 312 2 681 2 850 2 980Industrial countries 2 891 3 135 3 380 3 440 3 500Transition countries 3 154 3 389 2 906 3 060 3 180

Undernourishment Million people % of population1990-92 1997-99 2015 2030 1990-92 1997-99 2015 2030

World 815 14Developing countries 816 777 610 443 20 17 11 6Industrial countries 11 1Transition countries 27 6

The projections at a glance

10

Cereals Million tonnes % per annum1979-81 1997-99 2015 2030 1979 to 1989 to 1997-99 2015 to

1999 1999 to 2015 2030World

Production 1 442 1 889 2 387 2 838 1.4 1.0 1.4 1.2Food 706 1 003 1 227 1 406 1.9 1.4 1.2 0.9Feed 575 657 911 1 148 0.6 0.6 1.9 1.5

Developing countriesProduction 649 1 026 1 354 1 652 2.5 2.1 1.6 1.3Food 524 790 1 007 1 185 2.2 1.7 1.4 1.1Feed 113 222 397 573 3.8 4.4 3.5 2.5Net trade - 66 - 103 - 190 - 265

Meat Million tonnes % per annum1979-81 1997-99 2015 2030 1979 to 1989 to 1997-99 2015 to

1999 1999 to 2015 2030World

Production 132 218 300 376 2.8 2.7 1.9 1.5Food 130 214 297 373 2.8 2.7 1.9 1.5

Developing countriesProduction 45 116 181 247 5.5 5.9 2.7 2.1Food 44 116 184 252 5.6 6.1 2.7 2.1Net trade - 0.2 - 1.2 - 3.9 - 5.9

Vegetable oils and oilseeds Million tonnes % per annum(oil equivalent) 1979-81 1997-99 2015 2030 1979 to 1989 to 1997-99 2015 to

1999 1999 to 2015 2030

WorldProduction 50 104 157 217 4.1 4.3 2.5 2.2Food 37 67 98 130 3.3 2.8 2.3 1.9Industrial use 8 23 45 71 6.1 6.9 3.9 3.1

Developing countriesProduction 29 68 109 156 5.0 4.7 2.8 2.4Food 21 45 73 102 4.3 3.6 2.9 2.2Industrial use 3 13 26 41 8.2 10.2 4.4 3.1Net trade 1.5 4.0 3.4 3.5

Arable land Total Irrigated(million ha) 1997-99 2015 2030 1979-81 1997-99 2015 2030

World 1 608 210 271Developing countries 956 1017 1076 151 202 221 242Industrial countries 387 37 42Transition countries 265 22 25

Crop land and yields Harvested land (million ha) Yield (tonnes/ha)in developing countries 1979-81 1997-99 2015 2030 1979-81 1997-99 2015 2030

Wheat 96 111 113 118 1.6 2.5 3.1 3.5Rice (paddy) 138 157 162 164 2.7 3.6 4.2 4.7Maize 76 97 118 136 2.0 2.8 3.4 4.0All cereals 408 465 497 528 1.9 2.6 3.2 3.6% of total 60 55 53 51

The projections at a glance (continued)

11

Long-term Perspectives

The outlook for agriculture

So far, world agriculture has been able torespond to the rising demand for crop andlivestock products. Although the world’spopulation doubled between 1960 and 2000 andlevels of nutrition improved markedly, the pricesof rice, wheat and maize — the world’s majorfood staples — fell by around 60 percent. The fallin prices indicates that, globally, supplies not onlykept pace with demand, but even outstripped it.

Although global demand for agriculturalproducts has continued to rise, it has doneso less rapidly in recent decades. Between1969 and 1989 demand grew at an average of2.4 percent a year, but this fell to only 2 percentin the decade from 1989.

The rate of growth in world demandfor agricultural products has slowed,because population growth hasdeclined and fairly high levels of foodconsumption have been reached inmany countries. Growth in demandwill slow still further in the future.The world as a whole has theproduction potential to cope withdemand. However, developingcountries will become moredependent on agricultural imports,and food security in many poor areaswill not improve without substantialincreases in local production.

Apart from temporary factors (foremostamong them a decline in consumption in thetransition economies in the 1990s), there weretwo more enduring reasons for the slowdown:• The growth rate of world population peaked in

the late 1960s at 2 percent a year and slowedthereafter.

• A rising proportion of the world’s populationhad reached fairly high levels of foodconsumption, so the scope for furtherincrease was limited. By 1997-99, 61 percentof the world’s population were living incountries where average food consumptionper person was above 2 700 kcal per day.

Demand for agricultural products will continueto grow more slowlyThese factors will continue to influence trendsin demand over the next three decades. Forexample, world population will go on rising,but less rapidly, growing at an average of1.1 percent a year up to 2030, compared with1.7 percent a year over the past 30 years.

As a result, future demand for agriculturalproducts is expected to slow further — to1.6 percent a year for the period 1997-99 to2015 and to 1.4 percent for 2015 to 2030. Indeveloping countries the slowdown will be moredramatic, from 3.7 percent for the past 30 yearsto an average 2 percent for the next 30.

The forces underlying this slowdown can beseen in the example of China, which has beenone of the major engines of growth in thedemand for food and agricultural products inthe world and in the developing countries overthe past few decades. By 1997-99 the Chinese

12

World market prices for agricultural commodities, 1960 to 2000

Source: World Bank (2001a)

had reached an average daily food consumptionof 3 040 kcal — only 10 percent short of the levelin industrial countries. Over the next threedecades the country’s aggregate foodconsumption is expected to grow at only aquarter of the rate seen in the past threedecades, while its population will grow at a thirdof its past rate. Given the sheer size of China’spopulation, these shifts alone will have a hugeeffect on the global situation. Many othercountries, including some of the largest ones,will be undergoing very similar shifts that willfurther lower the growth of demand.

India’s daily average food energy intake perperson is still below 2 500 kcal, a level at whichthere is considerable scope for furtherincreases, while her population will be growingat an average of over 1 percent a year over thenext 30 years. Could India take over China’s roleas a major engine of growth in worldagricultural demand? This is not expected,because India’s cultural traditions favour

vegetarianism, which will hold back thecountry’s demand for meat and animal feeds atrates well below those seen in China.

Agricultural trade deficits of developingcountries will worsenTraditionally, the developing countries as awhole have had a net surplus in agriculturaltrade. In value terms this peaked at US$17.5billion in 1977. The trend since then has beenfor their imports to grow faster than theirexports. The agricultural trade balance of thedeveloping countries has gradually dwindleduntil, by the mid-1990s, it was more oftennegative than positive. The highest recordeddeficit was US$6 billion, in 1996.

This overall trend masks a complex picturewhich varies from one commodity to anotherand from one country to another. The drasticdecline in developing countries’ net surplus insugar, oilseeds and vegetable oils, for example,reflects growing consumption and imports inseveral developing countries and the effects ofprotectionist policies in the major industrialcountries. For commodities produced almostentirely in developing countries and consumedpredominantly in the industrial countries, suchas coffee and cocoa, slow growth in demandprevented the trade balance of the developing

By 1997-99, 61 percent of the world’s population were livingin countries where average food consumption per personwas above 2 700 kcal per day.

1 400

1 200

1 000

800

600

400

200

0

280

240

200

160

120

80

40

01960 1965 1970 1975 1980 1985 1990 1995 2000

Inde

x, 1

990

= 10

0

Year

Con

stan

t 199

0 U

S$/t

onne

Agriculture (index)Food (index)Maize (US$/tonne)Palm oil (US$/tonne)Rice (US$/tonne)Soybean (US$/tonne)Wheat (US$/tonne)

13

Net agricultural trade balance of developing countries, 1984 to 1999

Source: FAO data

countries from improving. Fluctuating and, onbalance, declining prices further contributed tothe problem.

The projections to 2030 show the agriculturaltrade deficit of developing countries growing stillfurther. In particular, net imports of cereals andlivestock products will continue to rise quiterapidly.

Production will keep pace with demand, butfood insecurity will persistDetailed analysis shows that, globally, there isenough land, soil and water, and enough

potential for further growth in yields, to makethe necessary production feasible. Yield growthwill be slower than in the past, but at the globallevel this is not necessarily cause for alarmbecause slower growth in production is neededin the future than in the past. However, thefeasible can only become the actual if the policyenvironment is favourable towards agriculture.

Globally, producers have satisfied effectivemarket demand in the past, and there is everylikelihood that they will continue to do so. Buteffective demand does not represent the totalneed for food and other agricultural products,

Other crop and livestock products

Meat (incl. live animals) anddairy products

Cereals and cereal-basedpreparations

Fruit and vegetables(fresh and processed, excl. bananasand cassava)

Bananas

Natural rubber

Tobacco

Sugar

Cocoa, coffee and tea

Oilcrop products (oilseeds, oils,cakes and meals)

All crop and livestock products,primary and processed

40

30

20

10

0

-10

-20

-30

-40

-50

Year

US$

bill

ion

(cur

rent

pri

ces)

1984 1989 1994 1999

14

Prospects for food and nutrition

because hundreds of millions of people lack themoney to buy what they need or the resourcesto produce it themselves.

Thus, even if there is sufficient potential forproduction in the world as a whole, there willstill be problems of food security at thehousehold or national level. In urban areas, foodinsecurity usually reflects low incomes, but inpoor rural areas it is often inseparable fromproblems affecting food production. In manyareas of the developing world, the majority ofpeople still depend on local agriculture for food

Progress in improving nutrition has beensignificantFreedom from hunger is not only a basic humanright: it is essential for the full enjoyment ofother rights, such as health, education andwork, and everything that flows from these.

The world has made significant progress inraising nutrition levels over the past threedecades. These levels are most commonlymeasured in terms of kilocalories per personper day. People in developing countries needbetween 1 720 and 1 960 kcal per day for basalmetabolism and light activity.

World average food consumption per personhas risen by almost a fifth, from 2 360 kcal per

and/or livelihoods but the potential of localresources to support further increases inproduction is very limited, at least underexisting technological conditions. Examples aresemi-arid areas and areas with problem soils.

In such areas agriculture must be developedthrough support for agricultural research andextension and the provision of credit andinfrastructure, while other income-earningopportunities are created. If this is not done,local food insecurity will remain widespread,even in the midst of global plenty.

person per day in the mid-1960s to 2 800 kcalper person per day today. The gains in the worldaverage reflect predominantly those of thedeveloping countries, given that the industrialand transition economies had fairly high levelsof food consumption already in the mid-1960s.Over the period to 1997-99, average daily percapita food consumption in developing countriesrose from 2 050 kcal to 2 680 kcal (see AnnexTable A3).

The proportion of the world’s populationliving in countries with low average food energyintakes has declined dramatically. In the mid-1960s, no less than 57 percent were living incountries with average intakes below 2 200 kcalper day. India and China both came into thiscategory. By 1997-99, although world populationhad almost doubled to nearly six billion, thisproportion had fallen to just 10 percent. Eventhe absolute numbers — which decline moreslowly because of population growth — fell byover two-thirds, from 1 890 million to 570million.

At the other extreme, the share of theworld’s population living in countries withaverage food energy intakes above 2 700 kcalper person per day has more than doubled, from30 percent to 61 percent. Rapid gains in some ofthe largest developing countries, including

Global progress in improving human nutrition will continue,but in numerical terms it will be slow. Even by 2030,hundreds of millions of poor people will remain under-nourished unless local food production is given higherpriority and inequality of access to food is reduced. However,the lower incidence of undernourishment will make theproblem more tractable through national and internationalpolicy interventions.

15

China, Brazil, Indonesia and Nigeria, accountfor much of this progress. India, however, hasyet to move into this category.

Over this same period, world annualconsumption of cereals for both food and feedhas doubled to 1.9 billion tonnes, while that ofmeat has more than doubled — no meanachievement considering popular fears thatthe world was running out of potential toincrease production. The main forces drivingthis achievement have included higher in-comes, which have increased effective demand,increased supplies, owing to improvementsin productivity, and the growth of trade andtransport links, which have allowed food deficitsin some areas to be covered by surpluses fromother areas.

Yet hundreds of millions remainundernourishedThis remarkable achievement has neverthe-less left out a massive number of people, whocontinue to fare badly. In 1997-99 there werestill 777 million undernourished people indeveloping countries — about one person in six.This represents only a modest decline from thefigure of 816 million for 1990-92.

In China, huge reductions in poverty raisednational average food consumption substantially— and this had a strong effect on the globalpicture. If China is removed from the picture,it becomes clear that the number of under-nourished people actually increased in the otherdeveloping countries, by almost 40 million.

The region with the largest number ofundernourished people in 1997-99 was SouthAsia, where 303 million or just under a quarterof the population remained undernourished.The region with the highest proportion was sub-Saharan Africa, where over a third of the totalpopulation, or 194 million people, wereundernourished.

In 1997-99, some 30 developing countriesstill had average per capita food consumption ofbelow 2 200 kcal per day. War and civil strifewere significant factors in no less than half ofthese countries. In most of them, food consum-ption today stands at levels below those attainedin the past. Some 23 of the 30 are in sub-Saharan Africa, while only 7 are in other regions.

Populations and incomes will continue to growFuture food consumption patterns are deter-mined by growth in population and in incomes,and by changes in dietary preferences.

The latest projections by the United Nations(UN) show a continuing slowdown in the growthof the world’s population. In the medium UNprojection, the 6.1 billion people of 2000 willgrow to 7.2 billion in 2015 and 8.3 billion in 2030,heading towards 9.3 billion in 2050.

In 1997-99 there were still 777 millionundernourished people in developingcountries — about one person in six.

Global progress in nutrition: energy intake levels by percentage of the world’s population, 1964-66 and 1997-99

Source: FAO data

2200

2200-2500

2500-2700

2700-3000

3000

kcal1997-9910%

25%

4%19%

42%

57%

1964-6620%

9%5%

9%

16

Perceptions of a continuing populationexplosion are false. In fact it is more than 30years since the world passed its peakpopulation growth rate, of 2.04 percent a year, inthe late 1960s. Since then the growth rate hasfallen to 1.35 percent. This is expected to fallfurther to 1.1 percent in the period 2010 to 2015and to 0.8 percent in 2025 to 2030. There will bea corresponding slowdown in the growth ofdemand for food.

The absolute numbers added each year arealso past their peak of 86 million a year,reached in the late 1980s. Even so, currentannual additions of around 77 million stillamount to almost a new Germany each year.The yearly increments will taper off only slowlyduring the study period: even by the period 2025to 2030 they will still be running at 67 million ayear. It is only by the middle of the century thatthese increments will have fallen significantly,to 43 million per year in 2045 to 2050. Almost allof these increases will be in the developingcountries.

By 2030 there will be substantial differencesin population growth rates among thedeveloping countries. While East Asia’spopulation will be growing at only 0.4 percent ayear, that of sub-Saharan Africa will still begrowing at 2.1 percent. By 2030, every thirdperson added to the world’s population will bea sub-Saharan African. By 2050, this will rise toevery second person.

The second major factor determining thedemand for food is growth in incomes. Thelatest World Bank assessment of futureeconomic growth is less optimistic than itspredecessors, but it still projects a rise of 1.9percent a year in per capita incomes between2000 and 2015, higher than the 1.2 percent seenin the 1990s.

What will happen to the incidence of povertyunder this overall economic scenario is of greatimportance to food security because poverty andhunger are closely associated. The World Bankhas estimated the implications of its economicgrowth projections for poverty reduction by theyear 2015. They are that:• It is possible to achieve the goal of halving the

proportion of people living in absolute poverty

— defined as an income below US$1 per day —by 2015, over the 1990 level.

• However, it is unlikely that the number of poorpeople can also be halved. This will declinefrom 1.27 billion in 1990 to 0.75 billion in 2015.

• Much of the decline will be due todevelopment in East and South Asia. Indeed,about half of the decline of 400 millionprojected for East Asia has already occurred.

• Only in sub-Saharan Africa, where incomesare expected to grow very slowly, are thenumbers living in poverty expected to rise,from 240 million in 1990 to 345 million in2015. By then, two out of five people in theregion will be living in poverty.

Average nutrition will improve, butundernourishment will fall only slowlyIn the light of these changes in population andincomes, progress in improving nutrition isexpected to continue, though more slowly thanin the past. Average per capita food consump-tion in developing countries is projected to riseby 6.3 percent, from 2 680 kcal in 1997-99 to 2850 kcal in 2015. This is a third of the riseachieved between 1974-76 and 1997-99.

The slowdown is occurring not because ofproduction limits but because many countrieshave now reached medium to high levels ofconsumption, beyond which there is less scopethan in the past for further increases. Hugecountries such as China, where per capitaconsumption rose from 2 050 kcal per day in themid-1970s to over 3 000 kcal per day today, havealready passed the phase of rapid growth. Moreand more countries will be attaining such levelsover the projection period.

By 2030, three-quarters of thepopulation of the developing worldcould be living in countries whereless than 5 percent of people areundernourished. Only 1 in 13 live insuch countries at present.

17

The World Food Summit of 1996 set a targetof halving the numbers of undernourished indeveloping countries by 2015, compared withthe base period of 1990-92. FAO’s study hasfound that the proportion of undernourishedpeople should fall significantly, from 20 percentin 1990-92 to 11 percent by 2015 and 6 percentby 2030. However, in numerical terms theWorld Food Summit target is unlikely to be met.The total number of undernourished peoplewill probably fall from 815 million in 1990-92 tosome 610 million by 2015. Not until 2030 willthe numbers fall to 440 million, therebyapproaching the 2015 target.

The proportion of the world’s populationliving in countries with per capita foodconsumption under 2 200 kcal per day will fallto only 2.4 percent in 2030. The reduction in thenumber of undernourished people will beimpressive in some regions: in South Asia, forexample, it could fall from 303 million in1997-99 to 119 million in 2030, while in EastAsia the number could halve from its currentlevel of 193 million.

In contrast, in sub-Saharan Africa andthe Near East and North Africa, there is likelyto be little or no decline in the numbers ofundernourished people, although the proportionwill approximately halve. By 2030, all regionsexcept sub-Saharan Africa should see theincidence of undernourishment decline tobetween 4 and 6 percent, down from the rangeof 9 to 24 percent today. In sub-Saharan Africa,15 percent of the population or 183 millionpeople will still be undernourished by 2030. Thiswill be by far the highest total for any region,and is only 11 million less than in 1997-99. Thefate of sub-Saharan Africa is therefore cause forserious concern.

As incomes rise, access to food shouldbecome more equal. This is because poorpeople spend a high proportion of increases intheir incomes on food, whereas there is anupper limit to the amount of food that richpeople want to eat. This greater equality willhave a significant effect on the numbers ofundernourished people. For example, in the44 countries that will have average per capitafood intakes of over 2 700 kcal per day in 2015,

Number of undernourished people by region, 1990-92 to 2030

Source: FAO data and projections

the number of undernourished people is ex-pected to be 295 million. But if inequality ofaccess to food were to remain unchanged attoday’s level, this number would be 400 million.

The decline in the numbers of under-nourished between now and 2030 will be slow,for several reasons:• Rapid population growth means that, although

the proportion of undernourished may fall, theabsolute number will fall much less and mayin a few cases even rise. This is an importantfactor in sub-Saharan Africa and the NearEast and North Africa.

• Economic growth will not be fast enough. Inthe Niger, for example, 3.3 million people or41 percent of the population were under-nourished in 1990-92. To achieve the WorldFood Summit target, the number of under-nourished will need to fall to 1.65 million or9 percent of the population by 2015. To bringthis about would require growth rates farabove what the Niger has seen over the pasttwo decades.

• Several countries start from highly adverseconditions, namely low national average food

South Asia East Asia Sub-SaharanAfrica

Near East andNorth Africa

Latin Americaand Caribbean

Mill

ion

peop

le

350

300

250

200

150

100

50

0

1990-92

1997-99

2015

2030

18

consumption, high incidence of undernourish-ment and high projected population growth.For example, nine developing countries hadproportions of undernourishment in 1990-92of over 50 percent (Afghanistan, Angola,Burundi, Democratic Republic of Congo,Eritrea, Ethiopia, Haiti, Mozambique andSomalia). In these countries the proportionof undernourished is expected to fall to39 percent by 2015 and to 25 percent by 2030.However, because of the relatively highgrowth rate of this group’s population, theabsolute numbers affected will rise to115 million in 2015 and may still be 106million in 2030. Even these figures are basedon projections for growth in food consumptionthat are much faster than the fastest seen inany comparable period in the past.

• In countries where average food intake iscurrently low and the majority of people arehungry, reducing inequality of access to food

has only a small impact on levels of under-nourishment. This is because few people areon diets that are more than barely adequate,so redistributing their food “surplus” does notgreatly improve matters. By 2015 there willstill be 41 countries with average food intakesof 2 500 kcal per day or less.

• In the future the threshold for definingundernourishment will rise, as ageingreduces the proportion of children in thepopulation. Since children’s food energyrequirements are lower than those of adults,the average calorie requirement in developingcountries will rise by around 3 percent by2030. If it were not for this rise in thethreshold, the number of undernourishedestimated for 2030 would be 370 millioninstead of 440 million.

The number of undernourished can bereduced more rapidly by affording increasedpriority to agriculture, increasing national foodproduction and reducing inequality of accessto food. These three measures should becombined with continuing interventions tocope with the consequences of local foodcrises, until the root causes of undernourish-ment have been removed.

It is possible for countries to raise nutritionlevels even in the absence of significanteconomic growth. Mali raised average foodconsumption by almost a third in the 1980s,although per capita household expenditure fellover this period. Other countries, such as Benin,Burkina Faso, Ghana, Mauritania and Nigeria,achieved similar quantum leaps at times of slowincome growth. The common characteristicseems to have been rapid growth in theproduction of food staples (cereals, roots andtubers), leading to improved self-sufficiency, atleast in cereals. Because most agriculture wasat or below the subsistence level, increasedproduction led directly to improved foodconsumption in rural populations.

The problem of undernourishment shouldbecome more tractableThe projections imply that the problem ofundernourishment should become more tract-able in future. This will work in two major ways:

Source: FAO data and projections

Dietary changes in developing countries,1964-66 to 2030

1964-66 1997-99 2030

3 000

2 500

2 000

1 500

1 000

500

0

kcal

/cap

ita/d

ay

Other

Pulses

Roots and tubers

Meat

Sugar

Vegetable oils

Other cereals

Wheat

Rice

19

Just as world average calorie intakes have

increased, so also people’s diets have

changed. Patterns of food consumption are

becoming more similar throughout the

world, incorporating higher-quality and more

expensive foods such as meat and dairy

products.

This trend is partly due to simple prefer-

ences. Partly, too, it is due to increased

international trade in foods, to the global

spread of fast food chains, and to exposure to

North American and European dietary habits.

Convenience also plays a part, for example

the portability and ease of preparation of

ready-made bread or pizza, versus root

vegetables. Changes in diet closely follow

rises in incomes and occur almost irrespec-

tive of geography, history, culture or religion.

However, cultural and religious factors do

explain differences between countries with

similar income levels. For example, Hindus

abstain from beef or meat in general, Mos-

lems and Jews from pork. Despite similar

income levels, Japanese people consume far

fewer calories from non-starchy foods than do

Americans, as do Thais compared with

Brazilians.

Dietary convergence is quite high

among the high-income countries of the

Organisation for Economic Co-operation and

Development (OECD), where food consump-

tion patterns show a 75 percent overlap with

How diets will change

• As the incidence of undernourishmentdiminishes, more and more countries willfind it easier to address the problem throughnational policy interventions. By 2030,three-quarters of the population of develop-ing countries could be living in countrieswhere less than 5 percent of people are

those in the United States, meaning that

75 percent of processed food products are

based on the same raw materials. Even

Japan has been moving closer to other

OECD countries, with the overlap rising from

45 percent in 1961 to about 70 percent in

1999. Convergence towards North American

dietary patterns is also occurring in other

groups of developing countries, though only

slowly in some cases, especially in land-

locked or politically isolated countries where

international influences permeate less

easily. However, cultural factors appear to

limit convergence to an upper ceiling of

around 80 percent, at least for the time

being.

These changes in diet have had an impact

on the global demand for agricultural

products and will go on doing so. Meat

consumption in developing countries, for

example, has risen from only 10 kg per

person per year in 1964-66 to 26 in 1997-99.

It is projected to rise still further, to 37 kg

per person per year in 2030. Milk and dairy

products have also seen rapid growth, from

28 kg per person per year in 1964-66 to 45 kg

now, and could rise to 66 kg by 2030. The

intake of calories derived from sugar and

vegetable oils is expected to increase.

However, average human consumption of

cereals, pulses, roots and tubers is expected

to level off.

undernourished, compared with 7.7 percentat present. This dramatic change will occurbecause the majority of the most populouscountries (Brazil, China, India, Indonesia,Islamic Republic of Iran, Mexico andPakistan) will shift to the “under-5-percent”category.

20

• The number of countries with severeproblems of undernourishment will becomesmaller over time. International policyresponses will tend to become more feasibleand effective, as the total effort need not bespread so thinly. For example, if the

projections come true, the number ofcountries with undernourishment of over25 percent will fall from 35 at present(accounting for 13 percent of the populationof the developing countries) to 15 in 2030(accounting for only 3.5 percent).

21

Poverty and agriculture

Undernourishment is not merely a symptom ofpoverty but also one of its causes. Poverty is notsimply a lack of income or consumption butincludes deprivation in health, education,nutrition, safety, legal and political rights, andmany other areas. All these dimensions ofdeprivation interact with and reinforce eachother.

Over the past decade, poverty and therelated issue of inequality have moved to the topof the international development agenda. Atvarious summits from the early 1990s onwards,world leaders have proclaimed theircommitment to poverty reduction and adopted aseries of related targets. These cover a widerange, from infant and child mortality to schoolenrolment, from gender equality to maternalmortality, from access to health andreproductive health services to the adoption of

Food and Agriculture in Nationaland International Settings

Except in most of sub-Saharan Africa,developing countries are makingprogress towards the UN goal ofhalving the incidence of poverty by2015. Growth in agriculture and innon-farm rural activities, as well asimprovements in nutrition, will becentral to continuing success.Sub-Saharan Africa’s continuingdescent into poverty is cause forserious concern.

national strategies for sustainable development.The UN Millennium Declaration, adopted inSeptember 2000, consolidated most of thesetargets, including that of halving the proportionof people living in extreme poverty by 2015.The international targets, and the indicatorsused to assess progress towards them, shouldbe viewed neither as finely tuned criteria toguide policy and spending priorities nor asaccurate measures of progress. In many poorcountries the necessary data are not reliableand may not even be up-to-date. Nor are theynecessarily comparable between differentcountries. But the targets are useful in drawingattention to persistent poverty, and ininfluencing opinion and creating a sense ofurgency among the public, politicians and thedevelopment community. The indicators canalso serve as rough guides to assess whetherprogress is being made.

Overall progress and prospectsAt the beginning of the twenty-first century,over 1.1 billion people are living in extremepoverty, subsisting on less than US$1 a day.Significant, but uneven, progress is being madetowards meeting the 2015 target of halving theproportion of people living in poverty indeveloping countries. This proportion fell from32 percent in 1990 to 25 percent in 1999.However, because of population growth, thereduction in numbers was less dramatic, from1 269 million to 1 134 million.

The regional picture was highly diverse. InEast Asia, poverty fell very steeply during the1990s. In South Asia, although the proportion of

22

poor fell, the total number remained almostconstant. In sub-Saharan Africa, the proportionremained virtually unchanged, while thenumber rose steeply.

The latest World Bank projections suggestthat the target of halving the proportion ofpeople living in poverty in the developingcountries by 2015 can be achieved. However,even if this target were met, because ofpopulation growth the result would be a fallof less than 30 percent in the absolute numberof poor. In sub-Saharan Africa the target seemsunattainable: projections suggest only a smallreduction in the proportion and a continued risein the number living in poverty.

Even the World Bank projections assumefaster economic growth rates than in the past.The Bank stresses that, if the slow growth of the1990s persists, then the number of people livingin extreme poverty will remain near currentlevels for the next 15 years.

Faster growth of incomes is essential forpoverty reduction everywhere. Reducinginequality is equally crucial, especially incountries where this is pronounced. Accordingto some estimates, countries where inequality ishigh will need twice as much growth as thosewhere it is low to meet the poverty target.

Selected targets associated with the United Nations MillenniumDevelopment Goals

The following targets are to be achieved by 2015, against a base year of1990:• Halve the proportion of the world’s people whose income is less than

US$1 a day• Halve the proportion of people who suffer from hunger• Halve the proportion of people without access to safe drinking water• Ensure full primary schooling for all children• Ensure equal access to all levels of education for boys and girls• Reduce under-five child mortality by two-thirds• Reduce maternal mortality by three-quarters• Halt and begin to reverse the spread of HIV/AIDS, malaria and other

major diseases.

Why better nutrition has to come firstFood and agriculture are centrally involved inboth the generation and the reduction ofpoverty.

Undernourishment is a characteristicfeature of poverty and a direct violation of auniversally recognized human right. It alsodeepens other aspects of poverty, in thefollowing important ways:• It leaves people more susceptible to illness.

Episodes of illness in turn reduce the intakeand absorption of food, producing a viciousdownward spiral in which hunger and diseasefeed off each other.

• When pregnant and nursing mothers areundernourished, babies are born underweightand start life with a nutritional handicap thatcan affect their health throughout their lives.

• Undernourishment can affect brain develop-ment in the womb and attentivenessin class, and so is associated with pooreducational performance.

• When energy and protein intakes areinadequate for the requirements of work,muscle mass and labour productivity can bereduced. Along with illness, this affects wagesand earnings. Studies have shown that a1 percent increase in body mass index (BMI, ameasure of weight for a given height) isassociated with an increase in wages of morethan 2 percent, at least over part of the rangeof BMIs.

• Micronutrient deficiencies can also reducework capacity. Surveys suggest that anaemiacaused by iron deficiency is associated with a17 percent loss of productivity in heavymanual labour.

• Investment and risk-taking are essential foreconomic growth, but people who live on theedge of starvation are likely to be extremelycautious about investing, since they cannotafford a drop in production or earnings.

• All this means that widespread hunger candepress the performance of whole economies.Studies in Bangladesh, India, Pakistan andViet Nam estimate that adult productivitylosses due to the combined effect of stuntingand deficiencies of iodine and iron consider-ably reduce the growth of incomes.

23

Growth in incomes is essential if under-nourishment is to be reduced, but it is notenough by itself. Better public services — suchas improved female and nutrition education, safedrinking water and improved health services andsanitation — are also needed. Interventions inthese areas must be carefully targeted towardsthe most vulnerable groups.

Agriculture holds the keyThe development community today shares thesame broad recipe for poverty reduction. Therecipe involves fostering pro-poor economicgrowth and favouring poor people’s access to allthe services and other factors that supportpoverty eradication and define an acceptablestandard of living: markets, credit and income-producing assets, basic education, health andsanitation services, safe water, transport andcommunications infrastructure, and so on.Providing access to these basic human rights isseen as an end in itself, but it will also boosteconomic growth.

Growth in the agricultural sector has acrucial role to play in reducing poverty. TheInternational Fund for Agricultural Development(IFAD) estimates that seven out of ten of theworld’s poor still live in rural areas. They includesmallholders, landless labourers, traditionalpastoralists, artisanal fishers and marginalizedgroups such as refugees, indigenous peoplesand female-headed households.

Many of the rural poor work directly inagriculture, as smallholders, farm labourersor herders. Their incomes can be boosted by

Income growth is essential ifundernourishment is to be reduced,but better public services — improvedfemale and nutrition education,safe drinking water, and improvedhealth services and sanitation — arealso crucial.

pro-poor measures, such as ensuring fairaccess to land, water and other assets andinputs, and to services, including educationand health.

Agricultural growth spreads its benefitswidely. Growth in the incomes of farmers andfarm labourers creates increased demand forbasic non-farm products and services in ruralareas. These include tools, blacksmithing,carpentry, clothes, processed food bought fromroadside kiosks, and so on. These goods andservices are often difficult to trade over longdistances. They tend to be produced andprovided locally, usually with labour-intensivemethods, and so have great potential to createemployment and alleviate poverty. Surveys infour African countries have shown that betweenone-third and two-thirds of income increases inrural areas are spent on such local goods andservices.

For the poor, the rural non-farm sectoroffers a relatively easy escape route from

Progress in poverty alleviation: number of people living inpoverty, 1990 to 2015

Source: World Bank (2001b)

Mill

ion

peop

le

199019992015 (World Bank projection)2015 (UN target)

East AsiaSouth AsiaLatin Americaand Caribbean

Near East andNorth Africa

Sub-SaharanAfrica

600

400

200

0

24

poverty. Rural non-farm enterprise oftenrequires little capital or training to set up and sooffers many of the rural poor opportunities tofind work and raise their incomes. Non-farmactivities provide 44 percent of rural jobs in Asiaand 25 percent in Latin America. In rural Indiathey provide 60 percent of the income of thepoorest 20 percent of the rural population.

But the rural non-farm sector cannot growindependently: agriculture must grow first, togenerate the increased demand for non-farmproducts. There can be a general rise in localwages only when growth in both farm and non-farm activities has soaked up most of the pool ofrural underemployment.

And agricultural growth alone may notalways produce a decline in rural poverty. Iflandholdings are very unequal, increasedincomes from farming may accrue almostentirely to large-scale farmers or absentee

Growth in agriculture and in associated rural non-farmemployment can have a broad impact in reducing poverty inrural areas, where seven out of ten of the world’s poor live.

landlords, who may either save it or invest itoutside the rural areas, on urban or importedgoods. In such cases the impact of agriculturalgrowth on poverty may be limited, and policiesto reduce inequality of access to assets such asland, water and inputs will be needed instead.

What economic policies at national levelfoster agricultural growth in developingcountries? During the 1950s and 1960s it waswidely believed that only industrial growth coulddeliver economic development. As a result,industry was protected while agriculture washeavily taxed or afforded low priority. By the endof the 1970s, there was increasing emphasis onthe structural reform of economies. It washoped that privatization, the liberalization ofinternal and external trade, lower taxes andreduced government intervention wouldproduce higher economic growth and reduce thebias against agriculture.

These measures have been widely adopted.However, there is little evidence to show thatthey have done much to increase growth, eitherin gross domestic product (GDP) as a whole orin agricultural GDP. This suggests that, badlyneeded though they were, these measures arenot enough in themselves and need to besupplemented with other policies.

International trade and globalizationFreer trade is highly prized as a route to peaceand prosperity. In developing countries,particularly in the least developed economies,freer trade in agriculture can raise incomesgreatly, be an important source of foreignexchange and act as a catalyst for overalldevelopment. For most countries, food importsare already an important source of supplies andwill continue to contribute to food security.

Rising agricultural trade deficits in developingcountriesThe trade patterns of developing countries havechanged rapidly over the past 40 years:

• Agricultural exports have grown modestlycompared to those of manufactured goods,resulting in a dramatic decline in the shareof agricultural exports in total traded merch-andise, from about 50 percent in the early1960s to about 6 percent by the year 2000.

• The overall agricultural trade surplus ofthese countries has virtually disappeared andthe outlook to 2030 suggests that they willbecome, as a group, net importers ofagricultural commodities, especially oftemperate-zone commodities.

• The least developed countries (LDCs), also asa group, became net importers of agricultural

25

Trade reform has lowered the barriersto trade, increased global economicintegration, enhanced productivityand boosted incomes — and willcontinue to do so. Not all countriesor stakeholders have been winners,but national and international policyinterventions could soften the impacton losers. Special measures couldensure that a greater share of thebenefits of trade go to developingcountries.

products as early as the mid-1980s. Theiragricultural trade deficit has been wideningrapidly and could quadruple by 2030.

Both policy and market factors are drivingthese changes. On the policy side, barriers totrade and support for domestic production inthe developed (mainly the OECD) countries haveheld back the growth of agricultural exports

from the developing world. These tradedistortions impose high costs and createwidespread inefficiencies. In the countries thatuse them, they exact higher prices and taxesfrom consumers and taxpayers. For othercountries, they limit access to export marketsand introduce unfair competition in domesticmarkets. They hold world commodity pricesdown and so hold back the development ofagriculture, especially in developing countrieswhere less government support is available.

On the market side, growth in agriculturalexports from developing countries has beenheld back by sluggish and largely saturateddemand in developed markets, in particular fortropical products such as coffee, cocoa and tea.

Ambitious goals, modest achievementsThe benefits of trade reform experienced bymany outward-oriented economies havecreated the momentum to continue reducingthe barriers to trade. Many developing countrieshad already liberalized aspects of their agri-cultural trade since the 1980s under structuraladjustment reforms. These reforms, and the fullrange of policies that affect agricultural trade,were subjected to systematic multilateral

Source: FAO data

Agricultural trade balance and share of agricultural exports in merchandisetrade, 1960 to 2000

% s

hare

1961 1970 1980 1990 2000

100

50

0Net imports

US$

bill

ion

Net exports

1961 1970 1980 1990 2000

20

15

10

5

0

-5

-10

Developing countries Least developed countries

The surplus of developing countrieshas been shrinking...

...as also has the share of agricultural exportsin total merchandise trade

26

controls for the first time by the UruguayRound’s 1994 Agreement on Agriculture (AoA).

The Agreement was hailed as a watershed,yet so far the results have been modest andoften disappointing. FAO studies have foundthat, for most agricultural commodities, theAoA’s impact on prices and levels of trade hasbeen negligible, as has its impact on manydeveloping economies. Producer support of alltypes remains high in developed countries: in2000 it totalled US$245 billion in the OECDcountries. This figure rises to US$327 billion ifmore general transfers to agriculture areincluded.

Tariffs continue to curb trade. Under the AoA,non-tariff barriers such as quotas were to bereplaced by equivalent tariffs. In addition,developed countries agreed to reduce all theirtariffs by an average of 36 percent, over a periodof six years, with a minimum of 15 percent forany one trade item. Developing countries agreedto reduce tariffs by 24 percent over a ten-yearperiod. The least developed countries were notrequired to make any reductions.

The reductions made since 1994 havecomplied with these goals, but it is not clearthat market access has improved significantly.Developed country tariffs have been cut by anaverage 37 percent, but the deepest cuts havebeen mainly for unprocessed tropical crops thatalready had low tariffs. Commodities alsoproduced in developed countries, and processedproducts, benefited much less. For example,maximum allowable tariffs agreed by theEuropean Union (EU) under the AoA were 86percent on beef and 215 percent on frozen beef,whereas they are only 6 percent on pineapplesbut 25 percent on processed pineapples.

Domestic support remains high. Governmentsupport for agriculture can also distort trade, byallowing domestic producers to sell at lowerprices than would otherwise be economicallyviable.

The AoA also covered domestic support.Several types of support, such as research,infrastructure and environmental programmes,were exempted. Developing countries could also

exclude measures of a developmental nature,such as agricultural and rural developmentprogrammes.

The AoA required developed countries tomake a 20 percent reduction in their support foragriculture, developing countries a 13.3 percentcut and least developed countries none. Thesecuts were to be made with reference to a 1986-88 base, over a period of six years for developedcountries and ten years for developingcountries.

In reality many countries have faced muchless pressure to reduce support for, andprotection of, their agricultural sector. This ismainly due to the fact that the commitments toliberalize were based on historically high levelsof support and protection. These so-called“bound” levels remained high enough tomaintain much of the protection previouslyenjoyed, even after the cuts had been imple-mented. Indeed, total support to agriculture inthe rich OECD countries was actually higher in1998-2000 than before the AoA.

Export subsidies are still substantial. The AoAbrought direct subsidies for agricultural exportsinto an international trade agreement for thefirst time. Indirect subsidies, such as exportcredit guarantees and food aid, were alsocovered. Developed countries agreed to reducetheir expenditure on subsidies by 36 percentand developing countries by 24 percent.Reductions in the volume of subsidized exportswere also negotiated, with reductions for eachcommodity of 21 percent required for developedand 14 percent for developing countries. Leastdeveloped countries undertook no commit-ments to reduce their subsidies. The EUaccounts for the bulk of direct export subsidies:in 1998 it spent US$5.8 billion, more than90 percent of all such subsidies covered by theAoA.

More liberalization would mainly benefitdeveloped countriesAccording to most studies, completeliberalization of agricultural trade couldproduce valuable overall welfare gains, butsome groups would win while others would

27

lose. The benefits would go mainly toconsumers and taxpayers in industrialcountries, where agriculture is most protected,and to developing country agriculturalexporters. In contrast, urban and landless ruralconsumers in developing countries might endup paying higher prices for some foodstuffs,especially cereals, milk, meat and sugar.Specific measures would be needed to helpsuch loser groups.

The results of studies on the impact ofagricultural trade liberalization vary accordingto the assumptions they make. For example, arecent study found that complete liberalizationwould boost global incomes by US$165 billion ayear. The largest benefits would arise fromreforms in developed countries, but the lion’sshare of these, amounting to some US$121billion, would also remain in these countries.Developing countries stand to gain significantly(by US$31 billion) only if they also liberalizetheir own trade.

The current FAO study also looked at theimpacts of gradually removing price supportsand other subsidies over the 30 years to 2030.The analysis focused on the expected priceeffects for consumers and producers, in bothdeveloped and developing countries. It foundthat international prices could rise moderately,while prices would fall substantially in countrieswith high levels of protection. Producers tradingat international prices would gain, while those

producing at inflated protected prices wouldlose. Like the study described above, the FAOstudy found that the benefits for consumers inhitherto protected OECD markets could be high,but it also stressed that high processing anddistribution costs in these countries could meanthat lower prices for raw products would nottranslate into substantially lower prices for thefinal consumer. Consumers in developingcountries, where processing and distributionmargins are much smaller, stood to lose moresignificantly. Trade liberalization would notchange the main conclusion of this study, thatdeveloping countries will increasingly becomenet importers of agricultural products — but itwould slow the process somewhat.

Why do developing countries stand to gain somuch less from trade liberalization thandeveloped countries? One reason is that manydeveloping countries have become netimporters of agricultural products, and modestincreases in world prices are unlikely to turnthem into net exporters. In the importingdeveloping countries, consumers stand to losemore from freer trade than domestic producersare likely to gain.

The finding that gains for producers indeveloping countries would often be smallreflects a number of factors:• Many studies show that a cut in OECD

subsidies would merely bring about anexchange of market shares between OECDcountries. This is because OECD tradedistortions are concentrated on temperate-zone commodities — products for which, inthe majority of developing countries, theproduction potential is limited more by agro-ecological conditions than by policydistortions abroad.

Eliminating all agricultural policydistortions could produce globalannual welfare gains of up to US$165billion, of which three-quarters wouldgo to developed countries.

Potential annual welfare gains fromagricultural trade liberalization

Source: Anderson et al. (2000)

Gains accruing to low-income countries

Gains accruing to high-income countries

If low-incomecountries liberalize

US$ billion

If all countriesliberalize

If high-incomecountries liberalize

0 60 120 180

28

• Where developing countries have a com-parative advantage — in such commodities ascoffee, cocoa, tea, spices and tropical fruits— developed countries’ import tariffs havealready been reduced and the effects of furtherliberalization are likely to be small.

• Higher and more stable international pricesare not always transmitted to farmers indeveloping countries. Inadequate infra-structure and inefficient marketing systemsinsulate many of them from world markets.

• Farmers in developing countries may not gainas long as domestic policies largely offset theprice incentives from international markets.Most developing countries heavily taxed theiragriculture throughout the 1970s and 1980s;many, including India, China and Pakistan,continued to do so during the 1990s.

How can trade liberalization benefit developingcountries?What measures and strategies would ensurethat the poorest and most vulnerable countriesand population groups receive an equitableshare of the benefits of trade liberalization?The aim should be to:• Eliminate direct and indirect export subsidies.• Rationalize and simplify access to OECD

markets. Specifically, rationalize and simplifytrade preferences, assist countries whosepreferences have been eroded throughmultilateral liberalization, and deepenexisting preferences for very poor countries.

• Reduce OECD tariffs and consumer taxes onprocessed agricultural products, with specialpreferences for products from developingcountries.

• Eliminate tariff escalation for tropicalcommodities, in the developing as well as thedeveloped countries. Tariffs are rising evenfaster in the former than in the latter group.The purchasing power of China’s or India’srapidly growing middle class could turn thesecountries into major importers of some tropicalagricultural products over the next 30 years.

• Create or expand safety nets and fooddistribution schemes, to ensure that low-income consumers are not penalized by risesin the prices of food imports.

If developing countries are to benefit fromfreer trade, their farmers will need to becomemore responsive to the rising and more stableinternational prices that should result fromsuch trade. A massive mobilization of resourcesis needed to improve agricultural productivity athome and thus competitiveness abroad. Themost important measures are increased creditsfor rural areas, and more investment in allaspects of support for agricultural productionand processing, including rural infrastructure(irrigation, transportation, storage andmarketing), research, education and training,and standard setting and quality control.

Substantial gains would also result fromother policy reforms. In developing countries,removing taxes on agricultural exports andtariffs on non-agricultural imports (machinery,fertilizers and pesticides) would improve theterms of agricultural trade and help farmerscompete on international markets. In developedcountries, removing trade barriers in labour-intensive manufacturing could bring benefits forfarmers in developing countries. For example, arapidly growing textile industry would createnew income opportunities for cotton farmers inthe tropics.

Non-agricultural exports now account formore than 90 percent of the total exports fromdeveloping countries, and more than 80 percentin the case of least developed countries. Deeperand broader preferential access to the marketsfor manufactured goods in some developedcountries could make an important contributionto food security in the least developed countries,providing them with the means to finance theirhuge and rapidly increasing food import needsin the future.

Does globalization disadvantage the poorestcountries?Globalization is a modern word for a processthat has been going on for centuries. Newtechnologies in the fields of transport andcommunications, from advances in sailing andnavigation to the steamship and the telegraph,have often reduced the cost of shifting goodsaround the world in the past, leading toincreased economic integration. Recently,

29

such technologies have included roll-on roll-offcontainer systems and the Internet, while lowertrade barriers have further eased the movementof goods and capital.

Globalization has brought lower prices toconsumers, and investment and employment tonewly industrializing countries. But it has alsoraised widespread public concern over the fateof the poorer developing countries, which arealleged to have been left further and furtherbehind as the rest of the world advances.

There is strong evidence that countries canbe disadvantaged in the global marketplace bytheir geographical endowments. Lack ofinfrastructure can make it hard to getperishable products to markets, increasingmarketing costs and so deterring investment.As new investment heads for better-endowed

areas, those countries and regions with physicaland infrastructural handicaps may be bypassed,falling further and further behind and findingthemselves trapped in a vicious circle ofdisadvantage.

Most poor countries are located in thetropics, where the higher incidence of crop andlivestock diseases and pests and excessive orinadequate rainfall are further factorscompromising their ability to participate inglobal agricultural markets. Distance from thesea and a lack of navigable waterways canconstitute additional disadvantages. OutsideEurope, average incomes in landlockedcountries are only a third of those in countrieswith a seaboard.

Sub-Saharan Africa, located mainly in thetropics and with a high proportion ofproblematic soils, suffers multiple handicaps inthe global marketplace. Only 21 percent of thisregion’s population live within 100 km of thecoast or of a navigable river, against 89 percentin high-income countries. The proportion of thepopulation that is landlocked is seven timeshigher than in rich countries. Landlockedcountries in Africa have average freight costsalmost three times higher than in high-incomecountries.

In contrast, regions of the United States,Western Europe and temperate-zone EastAsia within 100 km of a coastline account fora mere 3 percent of the world’s inhabited landarea. Yet they house 13 percent of the world’spopulation and produce at least 32 percent ofthe world’s GDP.

Combining data on population and in-come levels provides a revealing picture ofthe distribution or density of incomes overdifferent countries and regions. It underscoresthe importance of infrastructure and/orgeographical location, showing that:• Nearly all landlocked countries in the world

are poor, except for a few in Western andCentral Europe which are deeply integratedinto the regional European market andconnected by multiple low-cost trade routes.

• Coastal regions, and regions linked to coastsby navigable waterways, are strongly favouredrelative to the hinterlands.

Sources: World Bank data and Gallup et al. (1999)

The conditions for market integrationdiffer vastly across regions

Land

lock

edpo

pula

tion

(%)

Popu

latio

n w

ithin

100

km o

f coa

st

or ri

ver (

%)

Tele

phon

es p

er

100

peop

leEl

ectr

ic p

ower

capa

city

(kW

/100

peo

ple)

Ocea

n fr

eigh

t

rate

s fo

r cer

eals

(US$

/tonn

e)

170

150

100

50

0

Sub-Saharan Africa

Latin America

High-income countries

30

• Sub-Saharan Africa stands out as the regionthat is most disadvantaged in terms ofunfavourable agro-ecological conditions aswell as inadequate transport andcommunications infrastructure.

Does globalization concentrate too much powerin the hands of multinationals?Globalization is often charged with shiftingpower away from national governments tomultinational enterprises (MNEs). MNEs havebeen accused of abusing market power,exploiting farmers and labourers around theworld, and exerting pressure on governments toreduce environmental and labour standards.

Today MNEs in food and agriculture operateacross many country borders. They are moreand more vertically integrated, covering thewhole sequence of operations from producingand marketing seeds, through purchasing thecrop, to food processing and distribution.

When they control large parts of the supplychain, these large corporations can exertmonopoly selling or buying power, therebyputting pressure on farmers and retailers.Through production contracts or joint ownershipin land or livestock operations, they can tiefarmers into buying the company’s inputs andselling their produce only to the company.Farmers may also lose entrepreneurial capacityand become more or less dependent workers ontheir own farms. It is also true that MNEs canand do move operations from country to country

in search of lower costs, including wage rates,and of lower labour and environmentalstandards.

Benefits of globalizationHowever, if the often heard demands for globalparity in wages and environmental standardswere met, this would remove a major competi-tive advantage of poorer countries and couldhalt the flow of investment towards them,seriously prejudicing their further development.

Countries that excluded MNEs would beexcluding the best available channels for gettingtheir products to the global marketplace. MNEsusually upgrade local skills, methods, standardsand technologies as they expand in a country.For example, in the late 1980s, in China’sHeilongjang province, the multinational Nestlébuilt rural roads, organized milk collectionpoints and trained dairy farmers in basic animalhealth and hygiene.

MNEs also force local firms to upgrade inorder to remain competitive. Recent researchshows that the greater the degree of opennessof a national industry to foreign competitors, thegreater its productivity. Indeed, the presence offoreign firms may be the single greateststimulus to improving productivity available inmany developing country settings.

The claim is often made that globalizationmakes the world’s poor poorer, but there is noevidence for this. Countries may, however,become poorer in a relative sense as they fail to

Growing industry concentration has led to asituation where just a few companies controllarge shares of the agricultural production andprocessing chain. On the production side, onlyfour companies control 50 percent of the USbroiler market and 46 of the US pork market.On the processing side, four companies controlover 80 percent of US beef packaging and 60percent of the pork packing market. Con-centration also extends into the agriculturalupstream sectors, resulting in a combination of

Sprawling giants

horizontal concentration and vertical integrationthroughout the entire agro-food chain. Cargill,for instance is not only amongst the top four beefand pork packers but also number one in termsof domestic grain handling, as well as grain andsoybean exports but also the second largestcompound feed producer and the number threeturkey producer. On the upstream side,Monsanto and Syngenta account together for 35percent of the global market for crop protectionand 19 percent of the one for seeds.

31

Multinational enterprises oftenupgrade local skills, methods,standards and technologies as theyexpand in a country. In so doing, theyforce local firms to upgrade in orderto remain competitive.

benefit from globalization. Recent researchconducted for the World Bank suggests thatopenness to international trade boostseconomic growth. Developing countries withpolicies that favour openness increased theirrate of GDP growth from 1 percent in the 1960sto 3 percent in the 1970s, 4 percent in the 1980sand 5 percent in the 1990s. In contrast, much ofthe rest of the developing world, containingabout 2 billion people, is becoming increasinglymarginalized. The aggregate growth rate ofthese countries was actually negative in the1990s.

Overall, the benefits of continuingglobalization are likely to outweigh the risks andcosts. Negative impacts can be mitigated byappropriate policies. A combination of measures

including openness, investments ininfrastructure, the promotion of economicintegration and limits on market concentrationand control, could make globalization work forthe benefit for the poor.

Income density in the world

Sources: GDP: World Bank (2001a) and FAO data;Population density: Oak Ridge National Laboratory (2000)

‹ 500500 - 1 1001 100 - 3 0003 000 - 8 1008 100 - 22 00022 000 - 60 00060 000 - 165 000165 000 - 500 000› 500 000No data available

GDP per km2

(in US$)

32

Prospects by Major Sector

Crop production

Cereals are still by far the world’s mostimportant sources of food, both for directhuman consumption and indirectly, as inputs tolivestock production. What happens in thecereal sector is therefore crucial to world foodsupplies.

Since the mid-1960s the world has managedto raise cereal production by almost a billiontonnes. Over the next 30 years it must do soagain. Is the task within its capabilities?

Growth of cereal demand slows downThe growth rate of world demand for cereals fellto 1 percent a year in the 1990s, down from 1.9percent in the 1980s and 2.5 percent in the1970s. World annual cereal use per person

(including animal feeds) peaked in the mid-1980s at 334 kg and has since fallen to 317 kg(1997-99 average).

This rapid decline was thought by some toherald a new world food crisis. It wasinterpreted as a sign that the world was hittingthe limits of its capacity for food production andwould soon experience serious threats to foodsecurity.

In fact, average cereal consumption perperson in developing countries has risensteadily throughout the past four decades. Theslowdown in the growth of world consumptionwas due not to production constraints but to aseries of factors that limited demand. Amongthese factors, some are ongoing andwidespread:• World population growth has been slowing.• Many large countries, especially China, are

reaching medium to high consumption levels,such that further rises will be much less rapidthan in the past.

• Persistent poverty has prevented hundreds ofmillions of people from meeting their foodneeds.

Other factors, however, are largely transient.These include:• A fall in demand in the transition economies.

This was the strongest factor during the1990s, when both consumption and imports inthese countries fell from the very high levelsthey had reportedly reached earlier.

• The use of cereals for animal feeds in the EUdeclined until the early 1990s, as highdomestic prices favoured cereal substitutes,which were largely imported. Growth in feed

Cereals: an extra billiontonnes needed

The 1990s saw a decline in the growth of world cerealconsumption. This was due not to limits in productioncapacity but rather to slower growth in demand, partlycaused by exceptional and largely transient factors.Growth in consumption will resume, leading to growingdependence on imports in developing countries. Thepotential exists for traditional and new exporters to fillthis gap, but problems of food security and environmentaldegradation will need to be addressed.

33

use resumed after EU policy reforms lowereddomestic prices.

• Consumption grew more slowly in oil-exporting countries after the effect of theinitial boom in oil prices on incomes andcereal imports had largely dissipated.

• Demand grew more slowly in the second halfof the 1990s in the East Asian economies,which were hit by economic crisis.

The influence of these transient factors isalready on the wane. Over the next 15 years theywill gradually cease depressing the growth incereal demand, which is projected to recover,rising to 1.4 per cent a year by 2015.

Looking further ahead, slower populationgrowth and the levelling off of food consumptionin many countries will continue to dampendemand, the growth of which is expected to slowto 1.2 percent a year over the period 2015 to2030. Nevertheless, the production task facingworld agriculture is massive. By 2030, an extrabillion tonnes of cereals will be needed eachyear. Unforeseeable events such as oil pricebooms, dramatic growth spurts or crises could,of course, alter effective demand over shortperiods, but will not greatly change the bigpicture.

Developing countries will become moredependent on importsIn the developing countries the demand forcereals has grown faster than production.The net cereal imports of these countries rosefrom 39 million tonnes a year in the mid-1970sto 103 million tonnes in 1997-99, representinga move from 4 percent of their annual cerealuse to 9 percent. This dependence on importsis likely to increase in the years ahead. By 2030the developing countries could be importing265 million tonnes of cereals, or 14 percentof their consumption, annually.

Though this increase may seem massive,it represents a lower rate of growth overthe next three decades than since the mid-1970s. If real food prices do not rise, andindustry and services grow as previously,then most countries will be able to afford toimport cereals to meet their needs. However,the poorest countries with the worst food

World demand for cereals, 1965 to 2030

Source: FAO data and projections

security also tend to be least able to pay forimports.

Exporters can fill the gapCan the rest of the world produce the exportsurpluses needed to fill the gap? It is worthexamining the experience of the past quartercentury. Between the mid-1970s and 1997-99the net annual imports of all cereal-importingcountries almost doubled, from 89 milliontonnes to 167 million tonnes.

Cereal exporters coped well with the spurtin demand, doubling their export levels.Traditional exporters such as Australia, NorthAmerica, Argentina and Uruguay played theirpart. They have the potential to continue to doso. But about half the total increase in exportscame from a new player, the EU. From being anet importer of 21 million tonnes of grain a yearin the mid-1970s, the EU became a net exporterof 24 million tonnes a year in 1997-99. Initially,much of this turnaround depended on heavy

The developing countries will become increasinglydependent on cereal imports. By 2030 they could beproducing only 86 percent of their own needs, with netimports amounting to some 265 million tonnes annually —almost three times present levels.

1964-66 1974-76 1984-86 1997-99 2015 2030

3 000

2 000

1 000

0

Mill

ion

tonn

esCoarse grains WheatRice (milled)

34

price support and protectionist policies. VariousEU policy reforms have since brought domesticprices broadly into line with international prices,but the EU is likely to remain a significant netexporter even if its trade is further liberalized.

The transition economies are anotherpossible source of future exports. Indeed, theyare already moving into surplus. Spare land isplentiful in parts of Eastern Europe and Russia,and the scope for increasing productivity byreducing losses and raising yields is high.FAO’s projections suggest that the transitioncountries could be net exporters of 10 milliontonnes of cereals a year by 2015 and 25 milliontonnes by 2030.

dependence of developing countries (excludingexporters Argentina and Uruguay) shouldcontinue to grow, with net wheat importsexpected to rise from 72 millions tonnes a yearin 1997-99 to 160 million tonnes in 2030.

Rice. This crop is overwhelmingly used for directhuman consumption, and made up 21 percent ofthe world’s cereal consumption by weight in1997-99. Average consumption per person indeveloping countries has been levelling off sincethe mid-1980s, reflecting economic developmentand income growth in major East Asiancountries. It has, however, been growing insome regions, including South Asia, where it isstill low. Consumption is expected to grow moreslowly in the future than in the past. Indeed,average consumption per person in developingcountries may well start to decline during theperiod 2015 to 2030. This will ease pressures onproduction, but given the slow yield growth ofrecent years, maintaining even modestincreases in production will be a challenge toresearch and irrigation policy.

Coarse grains. These include maize, sorghum,barley, rye, oats and millet, and some regionallyimportant grains such as tef (Ethiopia) or quinoa(Bolivia and Ecuador). About three-fifths ofworld consumption of coarse grains is used foranimal feed, but where food insecurity is highthese crops remain very important in directhuman consumption: in sub-Saharan Africa,80 percent of the grain harvest is used in thisway. Consumption of coarse grains has beenrising fast, driven mainly by growing use asanimal feed in developing countries. In thefuture, consumption may well grow faster thanthat of rice or wheat, in line with the growth ofthe livestock sector. Developing countries willaccount for a rising share of world production,from less than half at present to just underthree-fifths by 2030.

Oilcrops. This sector has been one of the world’smost dynamic in recent decades, growing atalmost double the speed of world agriculture asa whole. It covers a wide range of crops used notonly for oil but also for direct consumption,

Prospects for key cropsFood staplesWheat. The world’s major cereal crop accountedfor 31 percent of global cereal consumption in1997-99. A growing share of wheat is used foranimal feed in the industrial countries — 45percent of total use in the EU. Wheat use perperson in developing countries, overwhelminglyfor food, has continued to rise, and mostdeveloping countries are increasingly dependenton imports. Among the net importers are somemajor wheat producers, such as Egypt, IslamicRepublic of Iran, Mexico and Brazil. Over thecoming years wheat consumption is expected toincrease in all regions, including the transitioncountries as their consumption revives. Inseveral rice-eating countries, increases in wheatconsumption go hand in hand with constant ordeclining consumption of rice. The import

The transition countries becamelarge net importers of cereals overthe two decades to the early 1990s.They have since reversed this trendand could be net exporters of10 million tonnes annually by 2015and 25 million tonnes by 2030.

35

animal feeds and a number of industrial uses.Oil-palm, soybean, sunflower and rapeseedaccount for almost three-quarters of worldoilseed production, but olive oil, groundnut,sesame and coconut are also significant. Therapid expansion of production has meant thatoilcrops have accounted for a huge share of theexpansion of the world’s agricultural land, witha net increase of 75 million ha between 1974-76and 1997-99 — this at a time when the areaunder cereals shrank by 28 million ha.

With their high energy content, oilcrops haveplayed a key role in improving food energysupplies in developing countries. Just over oneout of every five kilocalories added toconsumption in the developing countries in thepast two decades originated in this group ofproducts. This trend looks set to continue andindeed intensify: 45 out of every 100 additionalkilocalories in the period to 2030 may comefrom oilseeds. The rapid growth in consumptionover the past few decades was accompanied bythe emergence of several developing countries— China, India, Mexico and Pakistan, amongothers — as major and growing net importers ofvegetable oils. The result has been that thetraditional surplus of the vegetable oils/oilseedscomplex in the balance of payments of thedeveloping countries has turned into a deficit in

recent years. This has happened despite thespectacular growth of exports from a few de-veloping countries that have come to dominatethe world export scene, namely Malaysia andIndonesia for palm oil and Brazil and Argentinafor soybean. In most other developing countries,the trend towards increased imports can beexpected to continue.

Roots, tubers and plantains. World consumptionof these crops as human food has been on thedecline, but for 19 countries — all of them inAfrica — they still provide more than a fifth, andsometimes as much as half, of all food energy.Cassava predominates in humid Central andWest Africa and in the United Republic ofTanzania and Madagascar, while plantains aremost important in Rwanda and cassava andsweet potato in West Africa and Burundi. Sincemost of these countries have low foodconsumption overall — less than 2 200 kcalper person per day — these crops play a crucialrole in food security. In the period to 1997-99,Ghana and Nigeria made considerable ad-vances in food security through the increasedproduction of these crops, but in most of theother 17 countries per capita consumptionstagnated or declined. The decline in worldconsumption of traditional roots and tubers has

Expansion of area cultivated by crop, 1974-76 to 1997-99

Source: FAO data

Oilcrops

Fibre crops

Pulses

Cereals

Million ha

Developing countriesRest of the world

-80 -60 -40 -20 0 20 40 60 80

Sugarcane and beet

Roots and tubers

36

been accompanied by a gradual shift towardspotato in some areas. A large part of this trendis explained by China, where millions of farmersand consumers have switched from sweetpotato to potato.

Average demand for roots, tubers andplantains is projected to rise again in develop-ing countries, with sweet potato and potatobecoming particularly important as animalfeeds. During the 1990s the use of importedcassava as feed in the EU skyrocketed becauseof high domestic prices for cereals, only to fallas reform of the Common Agricultural Policyreduced cereal prices. Cassava productionfor export as feed has been a major factorin expansion of the area cultivated in suchcountries as Thailand, a trend that is oftenassociated with deforestation.

Traditional export cropsBeyond these basic food crops, the agricultureand often the whole economy of many develop-ing countries depend to a high degree on theproduction of one or a few commodities destinedprincipally for export. In this category arecommodities such as banana, sugar, naturalrubber and tropical beverages (tea, coffee andcocoa).

The distinction between export crops andthose for the domestic market is not alwaysneat, either across or even within the developingcountries. For example, sugar is the export croppar excellence for Mauritius and Cuba but amajor import for Egypt, Indonesia and severalother countries. Vegetable oils and oilseeds(especially palm oil and soybean) are major andrapidly growing export crops for severalcountries (including Malaysia, Indonesia,Argentina and Brazil), but are heavily importedby countries such as India and China. Coffee andcocoa share the characteristic of beingproduced exclusively in the developing countriesbut consumed predominantly in the industrialones. Natural rubber used to belong to thiscategory, but more of it is now consumed in thedeveloping countries (half of world consumption,up from a quarter in the mid-1970s) as theyindustrialize. Cotton is in the same class, butmore so, with the developing countries having

turned into large net importers following thegrowth of their textiles industries and exports.

The economies of countries dependent onexports of these commodities are subject tochanging conditions in the world market. Slowgrowth in world demand, combined withincreasing supplies from the main producingand exporting countries, which compete withone another, have led to declining and widelyfluctuating prices in the markets for severalcommodities. This has been particularly markedfor coffee in recent years: per capitaconsumption in the industrial countries,accounting for two-thirds of world consumption,has been nearly constant for two decades, ataround 4.5 kg, while production has increased,with several new countries, such as Viet Nam,entering the market. The result is that the priceof Robusta coffee has nosedived, falling toUS$0.50 per kg by January 2002, one-fifth ofwhat it was in the mid-1990s.

For sugar and a few other commoditiesexperiencing faster growth in consumption,mainly in the developing countries, the earningsof developing country exporters have beencurbed by policies restricting access to markets,including policies favouring substitutesweeteners such as corn syrup. Such policiesare very common in the main industrialcountries that are, or used until recently to be,large importers. The EU used policies of thiskind to turn itself from a large net importer,which it was until the second half of the 1970s,into a large net exporter today.

Looking into the future, the scope for growthin world demand and in the exports ofdeveloping countries is greatest for thosecommodities whose consumption is growingfairly rapidly in the developing countriesthemselves, several of which are likely tobecome large importers. In this category belongsugar and vegetable oils and, to a lesser extent,natural rubber and tea. Banana and cocoa arealso becoming substantial import items inseveral developing countries, a trend thatshould intensify in the coming decades. In thesetwo commodities, but also in others such ascitrus and fruits and vegetables in general,there is still scope for growth in consumption

37

and imports in the industrial countries. Inparallel, the transition economies will play agrowing role as importers of tropical products, aprocess already under way. In contrast, the highconcentration of coffee markets in the industrialcountries, together with negligible growth inpopulation and per capita consumption here, donot augur well for the expansion of productionand exports in this commodity: a continuation ofthe current slow growth, of no more than 1.2percent yearly, seems the most likely outcome.

In conclusion, the agriculture, overalleconomy and food security of several developingcountries will continue to depend on severalcrops for which the world market conditions arenot only volatile but also, on balance, on adeclining trend as regards real prices. Thesecharacteristics of the market could be highly

Two countries, China and India, have been thefocus of fears that the world might run intoserious food shortages. Together they arehome to over a third of the world’s population.

Some analysts feared that China wouldbecome a permanent importer on an ever-increasing scale. This would raise food priceson the world market, reducing the ability ofother poor countries and people to buy food.

China (not including Taiwan Province) was alarge importer of cereals in most years up to1991, with typical net imports of 5 to 15 milliontonnes a year. However, in the 1990s thecountry turned this situation around. In allbut two of the eight years from 1992 to 1999,China was a net exporter of cereals, evenwhile domestic use rose from 295 to 310 kgper person per year.

In the 1960s and early 1970s it becamecommonplace to warn of impending famine inIndia and in South Asia as a whole. In the mid-1960s the region imported 10 million tonnes ofcereals a year — 11 percent of its consumption— but even so its cereal use per person waslow, only 146 kg per year.

Thirty-three years on, the region’s popula-tion had doubled and cereal use had risen to163 kg per person per year. Yet thanks to thegreen revolution, imports were only a third oftheir mid-1960s levels, running at less than 2percent of consumption. India had become asmall net exporter in most years since the late1970s. However, per capita use is still low inthe region, reflecting, among other things, thepersistence of widespread poverty and thevery low use of cereals as feed, given the lowconsumption of meat. If consumption hadgrown faster, it is an open question whetherimports would have been contained at suchlow levels.

The scares that went away

China: from net importer to net exporterof cereals

Source: FAO data

-25 -20 -15 -10 -5 0 5 10

Million tonnes

1974

1979

1984

1989

1994

1999

Net imports Net exports

38

detrimental to the development prospects ofthese countries. Countries that have failed in thepast to diversify their economies and reduce theirdependence on these traditional export cropshave had a growth record well below average.Their challenge is to change this scenario in thefuture. The experiences of countries such asMalaysia suggest that this can be done.

The environmental issues must be addressedA frequently voiced concern is that theadditional production required to meet worlddemand will be unsustainable, involvingdeepening levels of environmental damage thatwill undermine the natural resource base.

In the developed countries this concernrelates mainly to the increased use of fertilizersand other chemical inputs. Past increases haveled to serious problems of water and airpollution, and so will future ones unlesscounter-measures are taken.

Although the overuse of pesticides and otherchemical inputs is a problem in some high-potential areas, increasing production in thedeveloping world for the most part entailsenvironmental risks of a different kind:• In extensive farming and ranching systems,

the major risks are soil erosion, soil miningand deforestation, leading to declining yieldsand desertification.

• In intensive irrigated farming systems, themajor risks are salinization, waterlogging andwater scarcities.

Some methods for increasing and sustainingcrop production while minimizing environmentaldamage are already known and practised insome areas. Such methods need to beresearched and extended for all environments,with appropriate policies that will encouragetheir rapid spread also being devised andimplemented.

Although future demand for food and cashcrops will grow more slowly than in the past,meeting this demand will still require thecontinued expansion of farmland, togetherwith improvements in yield based on new plant

varieties and farming technologies.Questions have been raised about all of

these factors. Is there enough suitable landand water to expand the rainfed and irrigatedarea as much as will be needed, or is the worldrunning short of these vital inputs? Is therescope for the higher yields that will be required,or are yields approaching limits that cannot bebreached? Can biotechnology deliver a newgeneration of higher-yielding crops bettersuited to difficult environments? And arethere approaches to farming that can increaseand sustain production while improvingconservation? The following sections willexamine these questions.

Land, water and crop yields

The sources ofproduction growth

Increases in crop production derive from threemain sources: expansion of arable land,increases in cropping intensity (the frequencywith which crops are harvested from a givenarea) and improvements in yield.

Since the early 1960s, yield improvementshave been by far the largest source of increasein world crop production, accounting for almostfour-fifths or 78 percent of the increasebetween 1961 and 1999. A further 7 percent ofthe increase came from increased croppingintensity, while a mere 15 percent came fromexpansion of the arable area.

Yield improvement was by far the largestfactor not just in the developed world but also inthe developing countries, where it accounted for

39

70 percent of increased production. Expansionof the area cultivated accounted for just under aquarter of production growth in these countries.However, in areas with more abundant land,area expansion was a larger contributing factor.This was especially the case in sub-SaharanAfrica, where it accounted for 35 percent, and inLatin America, where the figure reached 46percent.

The projections suggest that these broadtrends for the developing countries willcontinue, at least until 2030: land expansion isexpected to account for 20 percent of productiongrowth, yield improvements for about 70percent and increased cropping intensity for theremainder. In sub-Saharan Africa and LatinAmerica, land expansion will still be important,but it is likely to be increasingly outweighed byyield increases.

The FAO study indicates that, for the worldas a whole, there is enough unused productivepotential, in terms of land, water and yieldimprovements, to meet the expected growth ineffective demand. However, this is a global

conclusion and there are several strongqualifications to bear in mind:• Effective demand expresses people’s

purchasing power rather than the real needfor food: wealthy consumers may indulge toexcess, while the very poor may not be able toafford even basic foods.

• Data suggesting that food is getting cheapermay be flawed, because they do not reflect theenvironmental costs of expanding andintensifying agriculture; moreover, the failureto internalize resource costs may curbinvestment in agricultural research, holdingback the potential for future growth in yields.

• Land or water scarcities and other problemswill most certainly continue to arise at countryand local levels, with serious consequencesfor poverty and food security.

Land resources

Is there enough potential cropland for futureneeds?It is often suggested that the world may beheading towards shortages of suitableagricultural land. FAO studies suggest that thiswill not be the case at the global level, althoughin some regions and areas there are alreadyserious shortages, and these may worsen.

Less new agricultural land will be openedup than in the past. Over the period 1961-63to 1997-99 the expansion of arable land indeveloping countries totalled 172 million ha, anincrease of 25 percent. In the next 30 years anincrease of only 120 million ha, or 13 percent,will be required. Adding an extra 3.75 million ha

In future, 80 percent of increased crop production indeveloping countries will have to come from intensification:higher yields, increased multiple cropping and shorter fallowperiods.

Sources of growth in production,1961 to 1999

Source: FAO data

100

80

60

40

20

0

%

Latin

Am

eric

a

and

Carib

bean

Wor

ldDe

velo

ping

coun

trie

sSo

uth

Asia

YieldCropping intensityArea expansion

40

a year may seem a daunting task — but it is lessthan the rate of 4.8 million ha a year that wasactually achieved over the period 1961-63 to1997-99. A slowdown in expansion is expected inall regions, but this is mainly a reflection of theslower growth in demand for crops.

There is still potential agricultural land thatis as yet unused. At present some 1.5 billion haof land is used for arable and permanent crops,around 11 percent of the world’s surface area. Anew assessment by FAO and the InternationalInstitute for Applied Systems Analysis (IIASA) ofsoils, terrains and climates compared with theneeds of and for major crops suggests that afurther 2.8 billion ha are to some degreesuitable for rainfed production. This is almosttwice as much as is currently farmed.

Of course, much of this potential land is inpractice unavailable, or locked up in othervaluable uses. Some 45 percent is covered inforests, 12 percent is in protected areas and

Fears of an imminent crunch betweenpopulation growth and land availabilityare unwarranted. Most future growthin crop production will stem fromimproved yields. In some countries,however, land shortages may bite.

3 percent is taken up by human settlements andinfrastructure. In addition, much of the landreserve may have characteristics that makeagriculture difficult, such as low soil fertility,high soil toxicity, high incidence of human andanimal diseases, poor infrastructure, and hillyor otherwise difficult terrain.

The pool of unused suitable cropland is veryunevenly distributed. By the end of the twentiethcentury, sub-Saharan Africa and Latin Americawere still farming only around a fifth of theirpotentially suitable cropland. More than half theremaining global land balance was in just sevencountries in these two regions: Angola,Argentina, Bolivia, Brazil, Colombia, DemocraticRepublic of Congo and the Sudan. At the otherextreme, in the Near East and North Africa87 percent of suitable land was already beingfarmed, while in South Asia the figure was noless than 94 percent. In a few countries of theNear East and North Africa, the land balance isnegative – that is, more land is being croppedthan is suitable for rainfed cropping. This ispossible where, for example, land that is toosloping or too dry for rainfed crops has beenbrought into production by terracing or irrigation.

More than 80 percent of the projectedexpansion in arable area is expected to takeplace in sub-Saharan Africa and Latin America.Although there is still surplus land in theseregions, the expansion may involve cutting backon long rotation and fallow periods. If fertilizer

Cropland in use and total suitable land (million ha)

Sources: FAO data and Fischer et al. (2000)

Industrial countries Transition countriesSouth Asia Near East andNorth Africa

East AsiaSub-Saharan AfricaLatin America and Caribbean

1066 1031

366 220 99

874

497

203 228 232 207 387 265

Arable land in use, 1997-99

Total suitable for rainfedcrop production

86

41

use does not rise to compensate, this may resultin soil mining and stagnant or declining yields.

In contrast, in South Asia and the Near Eastand North Africa, where almost all suitable landis already in use, there will be next to noexpansion in area. By 2030 the Near East andNorth Africa will be using 94 percent of itssuitable cropland, with a remaining surplus ofonly 6 million ha. In South Asia the situation willbe even tighter, with 98 percent already in

The projections suggest that thearable area in developing countrieswill increase by almost 13 percent or120 million ha over the years from1997-99 to 2030.

cultivation. In South and East Asia, more than80 percent of the increase in production willhave to come from yield increases, since only5 or 6 percent can come from expansion of thearable area.

Cropping intensities will rise in all develop-ing regions, on average from 93 percent to99 percent. This will occur through the shorten-ing of fallow periods and increased multiplecropping, made possible partly by growth in theirrigated area.

Is land becoming scarcer?There is widespread concern that the world maybe running out of agricultural land. The trendtowards scarcity associated with populationgrowth is aggravated by the conversion offarmland to urban uses, by land degradationand by other factors.

Certainly, much farmland is being taken overfor non-agricultural uses. Assuming a require-ment for housing and other infrastructure of40 ha per 1000 people, then world populationgrowth between 1995 and 2030 implies the needfor an additional 100 million ha of such non-agricultural land. Since most urban centres aresited on fertile agricultural land in coastal

plains or river valleys, when they expand theytake up more of this prime land. In China alone,more than 2 million ha were taken out ofagriculture in the ten years to1995.

Despite these losses, there is little evidenceto suggest that global land scarcities lie ahead.Between the early 1960s and the late 1990s,world cropland grew by only 11 percent, whileworld population almost doubled. As a result,cropland per person fell by 40 percent, from0.43 ha to only 0.26 ha. Yet, over this sameperiod, nutrition levels improved considerablyand the real price of food declined.

The explanation for this paradox is thatproductivity growth reduced the amount of landneeded to produce a given amount of food byaround 56 percent over this same period. Thisreduction, made possible by increases in yieldsand cropping intensities, more than matchedthe decline in area per person, allowing foodproduction to increase.

Land scarcity and the problems associatedwith it do of course exist at country and locallevels, with serious consequences for povertyand food security. In many places these arelikely to worsen unless remedial action is taken.

How serious is land degradation?Land degradation is the process by which thesoil’s current or future capacity to produce islowered by chemical, physical or biologicalchanges. Some analysts claim that acceleratingland degradation will offset productivityimprovements, while others believe theseriousness of this problem has been greatlyoverstated.

The truth is that the area of degraded land isnot known with much precision. Its assessmentis often based on expert judgement rather thanobjective measurement. For India alone,estimates by different public authorities varyfrom 53 million ha right up to 239 million ha.

The most comprehensive survey to date, theGlobal Assessment of Land Degradation(GLASOD), is now over ten years old. GLASODestimated that a total of 1964 million ha weredegraded, 910 million ha to at least a moderatedegree (with significantly reduced productivity)and 305 million ha strongly or extremely so (no

42

longer suitable for agriculture). Water erosionwas the most common problem, affectingalmost 1 100 million ha, followed by winderosion, which affected almost 600 million ha.

The impact of degradation on productivity isalso hard to assess. Its seriousness varieswidely from site to site over even smalldistances, and at the same site according tolocal weather, vegetation and farmingtechniques. Degradation is a slow process thatcan be masked by applying additional fertilizeror by changing the crops grown. GLASODreported in 1991 that almost all farmland inChina was degraded, yet between the early1960s and mid-1990s China tripled her riceproduction and increased her wheat productionsevenfold. Some studies suggest annualaverage losses in cropland productivity may bequite small, averaging only 0.2 to 0.4 percent ayear.

Degradation also has off-site costs, such asthe siltation of streambeds and dams, flooddamage, loss of fisheries and the eutrophicationof lakes and coastal waters. These costs areoften greater than on-site costs. However, theoff-site effects of degradation are not allnegative: losses in one place may result in gainselsewhere, as when soil eroded from uplandsboosts productivity in the alluvial plains where itis deposited.

Because it is difficult to quantify, the futureprogress of land degradation was not taken intoaccount in the projections made for this study.However, some projected or foreseeable trends,driven primarily by economic forces, will tend toreduce its extent and impact:• About a third of the harvested area in

developing countries in 2030 is expected to beirrigated land, which is generally flat,protected by bunds and little affected by

Human-induced soil degradation in the world

Source: Oldeman et al. (1991)

Water erosion

Wind erosion

Chemical deterioration

Physical deterioration Stable terrain

Non-used wasteland

Water bodies

Soil degradation types Other symbols

Severe degradation

43

erosion. A quarter of the rainfed land by thattime will have slopes of less than 5 degrees,also generally not prone to heavy erosion.

• The shift in livestock production to moreintensive systems will take some pressure offdryland pastures. However, in the developingcountries this will be partly offset by theencroachment of cropland, which will reducethe area remaining for extensive grazing.

• As people leave rural areas for urban centres,and farming for non-farming occupations,steep slopes and other marginal land will tendto be abandoned and will revert to scrub and

forest. This process has already occurredrapidly in some European countries. In Italy,some 1.5 million ha were abandoned in the1960s, 70 percent of which was sloping land.In some provinces, agricultural landdecreased by 20 percent.

Other trends tending to reduce landdegradation are likely, but their extent andintensity will depend heavily on the spread ofimproved agricultural and conservationpractices, without which land degradation mayworsen in many areas. The main practices andtheir potential impact are:

• Sloping land is particularly prone to watererosion, especially in wet areas whereslopes exceed 10 to 30 percent and conser-vation measures are lacking. In Nepal, forexample, some 20 to 50 tonnes of soil per haare estimated to be eroded each year fromfields in the hills and mountains, while up to200 tonnes per ha per year may be lost insome highly degraded watersheds. Cropyields in these areas fell by 8 to 21 percentin the 25 years to 1995. Around 45 percent ofthe world’s agricultural land has slopes ofmore than 8 percent, and out of this total9 percent has very steep slopes of over30 percent.

• Desertification, a term referring to landdegradation in arid and semi-arid areas,received a great deal of attention during the1970s and 1980s, when it was believed thatdeserts such as the Sahara were spreadingirreversibly. Estimates suggested that up to70 percent of the world’s 3.6 billion ha ofdrylands were degraded. Since then remotesensing has established that desert marginsebb and flow with natural climate changes,while studies on the ground are showing theresilience of crop and livestock systems andthe adaptiveness of farmers and herders.

• Salinization occurs in irrigated areas, usuallywhen inadequate drainage causes salts toconcentrate in the upper soil layers whereplants root. It is a problem mainly in the aridand semi-arid zones, where 10 to 50 percentof the irrigated area may be affected.Salinization can cause yield decreases of10 to 25 percent for many crops and mayprevent cropping altogether when it issevere. It is estimated that 3 percent of theworld’s agricultural land is affected. In EastAsia, however, the proportion is 6 percentand in South Asia 8 percent. For the arid andsemi-arid tropics as a whole, 12 percent ofagricultural land may be affected.

• Nutrient mining is also a serious problem.Farmers often use insufficient fertilizer toreplace the nitrogen, phosphorus andpotassium (NPK) harvested with their cropsand lost through leaching, while traceelements, such as iron or boron, may also bedeficient. A detailed study of Latin Americaand the Caribbean found nutrient depletionin all areas and for almost all crops exceptbeans. Net NPK losses in the region in1993-95 amounted to 54 kg per ha per year.Another study suggested net losses of 49 kgper ha per year in sub-Saharan Africa.

Principal types of land degradation

44

• No-till/conservation agriculture (NT/CA),which can maintain year-round soil cover andincrease organic matter in soils, therebyreducing water and wind erosion.

• Increased fertilizer consumption and moreefficient fertilizer use, which will reduceerosion by increasing root growth and groundcover.

• The use of irrigation, water harvesting,drought-tolerant crops and grazing-tolerantgrasses, which will improve crop andvegetation cover and reduce erosion indrylands.

• The cultivation of legumes, which can addnitrogen to soils and improve their stabilityand texture in mixed crop-livestock farmingsystems.

Irrigation and waterresources

A large share of the world’s crops is alreadyproduced under irrigation. In 1997-99, irrigatedland made up only about one-fifth of the totalarable area in developing countries. However,because of higher yields and more frequentcrops, it accounted for two-fifths of all cropproduction and close to three-fifths of cerealproduction.

This share is expected to increase further inthe next three decades. Based on the potentialfor irrigation, national plans for the sector andthe moisture needs of crops, the developingcountries as a whole can be expected to expandtheir irrigated area from 202 million ha in1997-99 to 242 million ha by 2030. This is a netprojection – that is, it is based on the

There will be no overall shortage ofland or water for irrigation, butserious problems will persist in somecountries and regions.

assumption that land lost due, for example,to salinization and water shortages will becompensated by rehabilitation or by thesubstitution of new areas.

Most of this expansion will occur in land-scarce areas where irrigation is already crucial:South Asia and East Asia, for example, will add14 million ha each. The Near East and NorthAfrica will also see significant expansion. Inland-abundant sub-Saharan Africa and LatinAmerica, where both the need and the potentialfor irrigation are lower, the increase is expectedto be much more modest – 2 million and4 million ha respectively.

Although the projected expansion isambitious, it is much less daunting than whathas already been achieved. Since the early1960s, no less than 100 million ha of newirrigated land have been created. The netincrease projected for the next three decades isonly 40 percent of that. The expected annualgrowth rate of 0.6 percent is less than a third ofthe rate achieved over the past 30 years.

The FAO study did not make projections forirrigation in the developed countries, whichaccount for around a quarter of the world’sirrigated area. Irrigation in this group ofcountries grew very rapidly in the 1970s, but bythe 1990s the pace of growth had slowed to only0.3 percent per year.

Is there enough irrigable land for future needs?As with land in general, it has been suggestedthat the world may soon experience shortagesof land suitable for irrigation. There is concern,too, that vast areas of presently irrigated landmay be severely damaged by salinization. Onceagain, at global level these fears seemexaggerated, though serious problems mayoccur at local level.

FAO studies suggest there is still scope forexpanding irrigation to meet future needs.However, irrigation potential is difficult toestimate accurately, since it depends oncomplex data on soils, rainfall and terrain. Thefigures should therefore be taken only as arough guide. The total irrigation potential indeveloping countries is nevertheless estimatedat some 402 million ha. Of this around half was

45

in use in 1997-99, leaving an unused potential of200 million ha. The projected increase by 2030would take up only 20 percent of this unusedpotential.

In some regions, however, irrigation willcome much closer to its full potential: by 2030,East Asia and the Near East and North Africawill be using three-quarters of their irrigablearea, and South Asia (excluding India) almost90 percent.

Is there enough water?Another frequently voiced concern is that muchof the world is heading for water shortages.Since agriculture is responsible for about 70percent of all the water withdrawn for humanuse, it is feared that this will affect the future offood production. Once again, at global levelthere seems to be no cause for alarm, but at thelevel of some localities, countries and regions,serious water shortages appear highly likely toarise.

The assessment of potential irrigated landused for this report already takes into accountthe limitations imposed by the availability ofwater. The renewable water resources availablein a given area consist of the amount added byrainfall and incoming river flow, minus theamount lost through evapotranspiration. Thismay vary greatly across regions. For example, inan arid region such as the Near East and NorthAfrica, only 18 percent of rainfall and incomingflows remain after evapotranspiration, whereasin humid East Asia the share is as high as 50percent.

The water used for irrigation includes,besides that actually transpired by the growingcrop, all the water applied to it, which may beconsiderable in the case of crops that areflooded, such as rice. In addition, there are thelosses through leakage and evaporation on theway to the fields, and the water that drains awayfrom the fields without being used by the crop.The ratio between the amount of water actuallyused for crop growth and the amount withdrawnfrom water sources is known as water useefficiency.

There are large regional differences in wateruse efficiency. Generally, efficiency is higher

where water availability is lower: in LatinAmerica, for example, it is only 25 percent,compared with 40 percent in the Near East andNorth Africa and 44 percent in South Asia.

In the developing countries as a whole, onlyabout 7 percent of renewable water resourceswere withdrawn for irrigation in 1997-99. Butbecause of differences in efficiency and in wateravailability, some regions were using a muchhigher proportion than others. In sub-SaharanAfrica, where irrigation is less widespread, only2 percent were used, and in water-rich LatinAmerica a mere 1 percent. In contrast, the figurein South Asia was 36 percent and in the NearEast and North Africa no less than 53 percent.

The projections for developing countriesimply a 14 percent increase in water withdrawalsfor irrigation by 2030. Even then, they will beusing only 8 percent of their renewable waterresources for irrigation. The shares in sub-Saharan Africa and Latin America will remainvery small.

Water availability is considered to become acritical issue only when 40 percent or more of

Irrigation and water resources, 1997-99 to 2030

Source: FAO data and projections

The projections for developingcountries imply a 14 percent increasein water withdrawals for irrigation by2030. One in five developing countrieswill face water shortages.

Near East andNorth Africa

Latin Americaand Caribbean

East Asia

South Asia

Sub-SaharanAfrica

Renewable water and water withdrawal (km3)

0 14 00012 00010 0008 0006 0004 0002 000

Renewable water resourcesWater withdrawal, 2030Water withdrawal, 1997-99

46

renewable water resources are used forirrigation. This is the level at which countriesare forced to make difficult choices betweentheir agricultural and their urban water supplysectors. By 2030 South Asia will be using thislevel, and the Near East and North Africa noless than 58 percent.

Out of 93 developing countries studied forthis report, 10 were already using more than40 percent in 1997-99 and another 8 were usingmore than 20 percent — a threshold which canbe considered to indicate impending waterscarcity. By 2030 two more countries will havecrossed this lower threshold and one in fivedeveloping countries will be suffering actual orimpending water scarcity.

Two countries, Libyan Arab Jamahiriya andSaudi Arabia, are already using more water forirrigation than their annual renewable re-sources, by drawing on fossil groundwaterreserves. Groundwater mining also occurs atlocal levels in several other countries of theNear East and North Africa, South Asia andEast Asia. In large areas of India and China,ground-water levels are falling by 1 to 3 metresper year, causing subsidence in buildings,intrusion of seawater into aquifers and higherpumping costs.

In these countries and areas, policy changesand investments will be needed to improve theefficiency of water use, together with innovationsto improve the capture and infiltration of water,such as water harvesting, tree planting and so on.

Potential for yield growth

Growth rates have slowed in the past decadeMost future increases in crop production will beachieved through improved yields. Yieldadvances have been uneven over the past threedecades.

Global cereal yields grew rapidly between1961 and 1999, averaging 2.1 percent a year.Thanks to the green revolution, they grew evenfaster in developing countries, at an average rateof 2.5 percent a year. The fastest growth rates

were achieved for wheat, rice and maize which,as the world’s most important food staples, havebeen the major focus of international breedingefforts. Yields of the major cash crops, soybeanand cotton, also grew rapidly.

At the other end of the scale, yields of millet,sorghum and pulses saw only slow growth.These crops, grown mainly by resource-poorfarmers in semi-arid areas, are ones for whichinternational research has not so far come upwith varieties that deliver large yield gainsunder farm conditions. There have been usefulincremental gains, however, and farmers’ yieldsare more stable than they used to be, thanks tothe introduction of traits such as early maturity.

Overall growth in cereal yields slowed in the1990s. Maize yields in developing countriesmaintained their upward momentum, but gainsin wheat and rice slowed markedly. Wheat yieldsgrew at an average of 3.8 percent per yearbetween 1961 and 1989, but at only 2 percent ayear in 1989 to 1999. For rice the respectiverates fell by more than half, from 2.3 percent to1.1 percent. This largely reflects the slowergrowth in demand for these products.

Is projected yield growth realistic?The slower growth in production projected forthe next 30 years means that yields will notneed to grow as rapidly as in the past. Growth inwheat yields is projected to slow to 1.1 percent ayear in the next 30 years, while rice yields areexpected to rise by only 0.9 percent per year.

Nevertheless, increased yields will berequired — so is the projected increase feasible?One way of judging is to look at the difference in

Growth in wheat and rice yieldsslowed markedly in the 1990s.Rice yields rose at an average of2.3 percent per year between 1961and 1989, but between 1989 and1999 this figure fell by more than half,to 1.1 percent.

47

performance between groups of countries.Some developing countries have attained veryhigh crop yields. In 1997-99, for example, thetop performing 10 percent had average wheatyields more than six times higher that those ofthe worst performing 10 percent and twice ashigh as the average in the largest producers,China, India and Turkey. For rice the gaps wereroughly similar.

National yield differences like these are dueto two main sets of causes:• Some of the differences are due to differing

conditions of soil, climate and slope. InMexico, for example, much of the country isarid or semi-arid and less than a fifth of theland cultivated to maize is suitable forimproved hybrid varieties. As a result, thecountry’s maize yield of 2.4 tonnes per ha isnot much more than a quarter of the UnitedStates average. Yield gaps of this kind, causedby agro-ecological differences, cannot benarrowed.

• Other parts of the yield gap, however, are theresult of differences in crop managementpractices, such as the amount of fertilizerused. These gaps can be narrowed, if it iseconomic for farmers to do so.

To find out what progress in yields is feasible,it is necessary to distinguish between the gapsthat can be narrowed and those that cannot.A detailed FAO/IIASA study based on agro-ecological zones has taken stock of the amountof land in each country that is suitable, invarying degrees, for different crops. Using thesedata it is possible to work out a nationalmaximum obtainable yield for each crop.

This maximum assumes that high levels ofinputs and the best suited crop varieties areused for each area, and that each crop is grownon a range of land quality that reflects thenational mix. It is a realistic figure because it isbased on technologies already known and doesnot assume any major breakthroughs in plantbreeding. If anything, it is likely to under-estimate maximum obtainable yields, becausein practice crops will tend to be grown on theland best suited for them.

The maximum obtainable yield can then becompared with actual national average yield to

Source: FAO data and projections

Crop yields in developing countries, 1961 to 2030

give some idea of the yield gap that can bebridged. The study showed that even a techno-logically progressive country such as France isnot yet close to reaching its maximum obtainableyield. France could obtain an average wheatyield of 8.7 tonnes per ha, rising to 11.6 tonnesper ha on her best wheat land, yet her actualaverage yield today is only 7.2 tonnes per ha.

Similar yield gaps exist for most countriesstudied in this way. Only a few countries are

The slower growth in productionprojected for the next 30 years meansthat yields will not need to grow asrapidly as in the past. Growth inwheat yields is projected to slow to1.1 percent and in rice yields to only0.9 percent per year in developingcountries.

5

4

3

2

1

0

Yiel

d (t

onne

/ha)

Pulses

1961-63

1997-99

2030

RiceMaize

Wheat

Soybean

Seedcotto

n

Groundnut

Sorghum

Millet

48

Exploitable yield gaps for wheat: actualversus obtainable yield

Sources: FAO data and Fischer et al. (2000)

actually achieving their maximum obtainableyield.

When real prices rise, there is every reasonto believe that farmers will work to bridge yieldgaps. In the past, farmers with good access totechnologies, inputs and markets haveresponded very quickly to higher prices.Argentina, for example, increased her wheatproduction by no less than 68 percent in just oneyear (1996), following price rises, although thiswas done mainly be extending the area underwheat. Where land is scarcer, farmers respondby switching to higher-yielding varieties andincreasing their use of other inputs to achievehigher yields.

It seems clear that, even if no more newtechnologies become available, there is still

scope for increasing crop yields in line withrequirements. Indeed, if just 11 of the countriesthat produce wheat, accounting for less than two-fifths of world production, were to bridge only halfthe gap between their maximum obtainable andtheir actual yields, then the world’s wheat outputwould increase by almost a quarter.

The outcome of research is alwaysuncertain, particularly if it is strategic or basicin nature. However, if new technologies dobecome available through the genetic and otherresearch currently under way, this could raiseyield ceilings still further, while possibly alsoreducing the environmental costs of cropproduction.

Given the right economic incentives, worldagriculture will respond to the demand

Fertilizer use, 1961 to 1999

Source: FAO data

Nut

rien

t app

licat

ions

(kg/

ha)

1961-63 1997-99 1961-63 1997-99 1961-63 1997-99 1961-63 1997-99 1961-63 1997-99 1961-63 1997-99 1961-63 1997-99

East A

sia

Industrial

countri

es

South A

sia

Near E

ast a

nd

North

Africa

Latin Am

erica

and C

aribb

ean

Tran

sition

countri

es

Sub-Sah

aran

Africa

200

100

0

Nitrogen

Phosphates

Potassium

German

y

Ukraine

United Stat

es

Turk

ey

Ethiopia

Brazil

Argentin

a

United R

epublic

of Tan

zania

8

6

4

2

0

Yiel

d (t

onne

s/ha

)

ActualObtainable

49

The role of technology

The development and dissemination of newtechnology is an important factor determiningthe future of agriculture. The FAO studyinvestigated three areas that are particularlycritical, namely biotechnology, technologies insupport of sustainable agriculture, and thedirections that should be taken by futureresearch.

Biotechnology: issuesand prospects

What is the current role of biotechnology?For thousands of years, human beings havebeen engaged in improving the crops andanimals they raise. Over the past 150 years,

One of the major ways in which farmers canincrease yields is by applying more fertilizer.A third of the increase in world cereal produc-tion in the 1970s and 1980s has been attributedto increased fertilizer use. In India this figurerises to half.

The level of fertilizer use varies enormouslybetween regions. North America, WesternEurope and East and South Asia accounted forfour-fifths of world fertilizer use in 1997-99.The highest rates, averaging 194 kg of nutri-ents per ha, were applied in East Asia, followedby the industrial countries with 117 kg per ha.At the other end of the scale, farmers in sub-Saharan Africa applied a mere 5 kg per ha.

World fertilizer consumption grew rapidlyin the 1960s, 1970s and 1980s, but slowedconsiderably in the 1990s. The slowdown in

Fertilizer: use will continue to rise, but slowly

industrial countries was due mainly to reducedgovernment support for agriculture and in-creased concern over the environmental impact.In transition countries, fertilizer consumptionalso fell rapidly, but for different reasons,namely recession and restructuring. Even indeveloping countries, the growth rate of ferti-lizer use in the 1990s was less than half of thatseen in earlier decades.

This slower growth is projected to continue.Global fertilizer consumption is expected togrow by an average 1 percent per year over thenext three decades (a little faster in developingcountries and a little slower in developed). Thefastest growth rates are expected in sub-Saharan Africa. Here fertilizer use is currentlyvery low, so fast growth rates can still meanonly small absolute increases.

expressed in the marketplace, as it has done inthe past. Of course, many poor farmers inmarginal environments will be in a position torespond only if they gain access to inputs,markets and technologies, and if the policyenvironment is favourable. In addition, research

must develop the varieties and techniquesneeded to raise yields in difficult environments.These measures are essential if poor farmersand their families are not to find themselvestrapped in poverty.

50

Biotechnology promises great benefitsfor both producers and consumers ofagricultural products, but itsapplications are also associated withpotential risks. The risks and benefitsmay vary substantially from oneproduct to the next and are oftenperceived differently in differentcountries. To reap the full potential ofbiotechnology, appropriate policiesmust be developed to ensure that thepotential risks are accuratelydiagnosed and, where necessary,avoided.

scientists have assisted their efforts bydeveloping and refining the techniques ofselection and breeding. Though considerableprogress has been achieved, conventionalselection and breeding are time-consuming andbear technical limitations.

Modern biotechnology has the potential tospeed up the development and deployment ofimproved crops and animals. Marker-assistedselection, for instance, increases the efficiencyof conventional plant breeding by allowing rapid,laboratory-based analysis of thousands of

individuals without the need to grow plants tomaturity in the field. The techniques of tissueculture allow the rapid multiplication of cleanplanting materials of vegetatively propagatedspecies for distribution to farmers. Geneticengineering or modification — manipulating anorganism’s genome by introducing oreliminating specific genes — helps transferdesired traits between plants more quickly andaccurately than is possible in conventionalbreeding.

This latter technique promises considerablebenefits but has also aroused widespread publicconcerns. These include ethical misgivings,anxieties about food and environmental safety,and fears about the concentration of economicpower and technological dependence, whichcould deepen the technological divide betweendeveloped and developing countries.

The spread of genetically modified (GM)crops has been rapid. Their area increased by afactor of 30 over the 5 years to 2001, when theycovered more than 52 million ha. Considerableresearch to develop more GM varieties is underway in some developing countries. China, forinstance, is reported to have the second largestbiotechnology research capacity after the UnitedStates.

However, the spread so far is geographicallyvery limited. Just four countries account for99 percent of the global GM crop area: theUnited States with 35.7 million ha, Argentinawith 11.8 million ha, Canada with 3.2 million ha

Source: ISAAA (2001)

Area of GM crops for different commodities and countries

Maize19%

Cotton13%

Canola5%

Soybean63%

Commodities(100% = 52.6 million ha)

Argentina22%

Australia1%

South Africa1%

China3%

Canada6%

Countries(100% = 52.6 million ha)

United States69%

51

Biotechnology: potential benefits, risks and concerns

Potential benefits• Increased productivity, leading to higher

incomes for producers and lower prices forconsumers.

• Less need for environmentally harmfulinputs, particularly insecticides.Scientists have developed maize and cottonvarieties incorporating genes from thebacterium Bacillus thuringensis (Bt)which produce insecticidal toxins. Virus-and fungus-resistant varieties are in thepipeline for fruits and vegetables, potatoand wheat.

• New crop varieties for marginal areas,increasing the sustainability of agriculturein poor farming communities. Thesevarieties will be resistant to drought,waterlogging, soil acidity, salinity orextreme temperatures.

• Reduced dependence on managementskills through built-in resistance to pestsand diseases.

• Enhanced food security through reducedfluctuations in yields caused by insectinvasions, droughts or floods.

• Higher nutritional values through higherprotein quality and content as well asincreased levels of vitamins and micro-nutrients (e.g. iodine or beta-caroteneenriched rice).

• Better health value and digestibility.Scientists are developing varieties ofsoybean that contain less saturated fatand more sucrose.

• Production of valuable chemicals andpharmaceuticals at lower cost than ispossible at present. Products envisagedrange from speciality oils and biodegradableplastics to hormones and human antibodies.

Risks and concerns• Products are tailored largely to the needs of

large-scale farmers and industrial process-ing in the developed world, with the resultthat resource-poor farmers in developingcountries will fail to benefit.

• Market concentration and monopoly powerin the seed industry, reducing choice andcontrol for farmers, who will pay ever higherprices for seed. One company alone controlsover 80 percent of the market for GM cottonand 33 percent for GM soybean.

• Patenting of genes and other materialsoriginating in the developing countries.Private-sector companies are able toappropriate without compensation theproducts resulting from the breeding effortsof generations of farmers and from researchconducted in the public sector.

• Technologies that prevent farmers re-usingseed. These require farmers to purchaseseed afresh every season and could inhibitadoption by poor farmers. In the worst case,ignorance of this characteristic could resultin complete crop failure.

• Food safety. This has received addedattention after a potentially allergenic maizevariety that was not registered for food useentered the food chain in the United States.

• The environmental impact of GM crops.There is a risk that inserted genes mayspread to wild populations, with potentiallyserious consequences for biodiversity, orcontaminate the crops of organic farmers.Genes for herbicide resistance couldencourage the overuse of herbicides, whilethose for insect resistance could generateresistance in insects, forcing the use ofmore toxic products to kill them.

and China with 1.5 million ha. The number andtype of crops and applications involved is also

limited: two-thirds of the GM area is planted toherbicide-tolerant crops. All commercially

52

grown GM crops are currently either non-foodcrops (cotton) or are heavily used in animalfeeds (soybean and maize).

Why do we need modern biotechnology?Globally, agricultural production could probablymeet expected demand over the period to 2030even without major advances in biotechnology.However, biotechnology could be a major tool inthe fight against hunger and poverty, especiallyin developing countries. Because it may deliversolutions where conventional breedingapproaches have failed, it could greatly assistthe development of crop varieties able to thrivein the difficult environments where many of theworld’s poor live and farm. Some promisingresults have already been achieved in thedevelopment of varieties with complex traitssuch as resistance or tolerance to drought, soilsalinity, insect pests and diseases, helping toreduce crop failures. Several applications allowresource-poor farmers to reduce their use ofpurchased inputs such as pesticides orfertilizers, with benefits to the environment andhuman health as well as farmers’ incomes.

Most biotechnology is generated andcontrolled by large private-sector companies,which have so far mainly targeted thecommercial farmers who can afford theirproducts. Nevertheless, there is some public-sector work directed towards the needs ofresource-poor farmers. In addition, most of thetechnologies and intermediate productsdeveloped through private-sector researchcould be adapted to solve priority problems inthe developing countries. If the poor of thesecountries are to reap this potential, national andinternational action is needed to foster private-public partnerships that will promote access tothese technologies at affordable prices. This isthe main policy challenge for the future.

What policies are needed to harness thepotential of biotechnology for the poor?In the case of GM crops, most of the commercialapplications developed so far are directedtowards reducing production costs, not towardsmeeting the needs expressed by consumers.The perception of the expected benefits and

potential risks of such crops, and of biotechno-logy as a whole, differ among regions, countries,interest groups and individuals. The urban andlandless poor in developing countries need

One of the most impressive successes inagricultural biotechnology is China’s experi-ence with Bt cotton.

Following research by various public- andprivate-sector partners, Bt cotton wasreleased to the country’s farmers in 1997. Itquickly became very popular, with the areadevoted to it expanding from 2000 ha in thefirst year to 70 000 ha in 2000. The reasons forthis popularity were mainly economic, butthere were important environmental andhuman health benefits too.

In general, cotton is very susceptible topests and normally requires many applicationsof insecticide, which are expensive, require agreat deal of extra labour, and often causehealth problems in farm workers. Farmersusing the new Bt variety needed over 80percent less insecticide than those using non-Bt varieties and only a third as manyapplications. They were able to cut both theirlabour and other input costs. Their yields werealso higher: 3.37 tonnes per ha as opposed to3.18 tonnes with non-Bt cotton. The overallcost of producing a kilogramme of cotton was28 percent lower.

There were positive effects onbiodiversity, with farmers and governmentextension agents reporting a greater variety ofinsects and more beneficial species in fieldswith Bt cotton. In addition there were consid-erable health benefits for the farmers: only5 percent of Bt cotton growers reportedpoisonings, against 22 per cent of growers ofnon-Bt cotton. The overall economic benefitsof Bt cotton were assessed at US$334 millionper year in 1999.

Bt cotton in China: a success story

53

Effects of Bt cotton in China

Source: Huang et al. (2002)

cheaper food. In contrast, for consumers indeveloped countries, where food is plentiful, thehealth and environmental concerns associatedwith biotechnology outweigh the possible costsavings. These consumers will be more inclinedto accept the new products if they can beassured of their safety through appropriateregulatory frameworks.

Greater and better targeted investments inGM research for developing countries will beneeded to ensure that the farmers of thesecountries have access to the resulting new cropvarieties. The focus should shift from pesticide-tolerant crops towards the characteristics thatmatter to resource-poor farmers: improvedresistance or tolerance to drought,waterlogging, salinity and extremetemperatures; improved resistance to pests anddiseases; better nutritional values; and higheryields. Such a shift could be based on newprivate-public partnerships, exploiting thegreater efficiency of private-sector research butunder the guidance of public-sector donors.Research funds could be made available on thebasis of public tenders.

Further change is on the horizonThe rapid progress made in both generating andextending new biotechnology applications,

together with the uncertain public response tothese applications, make it difficult to predictthe long-term prospects for these technologies,including their impact on future production.However, developments in the short term — thenext 3 years or so — are somewhat easier toforesee.

The success of Bt cotton in China has pavedthe way for further expansion of GM crops inthis country, which has considerable potentialfor GM products. China is a major producer ofsoybean, maize and tobacco — all crops forwhich GM traits have been developedelsewhere. Wide-scale adoption of GMtechnology in China could well provide theimpetus for other developing countries to followsuit.

While the adoption rates for GM technologiesin developing countries are likely to rise, theyare expected to slow in the developed world.This mainly reflects the impressive growth ofthe past, which limits the remaining potential.GM soybean, for instance, already accounts fortwo-thirds of the soybean area worldwide andfor an even larger share of the area in developedcountries. As the global area of such cropsexpands, other, more sophisticatedbiotechnology applications may gainimportance. Examples include GM-based

3 400

3 350

3 300

3 250

3 200

3 150

3 100

3 050

With Bt Without Bt

Yield(kg/ha)

70

60

50

40

30

20

10

0With Bt Without Bt

Pesticide use(kg/ha)

2.5

2.0

1.5

1.0

0.5

0

With Bt Without Bt

Production costs(US$/kg)

25

20

15

10

5

0

With Bt Without Bt

Farmers reportingpoisonings (%)

54

nutraceuticals or cosmetic applications. Asthese new applications are likely to produce abroader range of benefits than “merely”cheaper foods and feeds, consumers in thedeveloped countries may become more inclinedto accept them.

Towards sustainableagriculture

Given a conducive policy environment, the next threedecades should see the spread of farming methods thatreduce environmental damage while maintaining or evenincreasing production. In some cases these technologies willalso reduce the costs of production.

No-till/conservation agricultureThe negative impact that soil tillage can have onsoil biological processes and hence onproductivity has been increasingly recognized. Inresponse, no-till or conservation agriculture(NT/CA) has been developed. This form ofagriculture can maintain and improve cropyields, providing greater resilience againstdrought and other stresses.

Like organic farming, NT/CA sustainsbiodiversity and saves on the use of resources.However, unlike organic farming, it can becombined with synthetic inputs and GM crops.It involves three principal elements:• Minimal disturbance of the soil. No tillage

occurs and crops are planted directly throughthe soil cover. Besides reducing the loss ofnutrients to the atmosphere, this sustains soilstructure and ecology.

• Maintenance of a permanent cover of live ordead plant material. This protects the soilagainst erosion and compaction by rain andinhibits the growth of weeds.

• Crop rotation. Different crops are planted overseveral seasons so as to avoid the buildup of

pests and diseases and to optimize the use ofnutrients.

NT/CA can raise crop yields by 20 to 50percent. Yields are less variable from year toyear, while labour and fuel costs are lower. Oncedemonstrated to farmers at a given location,NT/CA tends to spread spontaneously over alarger area. The main obstacles to its spreadare the complexity of managing crop rotation,the transitional costs of switching to newpractices and, to a certain extent, theconservatism of agricultural extension services.Retraining, sometimes combined with increasedfinancial incentives, may be needed to speed thepace of adoption.

Integrated pest managementPesticides involve a range of hazards in theirproduction, distribution and application. Whenused conventionally, they can eliminate naturalpredators as well as target pests, and generateresistance in pests. They may also pollute waterand soil resources and cause a range of healthproblems to operators and their families.

Integrated pest management (IPM) aims tominimize the amount of pesticides applied byusing other control methods more effectively.Pest incidence is monitored, and action is takenonly when damage exceeds tolerable limits. Theother technologies and methods used includepest-resistant varieties, bio-insecticides andtraps, and the management of crop rotations,fertilizer use and irrigation in such a way as tominimize pests. Chemical pesticides, if they areused at all, are chosen for minimum toxicity andapplied in carefully calculated ways.

Many countries have successfully introducedIPM and have experienced increased productionaccompanied by lower financial, environmentaland human health costs as a result. Again,extension systems and policy frameworks inmany countries have tended to favour the use ofpesticides. These must be reformed if IPM is tospread faster in future.

Integrated plant nutrient systemsAll crop production uses up plant nutrients inthe soil. Conventional fertilizers usually replaceonly a few key nutrients, while others continue

55

No-till/conservation agriculture canraise crop yields by 20 to 50 percent.Yields are more stable, resilienceagainst drought improves and labourand fuel costs are lower, butmanagement is more complex.

to be depleted. Many resource-poor farmerscannot afford these fertilizers, resulting in soilmining. In other cases there is overuse, leadingto the pollution of soils and water resources.

An integrated plant nutrient system is one inwhich practitioners aim to optimize the use ofnutrients through a range of practices thatinclude the recycling of vegetable and animalwastes and the use of legumes to fixatmospheric nitrogen. External nutrients areused judiciously, in ways that minimize costsand reduce pollution. Managing the use offertilizers precisely can increase their efficiencyby 10 to 30 percent.

The promise of organic agricultureOrganic agriculture is a set of practices inwhich the use of external inputs is minimized.Synthetic pesticides, chemical fertilizers,synthetic preservatives, pharmaceuticals, GMorganisms, sewage sludge and irradiation areall excluded.

Interest in organic agriculture has beenboosted by public concerns over pollution, foodsafety and human and animal health, as wellas by the value set on nature and the country-side. Consumers in developed countries haveshown themselves willing to pay price premiumsof 10 to 40 percent for organic produce, whilegovernment subsidies have helped to makeorganic agriculture economically viable.

As a result, organic agriculture has expandedrapidly in Western countries. Between 1995 and2000, the total area of organic land in Europeand the United States tripled, albeit from a verylow base.

In 2001, some 15.8 million ha were undercertified organic agriculture globally. Almost

half of this was in Oceania, just under a quarterin Europe and a fifth in Latin America. Abouttwo-thirds of the area is organic grassland. Asa percentage of total agricultural land, organicagriculture is still modest — an average of2 percent in Europe. However, many Europeancountries have ambitious targets for expansion,with the result that Western Europe may havearound a quarter of its total agricultural landunder organic management by 2030.

With a number of large supermarket chainsnow involved, the market for organic foods isbooming and potential demand far outstripssupply. In many industrial countries, sales aregrowing at 15 to 30 percent a year. The totalmarket in 2000 was estimated at almostUS$20 billion — still less than 2 percent of totalretail food sales in industrial countries but asizeable increase over the value a decade ago.

Land area under organic management

Source: Willer and Yussefi (2002)

Are

a (t

hous

and

ha)

Canad

aIta

ly

United Stat

es

German

y

United K

ingdom

Spain

France

Austria

1 000

800

600

400

200

0

56

Demand is expected to continue to grow,perhaps even faster than the 20 percent or soachieved in recent years. The supply shortfalloffers opportunities for developing countriesto fill the gap, especially with out-of-seasonproduce.

In industrial countries, organic agricultureis based on clearly defined methods enforcedby inspection and certification bodies. Mostdeveloping countries, in contrast, do not yethave their own organic standards andcertification systems. In these countries,organic agriculture may in fact be more wide-spread than in the developed world but ispractised of necessity, since the majority offarmers are unable to afford or cannot obtainmodern inputs. Most organic crops for localconsumption are sold at the same price asother produce. However, many developingcountries are now producing organiccommodities in commercial quantities forexport to developed country markets. Theseexports can be expected to increase in thecoming years.

Organic agriculture offers manyenvironmental benefits. Agrochemicals canpollute groundwater, disrupt key ecologicalprocesses such as pollination, harm beneficialmicro-organisms and cause health hazards tofarm workers. Modern monoculture usingsynthetic inputs often harms biodiversity at thegenetic, species and ecosystem levels. Theexternal costs of conventional agriculture canbe substantial.

In contrast, organic agriculture sets out toenhance biodiversity and restore the naturalecological balance. It encourages both spatialand temporal biodiversity through intercroppingand crop rotations, conserves soil and waterresources and builds soil organic matter andbiological processes. Pests and diseases arekept at bay by crop associations, symbioticcombinations and other non-chemical methods.Water pollution is reduced or eliminated.

Although yields are often 10 to 30 percentlower than in conventional farming, organicagriculture can give excellent profits. Inindustrial countries, consumer premiums,government subsidies and agritourism boost

incomes from organic farms. In developingcountries, well-designed organic systems cangive better yields, profits and returns on labourthan traditional systems. In Madagascar,hundreds of farmers have found they canincrease their rice yields fourfold, to as much as8 tonnes per ha, by using improved organicmanagement practices. In the Philippines,organic rice yields of over 6 tonnes per ha havebeen recorded. Experiences of organicproduction in low-potential areas such asNorthern Potosí (Bolivia), Wardha (India) andKitale (Kenya) have shown that yields can bedoubled or tripled over those obtained usingtraditional practices.

Organic agriculture also has social benefits.It uses cheap, locally available materials andusually requires more labour, thereby increasingemployment opportunities. This is aconsiderable advantage in areas where, or attimes when, there is a labour surplus. Byrehabilitating traditional practices and foods,organic agriculture can promote social cohesion.

Certain policy measures are essential if theprogress of organic agriculture is to continue.Support for agriculture is increasingly shiftingfrom production goals to environmental andsocial goals, a trend that could favour organicagriculture. Agreed international standards andaccreditation are needed to remove obstacles totrade. Extensionists often promote the idea thatsynthetic inputs are best and may need trainingin organic methods. Research to solve technicalproblems needs to be stepped up. Secure landtenure is essential if farmers are to undertakethe long process of conversion to organicstandards. If these measures are put in place,organic agriculture could become a realisticalternative to traditional agriculture over thenext 30 years, at least at the local level.

Locally, organic agriculture couldbecome a realistic alternative totraditional agriculture over the next30 years.

57

Directions for research

Strengths and weaknesses of past researchThe green revolution has played a key role inthe major improvements in food supply overthe past 40 years. The yields of rice, wheat andmaize in developing countries have risen by100 to 200 percent since the late 1960s.

Yield gains were the primary focus of thegreen revolution. Breeding and selection led tothe development of improved crop varieties, butgreatly increased use of inputs, such asfertilizer, pesticides and irrigation water, wereneeded to get the best out of these varieties.The green revolution achieved its aims not justthrough research but through a package ofmethods and inputs pushed by national andinternational agencies, extension services andprivate-sector companies.

But this first green revolution had itsshortcomings:• It was heavily geared to the world’s three

leading cereal crops, which were suited to itsemphasis on maximizing yields. Other crops,including many that are important in sub-Saharan African, such as cassava, millet,sorghum, banana, groundnut and sweetpotato, needed a different approach.

• It was suited only to areas with good soils andwater supplies, and largely neglected themore marginal rainfed areas with problemsoils and uncertain rainfall.

• It relied on farmers being able to affordinputs, and did little for poor smallholderswith insufficient funds or access to credit.

• Finally, it largely ignored the possibleenvironmental consequences of high inputuse, such as the pollution of water and soilswith nitrates and pesticides.

Needed: a doubly green revolutionA second, doubly green revolution is now needed.Its goals, as with the first, must include increasedproductivity. But it must also aim for sustainability— minimizing or reducing the environmentalimpacts of agriculture — and for equity — makingsure that the benefits of research spread to thepoor and to marginal areas.

Productivity must increase on all the landswhere farmers seek a living, not just in the well-endowed areas. More varieties and packages forcrops other than the three key cereals need tobe developed. And the potential of resource-conserving approaches such as IPM needs to befully realized.

Research for the new green revolution needsto be genuinely multidisciplinary. It must covernot only the biological sciences, includinggenetic engineering alongside conventionalbreeding and agronomy, but also the socio-economic context in which farming occurs. Andit must focus not only on crops and animals buton the ecology of all life forms within the farm-ing system. Areas of special importance inecology include the interactions of plants, pestsand predators, and competition between cropsand weeds. Plant rooting systems and theavailability of nutrients and soil organic matteralso deserve more emphasis.

Above all, priority must be given to theneeds of the poor in the marginal, rainfedareas bypassed by the first green revolution.Scientists must engage in an interactivedialogue with all the stakeholders in the re-search process, especially farmers but alsopolicy makers, civic society and the generalpublic.

Research towards this second green revolu-tion is already under way in some locations. Itsfirst fruits have shown that it can be successful,

• Will the technology lead to higher productiv-ity across all farms, soil types and regions,not just well-endowed ones?

• How will the technology affect the seasonaland annual stability of production?

• How will the technology affect the eco-system and the sustainability of farming?

• Who will be the winners and losers from thetechnology — and how will it affect thepoor?

Key questions for researchers:

58

especially when farmers participate actively inthe design and testing of new technology.However, the research effort needs to be greatly

strengthened and the challenge of scaling upthe results of research has yet to be adequatelyaddressed.

Livestock: intensification and its risks

Livestock production currently accounts forsome 40 percent of the gross value of worldagricultural production, and its share is rising.It is the world’s largest user of agricultural land,directly as pasture and indirectly through theproduction of fodder crops and other feedstuffs.In 1999 some 3 460 million ha were underpermanent pasture — more than twice the areaunder arable and permanent crops.

Livestock provide not only meat, but dairyproducts, eggs, wool, hides and other goods.They can be closely integrated into mixedfarming systems as consumers of crop by-products and sources of organic fertilizer, whilelarger animals also provide power for ploughingand transport.

Livestock have a considerable impact onthe environment. Growth of the livestock sectorhas been a major factor contributing todeforestation in some countries, particularly inLatin America. Overstocking land with grazinganimals can cause soil erosion, desertificationand the loss of plant biodiversity. Public healthhazards are increasing with the intensificationof urban and peri-urban livestock production.Wastes from industrial livestock facilities can

pollute water supplies and livestock are majorsources of greenhouse gases.

Diets shift from staples to meatThe past three decades have seen major shiftsin human diets. The share of animal productshas risen, while that of cereals and otherstaples has fallen. And within the meat sectorthere has been a dramatic rise in the shareof poultry and, to a smaller extent, pig meat.These trends are likely to continue over the next30 years, though in less dramatic form.

As incomes rise, people generally prefer tospend a higher share of their food budget onanimal protein, so meat and dairy consumptiontends to grow faster than that of food crops. Asa result, the past three decades have seenbuoyant growth in the consumption of livestockproducts, especially in newly industrializingcountries.

Annual meat consumption per person indeveloping countries as a whole more thandoubled between 1964-66 and 1997-99, fromonly 10.2 kg per year to 25.5 kg — a rise of2.8 percent a year. The growth was much less(from 10 kg to 15.5 kg) if China and Brazil areexcluded. The rise was particularly rapid for

Meat and dairy products will providean increasing share of the humandiet, with poultry expanding fastest.Future demand can be met, but thenegative environmental consequencesof increased production must beaddressed.

Livestock are the world’s largest userof agricultural land: in 1999 some3 460 million ha were under permanentpasture — more than twice the areaunder arable and permanent crops.

59

poultry, where consumption per person grewmore than fivefold. Pig meat consumption alsorose strongly, though most of this rise wasconcentrated in China.

The overall rise was unevenly spread: inChina meat consumption has quadrupled overthe past two decades, whereas in sub-SaharanAfrica it has remained stagnant, at under 10 kgper person. Differences in meat consumptionbetween countries can be substantial becauseof differences in meat availability or in dietaryhabits, including the role of fish in the provisionof total animal protein. For example, meatconsumption in Mongolia is as high as 79 kg perperson, but overall diets are grossly insufficientand undernourishment is widespread. Meatconsumption in the United States and Japan,two countries of comparable living standards, is120 kg and 42 kg per person respectively, buttheir per capita consumption of fish and seafoodis 20 kg and 66 kg.

Future growth may slowLooking towards 2030, the trend towardsincreased consumption of livestock productswill continue in the developing countries.However, future growth in consumption of bothmeat and milk may not be as rapid as in therecent past, given the reduced scope for furtherincreases in major consuming countries.

In developed countries the scope for in-creased demand is limited. Population growth isslow and the consumption of livestock productsis already very high. At the same time healthand food safety concerns, focused on animalfats and the emergence of new diseases such asbovine spongiform encephalopathy (BSE) andvariant Creutzfeldt-Jakob disease (vCJD), are

Annual meat consumption per personin developing countries more thandoubled between 1964-66 and1997-99, but there were substantialdifferences between countries.

holding back demand for meat. Total meatconsumption in the industrial countries hasrisen by only 1.3 percent a year over the past tenyears.

In developing countries the demand for meathas grown rapidly over the past 20 years, at5.6 percent a year. Over the next two decadesthis rate is projected to slow by half. Part of thisslowdown will be due to slower populationgrowth and part to the same factor that is atwork in developed countries: the countries thathave dominated past increases, such as Chinaand Brazil, have now reached fairly high levelsof consumption and so have less scope forfurther rises. In India, which will rival China asthe most populous country in the world in the2040s, the growth of meat consumption may belimited by cultural factors in addition to thecontinued prevalence of low incomes, sincemany of India’s people are likely to remainvegetarians. However, India’s consumption ofdairy products is projected to continue to riserapidly, building on the successes achieved overthe past 30 years. In sub-Saharan Africa, sloweconomic growth will limit increases in bothmeat and dairy consumption.

The rise in poultry consumption looks set tocontinue, though a little more slowly than in the

Source: FAO data and projections

World average meat consumption perperson, 1964-66 to 2030

1964-66 1997-99 2030

50

40

30

20

10

0

Con

sum

ptio

n (k

g/ca

pita

/yea

r)

PoultryPig meatSheep and

goat meatBeef

60

past, from a global average of 10.2 kg perperson in 1997-99 to 17.2 kg by 2030. Muchsmaller increases in world per capitaconsumption are foreseen for both pig meatand beef.

Bigger herds, fatter animalsGiven the slower growth of demand, livestockproduction will also grow more slowly than inthe past. Moreover, increased efficiency in thesector could mean that extra demand can bemet by a smaller growth in the number ofanimals. In absolute terms, however, thenumber of animals will still need to rise con-siderably. The projections show an extra360 million cattle and buffaloes, 560 millionextra sheep and goats, and 190 million extrapigs by 2030 — rises of 24, 32 and 22 percentrespectively.

However, it should prove possible to meetmuch of the extra demand by increasingproductivity rather than animal numbers. Thereis ample scope for this in developing countries,particularly with regard to cattle productivity. In1997-99 the yield of beef per animal in develop-ing countries was 163 kg compared with 284 kgin industrialized countries, while average milkyields were 1.1 and 5.9 tonnes per year per cowrespectively.

Selection and breeding, together withimproved feeding regimes, could lead to fasterfattening and larger animals. The averagecarcass weight for cattle, for example, hasalready risen from 174 kg in 1967-69 to 198 kg30 years later; by 2030 it could reach 211 kg.The offtake rate should also rise, as animalswill be ready for market earlier.

The shift to more intensive production willcontinueA continued shift in production methods can beexpected, away from extensive grazing systemsand towards more intensive and industrialmethods.

Grazing on pasture still provides 30 percentof total beef production, but its market share isdeclining. In South and Central America,grazing is often pursued on land cleared fromrainforests, where it fuels soil degradation and

further deforestation. In semi-arid environ-ments, overstocking during dry periodsfrequently brings risks of desertification,although it has been shown that pastures dorecover quickly if stock are taken off and goodrains return.

Mixed farming, in which livestock providemanure and draught power in addition to milkand meat, still predominates for cattle. Aspopulations and economies grow, these multi-purpose types of farming will tend to give way tomore specialized enterprises.

Where land is scarce, more intensivesystems of stall-feeding emerge. In thesesystems, fodder is cut and brought to thestabled animals, leading to less soil damageand faster fattening. This trend too can beexpected to continue and accelerate.

More industrial and commercial forms ofproduction will gradually increase in bothnumber and scale. These intensive enterpriseswill make use of improved genetic material,sophisticated feeding systems, animal healthprophylactics and highly skilled management.In recent years, industrial livestock productionhas grown at twice the rate of more traditionalmixed farming systems and at more than sixtimes the rate of production based on grazing.At the turn of the century industrial enterprisesaccounted for 74 percent of the world’s totalpoultry production, 68 percent of its eggs and40 percent of its pig meat.

Current trends towards industrial andcommercial production could pose a threat tothe estimated 675 million rural poor whoselivelihoods depend on livestock. Without special

In recent years, livestock productionfrom industrial enterprises has growntwice as fast as that from moretraditional mixed farming systems andmore than six times faster than fromgrazing systems.

61

measures, the poor will find it harder tocompete and may become marginalized,descending into still deeper poverty. Yet, if thepolicy environment is right, the future growth indemand for livestock products could provide anopportunity for poor families to generateadditional income and employment. Because ofits low capital costs, and its ability to make useof wastes and communally owned resources,livestock production allows poor families toaccumulate assets and diversify risks, besidesserving as a valuable source of products thatimprove both cash income and family nutrition.Policy measures that will help the poor enterand stay in the expanding market for livestockproducts include the provision of low-costcredit, technical support — especially in animalhealth and quality matters — and better accessto markets through improved infrastructure andinstitutions.

The growing demand for livestockproducts offers an opportunity for the675 million rural poor who depend onlivestock to improve their livelihoods.

Environment and health problemsCommercial and industrial systems bring theirown environmental problems, which differ fromthose of extensive systems. The concentrationof animals, particularly in urban areas, leads toproblems of waste disposal and pollution.Higher animal densities and transport to moredistant markets often involve the frustration ofnatural animal behaviour, bringing distress.Increased trade in livestock products andfeedstuffs brings greater risk of diseasetransmission, both within and across nationalboundaries. This applies both to diseaseslimited to livestock, such as foot-and-mouth,and to those that may affect both livestock andhumans, such as avian flu.

Infectious animal diseases such asrinderpest or foot-and-mouth are still major

threats in developing countries. Increased tradecan spread them more widely, even to developedcountries. Eradication programmes are shiftingaway from countrywide control strategiestowards more focused and flexible approaches,with the aim of improving the cost-effectivenessof control.

In humid and subhumid Africa,trypanosomiasis (sleeping sickness) posesan enormous obstacle to human health andcattle production. Trypanocidal drugs, aerialspraying, adhesive insecticides, impregnatedscreens and traps and the use of sterile in-sects offer the promise of recovering infestedareas for mixed farming. This will improvehuman health and nutrition, as well as live-stock and crop production.

Industrial livestock enterprises useantibiotics on a large scale. This practice hascontributed to antibiotic resistance amongbacteria, including those that cause humandiseases. Resistance to antihelminthics isemerging among livestock parasites. Industrialenterprises also use growth hormones to speedfattening and increase the efficiency of con-version of feed into meat. Public concern hasled to restrictions on use in the EU, althoughnegative impacts on human health have notbeen proved.

Increased trade in livestock productsand feedstuffs brings greater risk ofdisease transmission, both within andacross national boundaries.

Promises and risks of biotechnologyBiotechnology will have a profound effect onthe future of livestock production. Somebiotechnology applications are already in use,while others are still under research.

Artificial insemination, already routine indeveloped countries, will spread in developingcountries. It can greatly increase the efficiencyof animal breeding.

62

The cloning of mammalian cells could alsoboost productivity and output, particularly fordairy cattle in developed countries. However,the problems with this technology must besolved: currently only 2 to 5 percent of attemptsto clone animals actually succeed, and clonedanimals often develop serious health problems.

Rapid advances in understanding the geneticmake-up of animals will provide additionalpotential for productivity growth. Genes that areimportant for economic performance, such asthose for disease resistance or for adaptation toadverse environmental conditions, can beidentified and transferred into more productivebackgrounds, either through marker-assistedselection or through GM. These applicationscould prove especially useful in developingcountries.

GM animals have so far been used mainly forbiomedical research or the production of human

proteins. GM cattle, sheep, pigs and chickensare now being produced experimentally, withthe intention of eventual use for human con-sumption. Despite signs of consumer resistanceto GM foods for direct human consumption,products from livestock fed with GM maize,soybean and cottonseed cake are already on themarket.

The main risks of genetic modification arisefrom potential side-effects on the environmentor on human health. These risks are particularlypronounced if there is insufficient testing beforewidespread release. There is also the risk ofnarrowing the genetic base and concentratingits control in the hands of large multinationalcorporations. Almost 5 000 breeds and strainsof farm animals have been identified. Some 600of these face extinction and many more may beat risk if the genetic resource base is notconserved.

Globally, some 660 million tonnes of cerealsare used as livestock feed each year. Thisrepresents just over a third of total worldcereal use.

This use of cereals is often perceived as athreat to food security, since it appears toremove from the market supplies of essentialfoods that would otherwise be available topoor countries and families, thereby raisingfood prices. However, it is important to realizethat if these cereals were not used as feed,they would probably not be produced at all, sowould not be available as food in any case.

The use of cereals as feed may actually helpfood security. The commercial livestock sectoris responsive to the price of cereals: whenevershortages raise prices, livestock producerstend to switch to other feeds, releasing morecereals for food use. As a result, the food useof cereals may contract less than it would have

Cereals used as feed: threat or safety valve?

done otherwise. In short, the use of cereals asfeed serves as a buffer, protecting food intakesfrom supply variations.

In recent years the use of cereals as feedshas declined in relative terms. One reason isthe growing use of cereal substitutes inlivestock feed rations. Another is the collapseof the livestock sector in the transitioncountries, which led to reduced demand forfeed in these countries. A third factor is theshift of meat production to poultry, which aremuch more efficient converters of feed thanother livestock species.

Over the next three decades growth in theuse of cereals as feeds is projected to behigher than in the recent past, accounting forhalf the additional use of cereals. This is partlybecause the transition countries will resumetheir agricultural growth and partly becausethe shift into poultry is expected to be slower.

63

India’s white revolution

Launched in 1970, India’s Operation Floodhas had an impact comparable to the greenrevolution on rural incomes and food prices. Ithas turned India’s dairy sector around.

Milk consumption per person had beenfalling, from 39 kg in 1961 to only 32 kg in 1970.Since then it has risen rapidly, reaching 65 kgper person by 1999. Prices of milk to consum-ers have fallen, while the incomes of Indiandairy farmers have quadrupled.

Operation Flood was created and led bynational institutions and supported by the WorldBank and the EU. It began with the selling offood aid, the profits from which were used tostrengthen dairy cooperatives and smallholdermanagement. Local cows were crossed withspecialized dairy breeds to produce a robust yetproductive animal adapted to local conditions.Artificial insemination, veterinary services andother inputs were provided, leading to improvedmilk yields, longer lactation periods and shortercalving intervals. Operation Flood also focusedon improving smallholders’ access to markets,

opening new marketing channels for remoterural producers and thereby reducing both theneed for middlemen and the seasonal varia-tions in milk prices that had previouslydiscouraged producers. Milk collection andchilling centres were established, minimizingwaste due to spoilage.

Operation Flood has greatly helped India’srural poor. Three-fifths of the operation’s9 million producers are marginal or small-scale farmers or landless people. The impacton women has been particularly marked. Sixthousand village-level Women’s Dairy Coop-erative Societies have been formed. As womenhave shifted into dairy production, they havefreed up employment opportunities, especiallyon construction sites where they traditionallyworked as unskilled labourers. Money earnedfrom the dairy industry has been used to keepchildren in school. Older female siblings,relieved of the need to stay at home to care foryounger children, now have the option ofcontinuing their education.

Milk consumption in India, 1961 to 1999

Source: FAO data

1961 1965 1970 1975 1980 1985 1990 1995 1999

70

60

50

40

30

0

Con

sum

ptio

n (k

g/ca

pita

/yea

r)

Operation Flood begins

64

Forests and other wooded areas perform keyeconomic and ecological functions. Not only dothey provide goods and livelihoods but they alsoprotect soils, regulate water flow and retaincarbon that might otherwise add to greenhousegases. Forests also shelter much of the world’sterrestrial biodiversity.

In 2000 the world had some 3 870 million haof forests, covering 30 percent of its land area.Tropical and subtropical forests comprised56 percent of the forest area, while temperateand boreal forests accounted for the rest.Natural forests were estimated to constituteabout 95 percent of global forests, whileplantation forests constitute around 5 percent.

Altogether, 51 percent of global forests areavailable for wood supply. Some 12 percent offorests are in legally protected areas, while theremaining 37 percent are physically inaccessibleor otherwise uneconomic for wood supply.

More than half the wood biomass consumedglobally is burned as fuel. Most fuelconsumption occurs in developing countries,where wood is often the primary source of

Towards sustainable forestry

energy. Asia and Africa together consume morethan three-quarters of global fuelwood, mostlyin domestic cooking, though cottage industriessuch as food drying and brick-making alsoconsume large volumes in some countries.

Industrial roundwood currently comprisesabout 45 percent of global wood production.Interestingly, annual per capita woodconsumption in developed and developingcountries is about equal, at just over 0.5 m3

per person. However, almost 80 percent of woodconsumption in developed countries is in theform of industrial wood products, while indeveloping countries well over 80 percent isburned as fuel.

World trade in wood does not lend itselfto easy generalizations. Production and tradingpatterns are highly diverse, both regionally andamong different commodities. In 2000,temperate and boreal zones accounted for80 percent of world industrial roundwoodproduction and 83 percent of roundwoodexports. However, these zones also accountedfor 85 percent of wood product consumption.Also in 2000, tropical areas were net exportersof wood products to the tune of around59 million m3 per year, though this was lessthan 4 percent of global consumption.

From deforestation to reforestationIt is often suggested that the world faces adeforestation crisis. Certainly, in somecountries the picture is alarming and a rapiddecline in forest area continues. During the1990s, the total forest area shrank by a net9.4 million ha each year, an area about threetimes the size of Belgium. Over the decade as awhole, the area lost was bigger than Nigeria.

It is true that if current deforestation ratesare projected into the future, then by 2030natural tropical forests will shrink by a further24 percent. However, deforestation was slowerin the 1990s than in the 1980s and will probablycontinue to slow during the first decades of thenew century.

Globally, deforestation is slowingdown. At the same time, theproductivity of timber processingis improving, helping to meet therising demand for wood. However,hotspots of deforestation are likelyto persist, undermining biodiversityand the provision of other economicand environmental benefits fromforests. The major challenge willbe to improve the sustainablemanagement of forests and toensure equitable distribution ofthe benefits of forest use.

65

The picture varies considerably fromregion to region. Deforestation was most rapidin the tropics, where 1990s losses averaged12.3 million ha per year. Africa lost 5.3 millionha a year and South America 3.7 million ha.In contrast, annual losses in Asia were only0.4 million ha, while non-tropical areas actuallyadded 2.9 million ha a year to their forests.

Net deforestation is now slowing in manydeveloping countries. For more than a decade,countries such as China, India, Libyan ArabJamahiriya, Turkey and Uruguay have plantedmore forest than they have cut. By 2000, othercountries, such as Algeria, Bangladesh, Gambiaand Viet Nam, had also begun accumulatingnet forest area. Some countries, for exampleThailand and the Philippines, have imposedcomplete bans on natural forest harvesting,although these may not last and are difficultto implement. In many developing countries,population growth and dependence on agri-culture will lead to continuing loss of forests.However, overall rates of deforestation will slowfurther in the coming decades. Social, economicand political trends will contribute to the

During the 1990s, the area of tropicalforests shrank by a net 12.3 million haeach year, but non-tropical areasactually added 2.9 million ha a yearto their forests.

Source: FAO (2001)

slowdown in deforestation in developingcountries. Urbanization will reduce the need toopen up new frontier land to create livelihoods.It will also drive a shift from wood to fossil fuelsand electricity.

This slowing is an integral part of the cycleof economic development. In the initial stages ofdevelopment, rapidly growing populations arestill heavily reliant on agriculture and fuelwoodand some countries may depend on timberexports to generate foreign currency, with theresult that deforestation may be rampant. Ascountries grow richer and more urbanized, theneed to clear forests declines and the value

Forest area as a percentage of country land area

66

placed on natural environments rises. More andmore forests are protected or managedsustainably.

In developed countries, populations aregrowing only slowly and forest areas are mostlyincreasing as marginal farmland is set asideand regenerates as secondary natural forest.

Wood products: growing demand, growingproductivityDemand for forest products will continue togrow as world population and incomes grow.The most recent FAO projections estimate thatby 2030 global consumption of industrialroundwood will rise by 60 percent over currentlevels, to around 2 400 million m3. Substantialrises are also likely in the consumption of paperand paperboard products.

Will the world’s forest resources be able tocope? Until the early 1990s, expert assessmentswere pessimistic, but most experts today nolonger foresee a crisis in the supply of wood.Projections of wood consumption are lower now,partly because of lower world population growth.In addition, there have been improvements inforest management and in harvesting andprocessing technologies, increases in plantation

establishment, and an expansion of the role oftrees outside forests.

The production of wood-based materials iscontinually increasing in efficiency, reducing thepressure on forest resources. Not only is theremore recycling of paper and wood. The pastdecade has also seen a shift from industrialroundwood and sawnwood to wood-basedpanels, which make much fuller use of timber.Global production of sawn timber has remainedlargely static since 1970, yet that of wood-basedpanels has more than doubled, while theproduction of paper and paperboard has almosttripled.

In the future, the key questions will not bewhether there will be enough wood but ratherwhere it should come from, who will produce it,and how it should be produced.

There has been a shift in the sources ofwood, away from poorly regulated wild foreststowards plantations and sustainably managedforests and woodlands. Industrial roundwoodproduction from plantations is expected todouble by 2030, from 400 million m3 today toaround 800 million m3. Thus increasedplantation supplies will meet much of thegrowth in demand for wood during this period.

Forest area changes (million ha), 1990 to 2000

Source: FAO (2001)

Natural forest

Other landuse classes

Forest plantations

Tropical areas Non-tropical areas

Natural forest

Other landuse classes

Forest plantations

Deforestation

1990: 1 9452000: 1 803

1990: 1 8632000: 1 879

1990: 6 2802000: 6 252

1990: 1072000: 119

1990: 482000: 68

142

10

8

10 5

7

4

26

Afforestation Reforestation Natural expansion of forest

1990: 2 8192000: 2 943

67

Another greatly expanded source of wood will betree cultivation outside forests.

Changes in the conditions of trade areunlikely to be dramatic, as most significant tariffbarriers have already been reduced to moderatelevels or completely removed — though eco-labelling and environmental regulations willdoubtless increase. However, there will bemajor shifts in the directions of internationaltrade, as developing countries increase theirper capita consumption of industrial wood. Insome of the richer countries, per capita con-sumption is currently at least tenfold that ofmany developing countries.

Stressing forestry’s environmental servicesIncreasing realization of the importance ofenvironmental values and services is helpingefforts to conserve forest and tree resources.As the wider environmental services of treesare recognized, the planting or conservation oftrees and forests is being fostered by develop-ment projects and programmes as a means ofpreventing erosion, regulating water flow andso avoiding downstream flooding, and controllingdesertification or salinization. The trend towardstree and forest planting and conservation islikely to continue.

A shift in attitudes has led to a rise in thevalue attached to environment and natureconservation by non-governmental anddevelopment organizations. There is increasingpressure to adhere to acceptable standards ofnatural resource management in all efforts tostimulate economic growth and promotelivelihoods for poor rural people. Theemergence of democratic institutions andimproved access to information are helping thisprocess.

Shifts in consumer values, especially in thericher developed countries, have led to anincrease in environmentally consciouspurchasing. The spread of eco-labelling nowallows consumers to choose products fromsustainably managed forests.

Ecotourism is another outcome of the sameshift. This is currently estimated to constitutearound 7 percent of global tourism, a share thatis expected to increase. Paradoxically, a high

volume of ecotourists can place great pressureon sites that offer memorable experiences.Nevertheless, ecotourism can prove a valuablesource of income for local communities andhence an economic incentive to conserveremaining forests.

Growing concern over global warming hasfocused attention on the potential role of forestsin regulating carbon dioxide levels in theatmosphere. Forests store large amounts ofcarbon in trees, understory vegetation, litter andsoil. Globally, they contain some 1 200 billiontonnes of carbon, just over half the total in allterrestrial vegetation and soils.

New forests, or degraded forests that areallowed to regenerate, absorb and store carbonas they grow. Conversely, when they are clearedor degraded, forests can become a substantialsource of carbon dioxide emissions. Accordingto the Intergovernmental Panel on ClimateChange (IPCC), measures such as reduceddeforestation, forest regeneration andplantation development could reduce carbondioxide emissions by the equivalent of 12 to15 percent of all emissions from fossil fuelsbetween 1995 and 2050. However, it is not yetclear to what extent this potential will bereflected in formal international agreements onclimate change.

Sustainable forest managementThe set of principles and practices known assustainable forest management (SFM) isincreasingly accepted as the core paradigm forforestry development. SFM implies broadeningthe focus of management from wood productionto include more emphasis on equitable and

Measures such as reduced deforestation, forest regenerationand plantation development could reduce carbon dioxideemissions by the equivalent of 12 to 15 percent of all theemissions from fossil fuels between 1995 and 2050.

68

participatory development and on environmentalconsiderations.

If forest development is inequitable, thepoor who are excluded from it will continue todepend on land and forest resources, but willexert greater pressure on the remaining areasto which they have access and may encroachillegally on protected areas or those allocatedto large-scale enterprises. Hence an importantaspect of SFM is its emphasis on providingsustainable livelihoods for the estimated350 million of the world’s poorest and mostmarginalized people who depend on forestecosystems.

Non-wood forest products (NWFPs), suchas wild foods, plants and medicinal herbs, arecrucial to this vulnerable group. The majorityare subsistence goods or are traded only in localmarkets. However, an estimated 150 NWFPsare traded internationally. While dependenceon many subsistence products may decline,increasing demand for ethnic foods and medicinesmay lead to the more systematic cultivation ofsome NWFPs. Access to knowledge and techno-logy will be critical if local communities are tobenefit from this trend.

Under the participatory developmentassociated with SFM, the primary responsibilityof forestry departments will shift frommanagement to policy development and

regulatory functions. Responsibility formanagement will pass largely to the privatesector, including farmers and localcommunities.

The environmental goals of SFM will includeincreasing the area of protected forest andreversing the loss of biomass, soil fertility andbiodiversity that occurs when forests aredegraded. Unsustainable forestry practices willbe discouraged and logging techniques thatreduce negative impacts on the forest as awhole will be encouraged. Improved security ofland and tree tenure will encourage treeplanting, both inside and outside forests.

There has been progress towards the wideradoption of SFM, though that progress has beenuneven. At one end of the spectrum, forestmanagement is carefully monitored againstagreed social and environmental criteria. At theother, substantial tracts of (mainly tropical)forests are still managed poorly or not at all,leaving them vulnerable to careless orunscrupulous degradation.

Advances in remote sensing and in dataprocessing and exchange will make it easier fornational and international bodies to monitorforest management practices. But if SFM is tosucceed, it will be crucial to strengthen thedeveloping world’s forestry institutions, whichare still severely under-resourced.

End use Typical products

Food products and additives Wild meat, edible nuts, fruits, honey, bamboo shoots, birds’ nests,oil seeds, mushrooms, palm sugar and starch, spices, culinaryherbs, food colorants, gums, caterpillars and insects, fungi

Ornamental plants Wild orchids, bulbs, cycads, palms, tree ferns, succulent plants,carnivorous plants

Animals and animal products Plumes, pelts, cage birds, butterflies, lac, cochineal dye, cocoons,beeswax, snake venom

Construction materials Bamboo, rattan, grass, palm, leaves, bark fibres

Organic chemicals Phytopharmaceuticals, aromatic chemicals and flavours, fra-grances, agrochemicals/insecticides, biodiesel, tans, colours, dyes

Souce: FAO data

Some non-wood forest products

69

World fisheries: a choice of futures

Increasingly, biodiversity is seen not just asa source of genetic material, medicines orother commercial products but as having valuein itself. It has been estimated that forests,especially tropical rainforests, harbour asmuch as half the world’s biodiversity.

Globally, more than 30 000 protected areashave been established. The goal of the WorldConservation Union (IUCN) is that 10 percentof each country’s land area should be undersome form of protection. At present, some80 countries have attained this level, butaround 100 countries still have less than5 percent.

The World Conservation Monitoring Centreestimates that only 6.4 percent of the areaof forest biomes is under some form ofprotection at present — and as little as3.6 percent in the case of temperate broad-leafforests. These shortfalls reflect the unevendistribution of forest ecosystems amongcountries, in addition to an overall failure tomeet the IUCN goal.

Almost 9 percent of tropical rainforestsare protected, but in many developing coun-tries this protection is only nominal. Theseforests continue to suffer serious encroach-ments, including logging, deliberate burning,poaching and other forms of clearing ordegradation.

The prospects for future expansion ofprotected areas are more modest than inthe recent past. In many countries whereconservation efforts fall below the IUCN goal,there are already intense pressures on theseareas and strong conflicts between economicand environmental objectives. During thenext 30 years the total land area under strictprotection will increase only moderately.Other means must be found to conservebiodiversity, including the on-farm productionand conservation of trees and the conservationof germplasm in genebanks. Larger areascould also be placed under SFM, which affordsconservation high priority as a managementobjective.

The role of forests in protecting biodiversity

Fisheries play an important role in the worldfood economy. Worldwide, more than 30 millionfishers and fish farmers and their families gaintheir livelihoods from fisheries. Most of themare poor artisanal fisher families in developingcountries.

Globally, fish provide about 16 percent of theanimal protein consumed by humans, and are avaluable source of minerals and essential fattyacids. Ocean and freshwater fish are also anincreasingly important recreational resource,both for active users such as anglers and forpassive users such as tourists, sports diversand nature-lovers.

The marine fisheries catch levelledoff during the 1990s. Aquaculture grewrapidly, allowing continued growth intotal fish production. With manymarine stocks now fully exploited oroverexploited, future fish supplies arelikely to be constrained by resourcelimits. Achieving effective governanceof world fisheries is crucial.

70

As marine capture fisheries level off,aquaculture expandsOver the past three decades world productionof fish has more than kept pace with humanpopulation growth, with the result that theamount of fish available per person hasincreased. The recent stagnation of capturefisheries has been balanced by the rapid build-up of aquaculture.

Total annual fish production almost doubledbetween 1970 and 1999, from 65 million tonnesto 125 million tonnes. This rise was the outcomeof two contrasting trends: growth in capturefisheries followed by a levelling off in the 1990s,and dramatic growth in aquaculture during the1990s.

Since the1950s, increases in marine capturelevels have been made possible by advances infishing technology and efficiency, includingsynthetic fibres for fishing gear, on-boardfreezing, electronic fish finding and improvednavigation. However, as more and more fisheryareas and fish stocks reached full utilization orwere overfished, the growth of marine catchesbegan to flatten out. During the 1990s, marinecatches fluctuated between 80 and 85 milliontonnes per year, despite the discovery of newstocks.

What made possible the continuing rise inoverall fish production was the rapid growth ofaquaculture, which expanded by 10 percent peryear during the 1990s. The share of aqua-culture in world fish production doubled overthe decade, reaching 26 percent in 1999.

So far, aquaculture has been heavily con-centrated in Asia, which provided 89 percent ofworld production in 1999. A growing diversity ofspecies is now cultured. Until the mid-twentiethcentury the range was limited to oysters,mussels, carps, trouts and shrimps. However,since the 1950s scientists have gradually solvedthe problem of artificial reproduction fordifferent carps, salmonids and other species.

The overall increase in fish production hasbeen paralleled by a steady growth in con-sumption. Fish now account for an average of30 percent of the animal protein consumed inAsia, approximately 20 percent in Africa andaround 10 percent in Latin America and theCaribbean. By 1999 global average intake offish, crustaceans and molluscs reached 16.3 kgper person, an increase of more than 70 percentover the 1961-63 level.

Fisheries are also a significant source oflivelihoods. In developed countries, employmentin fishing has declined due to improvements inproductivity and the collapse of some importantfisheries. In contrast, in developing countriesfisheries employment has continued to expand.Over 90 percent of the people fully employed inthe fisheries sector in the early 1990s were inthe developing or transition economies.

Nearly 40 percent of all fish productionis now internationally traded. As a result,fisheries are increasingly seen as a powerfulmeans of generating hard currency. Developingcountries’ gross earnings from fish exports havegrown rapidly, from US$5.2 billion in 1985 toUS$15.6 billion in 1999, a level that far exceedsearnings from commodities such as coffee,cocoa, banana or rubber.

Fish consumption may be constrained byresource limitsFish consumption per person is expected tocontinue to rise. If it were to be determinedsolely by income growth and dietary changes,

The continuing rise in overall fishproduction was made possible by thegrowth of aquaculture at 10 percentper year during the 1990s. Thecontribution of aquaculture to worldfish production doubled over thedecade, reaching 26 percent in 1999.

Catches in inland waters, however,continued to grow moderately, from 6.4 milliontonnes per year in 1990 to 8.2 million tonnes in1999 — though the true inland total may bemuch higher, as produce is often bartered, soldor consumed locally without being formallyrecorded.

71

average intake could reach as high as 22.5 kgper person by 2030. Combined with populationgrowth, this would imply a total annual demandfor fish of 186 million tonnes by 2030 — almostdouble the present level. However, since supplywill probably be limited by environmentalfactors, a more likely range for demand is 150to 160 million tonnes, or between 19 and 20 kgper person.

The regional picture will be very diverse.Health and diet quality concerns will boostconsumption in North America, Europe andOceania, but slow population growth will meanslow increase in overall demand.

In sub-Saharan Africa and the Near East andNorth Africa, fish consumption per person maystagnate or even decline, despite current lowlevels. In Africa, local wild stocks are almostfully exploited and, except in Egypt, aquaculturehas barely begun. Per capita demand in SouthAsia, Latin America and China may increaseonly gradually, while in the rest of East Asia itwill almost double, reaching 40 kg by 2030.

Fish consumption by region, 1961-63 and1997-99

Source: FAO data

Asian aquaculturists should be able to increaseproduction, and any remaining shortfall can bemet by imports.

There is a growing trend to market fish freshfor human consumption. This is because thecosts of delivering fresh fish to markets arefalling and consumers are willing to pay apremium for this product. Demand for fish mealand fish oil will continue to grow rapidly. Theseproducts are used for livestock and aquaculturefeeds, and at present account for about aquarter of world fish production. So far the rawmaterial for fish meal and oil has been suppliedby capture fisheries, and in all likelihood thiswill continue. However, the competition forsmall surface fish will become more intense,and the fish meal and oil industry will need toexploit other raw materials, such asmesopleagic fish and krill. Rising prices willalso drive a switch to substitute feeds. However,a satisfactory replacement for fish oil has notyet been found.

Aquaculture and marine ranching will continueto expandOver the next three decades, the world’sfisheries will meet demand by continuing thesame shift from fish capture to fish cultivationthat gained momentum in the 1990s.

The share of capture fisheries in worldproduction will continue to decline. The maxi-mum sustainable marine production has beenestimated at around 100 million tonnes a year.However, this is higher than the annual catchesof 80 to 85 million tonnes achieved during the1990s, and assumes that large quantities ofhitherto underexploited aquatic resources willbe used, including krill, mesopelagic fish andoceanic squids.

As in the 1990s, most of the shortfall willbe made up by aquaculture, which will probablycontinue to grow at rates of 5 to 7 percent ayear, at least until 2015.

Aquaculture species will be improved.Traditional breeding, chromosome manipulationand hybridization have already made significantcontributions. In future the use of new techno-logies, such as genetic modification, can beexpected. Already, a gene that codes for an anti-

30

25

20

15

10

5

0

1961-63

1997-99

Industrial

countri

es

South A

sia

Near Eas

t and

North A

frica

Sub-Sah

aran

Africa

Latin A

meric

a

and C

aribbean

Tran

sition

countri

es

East A

sia

kg/p

erso

n/ye

ar

72

freeze protein in the Arctic flounder has beentransferred to Atlantic salmon to increase itstolerance of cold waters. Currently, however,no commercial aquaculture producer ismarketing such transgenic species for humanconsumption. If this field is to progress, publicconcerns about GM organisms will need tobe addressed through risk assessments andthe development of policy guidelines forresponsible use.

Additional species will be domesticated foraquaculture. For halibut, cod and tuna, whichhave been fished in high volumes in capturefisheries, aquaculture production couldeventually be high. If commercially viabletechnology is developed soon, by 2015 thecultured production of cod could reach 1 to2 million tonnes per year.

Environmental concerns will probably shiftthe focus of aquaculture away from coastalzones into more intensive inland systems.Marine ranching will also expand, though itslong-term future will depend on solutions to theproblems of ownership surrounding releasedanimals. At present, only Japan is engaged insea ranching on a large scale.

Social and political pressure will also driveefforts to reduce the impact of capture fisheries,for example by making use of the unwantedcatch of non-target species and by using moreselective fishing gear and practices. Increasinguse of eco-labels will enable consumers tochoose sustainably harvested fish products, a

trend which will encourage environmentallysensitive approaches in the industry.

Towards sustainable fisheriesThe single most important influence on thefuture of wild capture fisheries is theirgovernance. Although in theory renewable, wildfishery resources are in practice finite forproduction purposes. If they are overexploited,production declines and may even collapse.

Resources must therefore be harvested atsustainable levels. In addition, access must beequitably shared among producers. As fishresources grow increasingly scarce, conflictsover access are becoming more frequent.

The principal policy challenge is to bring thecapacity of the global fishing fleet back to a levelat which fish stocks can be harvested sustainably.Past policies have promoted the buildup of excesscapacity and incited fishermen to increase thecatch beyond sustainable levels. Policy makersmust act fast to reverse this situation.

There are numerous measures that couldencourage sustainable use and remove theperverse incentives to overfish. Fisheries basedon clearly defined rights of access will need tobecome more common: experience shows thatwhen these rights are not merely in place butare understood and observed by users, conflictstend to be minimized.

Laws and institutions need to be establishedor strengthened to limit and control access tomarine fish stocks, both by larger ocean-going

Source: FAO data

State of the world’s fishery stocks, 1998The maximum sustainable marineproduction is estimated at around100 million tonnes a year, comparedto annual catches of 80 to 85 milliontonnes in the 1990s. But the estimateassumes that large quantities ofhitherto underexploited resourceswill be used, including krill andoceanic squids.

Moderately exploited21%

Underexploited4%

Depleted9%

Recovering1% Fully exploited

47%

Overexploited18%

73

Biodiversity comprises four main elements:variability within species, among species,among ecosystems and among larger ecologi-cal complexes. It is a key ingredient of sustain-able fisheries in the future.

Altogether, over 1 100 species of fish,mollusc and crustacean are taken in capturefisheries, while over 300 species are used inaquaculture. Biodiversity in wild populationsallows adaptation to the changing environ-ment, while in farmed fish it allows continuedbreed improvement.

Human fishing activities have had a power-ful impact on aquatic biodiversity. The currenthigh level of impact may limit capture fishingin future, unless the governance and manage-ment of ocean and freshwater resources aregreatly improved.

Damage is done by several unsustainablefishing practices. These include: the use ofpoison and dynamite near coral reefs; non-selective fishing gears that capture marinemammals, unwanted species or fish that aretoo small; and bottom trawling, which disturbsthe ecology of the ocean floor.

Perhaps the major ecological impact stemsfrom the sheer extent of human fishing. Manyfishing areas and stocks are fished up to orbeyond the sustainable limit, while fishingpressure appears to have altered the distribu-tion and size of some fish.

The overall impact on ocean ecology is onlysketchily known, but appears to be significant.Statistics on fish landings suggest that therehas been a reduction in the numbers of largerpredatory fish, shifting the balance of catchestowards fish that eat lower down the foodchain. As high-value species such as bottom-dwellers or large surface fish such as tuna areoverfished, they are gradually replaced by

The changing ecology of the oceans

shorter-lived and smaller surface-dwellingand schooling fish. Numbers of smaller fishare also boosted in some areas by increasedplankton production.

By 1998, some 12 out of the FAO’s 16 worldfishing regions had production levels at orbelow their historical maximum. Indeed, theAntarctic, the Southeast and NorthwestAtlantic and the Southeast Pacific had fallenbelow half their maximum past productionlevels.

In terms of stocks of main species, FAOestimates suggest that, by the end of the1990s, only a quarter of stocks were moder-ately exploited or underexploited and 1 percentwere recovering. Nearly half of all stocks wereexploited up to their maximum sustainableyield and were thus potentially on the brink ofbeing overexploited. More than a quarter ofstocks were overfished or depleted.

Such developments have raised concernamong environmentalists and otherstakeholder groups. In response, fisheryadministrations are working to minimize ormitigate the negative impacts on genetic andbiological diversity. Measures include thedevelopment and use of selective fishing gearsthat reduce the capture of marine mammals,undersized target species and unwantedbycatch, direct controls on the total catchallowed of various species and, in some cases,outright fishing bans and moratoriums.

Unfortunately, inappropriate fishing andaquaculture activities are only one of thethreats to aquatic biodiversity. Additionalthreats include pollution, loss of habitat andhabitat degradation. Such threats oftencombine to aggravate pressure on biodiversity.The whole range of threats must be addressedif aquatic biodiversity is to be protected.

74

vessels and by local artisanal fishers.Increasingly, the responsibility for managingfisheries will have to be devolved to fishinginterests and other stakeholder groups.Traditional arrangements in fishing communitiescan be incorporated into new managementregimes. However, the need to control entry intoartisanal fisheries will become more pressing.Indeed, if this issue is not tackled, a large

number of fisher households may be forced outof fishery and, unless there are alternativelivelihoods, into poverty.

If the world’s fisheries are to achieve theirfull potential, the major policy and manage-ment challenges must be met, and the culturaland social concerns of all stakeholder groupsmust be addressed. These are enormouschallenges, yet they are not insurmountable.

75

Prospects for the Environment

Agriculture and the environment

Previous sections of this report have alreadydiscussed the environmental aspects ofagriculture as they relate to each sector. Thissection discusses the overarching or cross-sectoral environmental issues and provides anoverview of the major trends in agriculture thatcan be expected to affect the environment overthe next 30 years.

methods of agriculture, forestry and fishingare the leading causes of loss of the world’sbiodiversity. The overall external costs of allthree sectors can be considerable.

Agriculture also affects the basis for its ownfuture through land degradation, salinization,the overextraction of water and the reduction ofgenetic diversity in crops and livestock. However,the long-term consequences of these processesare difficult to quantify.

If more sustainable production methods areused, the negative impacts of agriculture on theenvironment can be attenuated. Indeed, insome cases agriculture can play an importantrole in reversing them, for example by storingcarbon in soils, enhancing the infiltration ofwater and preserving rural landscapes andbiodiversity.

Agriculture has a vast impact on the earthAgriculture accounts for the major share ofhuman use of land. Pasture and crops alonetook up 37 percent of the earth’s land area in1999. Over two-thirds of human water use is foragriculture. In Asia the share is four-fifths.

Crop and livestock production have aprofound effect on the wider environment. Theyare the main source of water pollution bynitrates, phosphates and pesticides. They arealso the major anthropogenic source of thegreenhouse gases methane and nitrous oxide,and contribute on a massive scale to other typesof air and water pollution. The extent and

Over the next 30 years, manyagriculture-related environmentalproblems will remain serious.However, some problems may deepenmore slowly than in the past and somemay even be reversed.

Percentage of annual nitrogen emissionsfrom different sources

Source: Adapted from Mosier and Kroeze (1998)

Ocean20%

Industrial sources9%

Biomass burning3%

Cattle14% Agricultural soils

14%

Natural soils40%

Total emissions: 15 million tonnes per year

76

Fertilizers, manure and pesticides are majorcauses of water pollutionPollution of groundwater by agriculturalchemicals and wastes is a major issue in almostall developed countries and, increasingly, inmany developing countries.

Pollution from fertilizers occurs when theseare applied more heavily than crops can absorbor when they are washed or blown off the soilsurface before they can be incorporated. Excessnitrogen and phosphates can leach into ground-water or run off into waterways. This nutrientoverload causes eutrophication of lakes,reservoirs and ponds, leading to an explosion ofalgae which suppress other aquatic plants andanimals.

The crop projections to 2030 imply a slowergrowth of nitrogen fertilizer use than in the past.If efficiency can be improved, the increase intotal fertilizer use between 1997-99 and 2030could be as low as 37 percent. However, currentuse in many developing countries is veryinefficient. In China, the world’s largestconsumer of nitrogen fertilizer, up to half thenitrogen applied is lost by volatilization andanother 5 to 10 percent by leaching.

Insecticides, herbicides and fungicides arealso applied heavily in many developed anddeveloping countries, polluting fresh water withcarcinogens and other poisons that affecthumans and many forms of wildlife. Pesticidesalso reduce biodiversity by destroying weedsand insects and hence the food species of birdsand other animals.

Pesticide use has increased considerablyover the past 35 years, with recent growth ratesof 4 to 5.4 percent in some regions. The 1990sshowed signs of declining use of insecticides,both in developed countries such as France,Germany and the United Kingdom and in a fewdeveloping countries, such as India. In contrast,herbicide use continued to rise in mostcountries.

As concern about pollution and the loss ofbiodiversity grows, future use of pesticides maygrow more slowly than in the past. In developedcountries, use is increasingly restrained byregulations and taxes. In addition, use will becurbed by the growing demand for organic

crops, produced without chemical inputs. Thefuture is likely to see increasing use of “smart”pesticides, resistant crop varieties andecological methods of pest control (IPM).

Agriculture as a cause of air pollutionAgriculture is also a source of air pollution.It is the dominant anthropogenic sourceof ammonia. Livestock account for about40 percent of global emissions, mineralfertilizers for 16 percent and biomass burningand crop residues for about 18 percent.

Ammonia is even more acidifying thansulphur dioxide and nitrogen oxides. It is one ofthe major causes of acid rain, which damagestrees, acidifies soils, lakes and rivers, andharms biodiversity. As other acidifying gasessuch as sulphur dioxide are brought undertighter control, ammonia may in time becomethe leading cause of acidification. Emissionsof ammonia from agriculture are likely tocontinue rising in both developed and develop-ing countries. The livestock projections imply a60 percent increase in ammonia emissions fromanimal excreta.

The burning of plant biomass is anothermajor source of air pollutants, including carbondioxide, nitrous oxide and smoke particles. It isestimated that humans are responsible forabout 90 percent of biomass burning, mainlythrough the deliberate burning of forestvegetation in association with deforestationand of pastures and crop residues to promoteregrowth and destroy pest habitats. The massiveforest fires in Southeast Asia in 1997 burned atleast 4.5 million ha and covered the region witha pall of smoke and haze. The burning oftropical savannas is estimated to destroy three

Projections suggest that, by 2030,emissions of ammonia and methanefrom the livestock sector of developingcountries could be at least 60 percenthigher than at present.

77

times as much dry biomass each year as theburning of tropical forests.

Pressures on biodiversityAs their numbers and needs have grown, humanbeings have taken up an increasing share of theplanet’s surface area and resources for theirown needs, often displacing other species in theprocess. Estimates of the total number ofspecies on earth vary wildly. The number thathas been scientifically described is around1.75 million, but the true total is unknown andmay be anything from 7 to 20 million or more.Estimates of losses of biodiversity to extinctionover the coming decades vary widely, from 2 to25 percent of all species.

Loss of biodiversity owing to agri-cultural methods continues unabated,even in countries where nature ishighly valued and protected.

Agriculture, forestry and fisheries areperhaps the most significant of humanpressures on the biodiversity of land and sea.

Species richness is closely related to thearea of a wild habitat. As the area declines, sodoes the number of species it harbours, thoughat a slower rate. Deforestation, fieldconsolidation, with the accompanying reductionin field margins and hedgerows, and drainage ofwetlands for farming reduce the overall areaavailable for wildlife and fragment naturalhabitats. Grazing lowers species richness inpastures.

Agricultural intensification adds its ownproblems. Pesticides and herbicides directlydestroy many insects and unwanted plants, andreduce food supplies for higher animals. Henceloss of biodiversity is not limited to the landclearing stage of agricultural development butcontinues long afterwards. It is unabated even indeveloped countries where nature is highlyvalued and protected.

Agriculture is an increasinglysignificant source of greenhousegases, as well as a potential route tothe mitigation of climate changethrough carbon storage in soils andvegetation.

Some of the affected life forms may beimportant soil nutrient recyclers, croppollinators and predators of pests. Others arepotentially a major source of genetic materialfor improving domesticated crops and livestock.

The pressures on biodiversity over the nextthree decades will be the outcome of conflictingtrends. Extensive methods will tend to give wayto intensification, which may in turn give way toorganic agriculture or to NT/CA.

Loss of wildlife habitat to farming willcontinue, but at a slower pace. Deforestation willslow down, and extensive grazing will give wayincreasingly to industrial livestock production.Although intensification entails its own range ofenvironmental risks related to pesticides,chemical fertilizers and animal wastes, theincreasing inclusion of environmentalconsiderations in agricultural policy will help tocounteract these.

Reducing agriculture’s pollution tollThe spread of NT/CA will help to improve soilstructure and reduce erosion. IPM will reducepesticide use, while programmes to improve themanagement of plant nutrition should reducethe overuse of chemical fertilizers.

Other policies will help to reduce the conflictbetween agricultural intensification andenvironmental protection. Tighter regulationsand national strategies on the management ofanimal waste and the use of chemical fertilizersand pesticides may be required, along with theremoval of subsidies on chemical and fossilenergy inputs. Pesticides should be subjected tomore rigorous testing, and residue buildup moreclosely monitored.

78

Agriculture and climate changeAgriculture as source and sink Agriculture is a major source of greenhousegas emissions. It releases large quantities ofcarbon dioxide through the burning of biomass,mainly in areas of deforestation and grassland.

Agriculture is also responsible for up to halfof all methane emissions. Though it persists fora shorter time in the atmosphere, methane isabout 20 times more powerful than carbondioxide in its warming action and is therefore amajor short-term contributor to globalwarming. Current annual anthropogenicemissions are around 540 million tonnes andare growing at around 5 percent per year.

Livestock alone account for about a quarterof methane emissions, by way of gutfermentation and the decay of excreta. Aslivestock numbers grow, and as livestockrearing becomes increasingly industrial, theproduction of manure is projected to rise byabout 60 percent by 2030. Methane emissionsfrom livestock are likely to increase by the sameproportion.

Irrigated rice farming is the other mainagricultural source of methane, accounting forabout a fifth of total anthropogenic emissions.The area used for irrigated rice is projected toincrease by about 10 percent by 2030. However,emissions may grow more slowly, because anincreasing share of rice will be grown withbetter controlled irrigation and nutrientmanagement, and rice varieties may be usedwhich emit less methane.

Agriculture is a key source of anotherimportant greenhouse gas: nitrous oxide. This isgenerated by natural processes, but is boostedby leaching, volatilization and runoff of nitrogenfertilizers, and by the breakdown of cropresidues and animal wastes. Livestock accountfor about half of anthropogenic emissions.Annual nitrous oxide emissions from agricultureare projected to grow by 50 percent by 2030.

Agriculture can help mitigate climate changeFarming can also be a sink for carbon. However,it is generally believed that soils, like other

biological sinks (e.g. vegetation), have aninherent upper limit for storage. The totalamount that can be stored is crop- and location-specific and the rate of sequestration declinesafter a few years of growth before eventuallyreaching this limit. In 1997-99 an estimated590 to 1 180 million tonnes of carbon werelocked up in cropland soils alone, in the form ofsoil organic matter from crop residues andmanure. Projections of increased cropproduction imply that by 2030 this total couldrise by 50 percent.

Other changes could boost the total evenfurther. If only 2 million of the current 126million ha of saline soils were restored eachyear, they could account for an extra 13 milliontonnes of carbon annually. In developedcountries, permanent set-aside land cansequester large amounts of carbon if it is leftunmanaged, or reforested.

Depending on agroclimatic conditions,NT/CA can lock up 0.1 to 1 tonne of carbon perha per year, in addition to cutting carbon dioxideemissions by over 50 percent through thereduced use of fossil fuel in ploughing. Thegrowth potential for NT/CA is considerable. Ifanother 150 million ha of rainfed cropland isconverted to NT/CA by 2030 and the averagesequestration rate on land managed in thisway is 0.2 to 0.4 tonne per ha per year, a further30 to 60 million tonnes of carbon could besoaked up annually during the first few yearsafter conversion.

Should any of these practices be discontinued,the sequestered carbon would be released over aperiod of a few years. Agricultural carbon sinks ofthis kind are needed to “buy time” in which tocope with carbon dioxide emissions at source.

Climate change will have very diverse impactson agricultureClimate change will affect agriculture, forestryand fisheries in complex ways, positive as wellas negative.

Global carbon dioxide concentrations in theatmosphere are expected to rise from 350 ppm

79

In the next three decades, climatechange is not expected to depressglobal food availability, but it mayincrease the dependence ofdeveloping countries on food importsand accentuate food insecurity forvulnerable groups and countries.

to over 400 ppm by 2030. Carbon dioxide causesplant stomata to narrow, so water losses arereduced and the efficiency of water use im-proves. Increasing atmospheric concentrationsof carbon dioxide will also stimulate photo-synthesis and have a fertilizing effect on manycrops.

Average global temperatures are projectedto rise by between 1.4 and 5.8 OC by 2100. By2030 the increase will be rather lower than this,between 0.5 and 1 OC.

The rise will be greater in temperate latitudes.Here global warming may bring benefits foragriculture. The areas suitable for cropping willexpand, the length of the growing period willincrease, the costs of overwintering livestock willfall, crop yields will improve and forests may growfaster. These gains may, however, have to be setagainst the loss of some fertile land to flooding,particularly on coastal plains.

In less well-watered areas, especially in thetropics, the rise in temperatures will increaseevapotranspiration and lower soil moisturelevels. Some cultivated areas will becomeunsuitable for cropping and some tropicalgrasslands may become increasingly arid.

Temperature rise will also expand the rangeof many agricultural pests and increase theability of pest populations to survive the winterand attack spring crops. In oceans, temperaturerise may reduce plankton growth, bleach coralreefs and disrupt fish breeding and feedingpatterns. Cold-water species such as cod mayfind their range reduced.

Higher global temperatures will also bringhigher rainfall. However, this will be unevenly

distributed between regions. Indeed, sometropical areas such as South Asia and northernLatin America are projected to receive lessrainfall than before.

The climate is also expected to becomemore variable than at present, with increases inthe frequency and severity of extreme eventssuch as cyclones, floods, hailstorms anddroughts. These will bring greater fluctuationsin crop yields and local food supplies and higherrisks of landslides and erosion damage.

Mean sea level is projected to rise by 15 to20 cm by 2030 and by 50 cm by 2100. The risewill lead to the loss of low-lying land throughflooding, seawater intrusions and storm surges.Subsidence due to the overextraction ofgroundwater may exacerbate the intrusionproblem in some areas. There will also bedamage to vegetable growing and aquaculturein low-lying areas and to fisheries dependent onmangrove swamps for their spawning grounds.The impact will be most serious in coastalzones, especially heavily populated deltas usedfor agriculture, of the kind found in Bangladesh,China, Egypt, India and mainland SoutheastAsia. In India alone, losses by 2030 could rangefrom 1 000 to 2 000 km2, destroying perhaps70 000 to 150 000 livelihoods.

There are still considerable uncertainties inmost of the projections. The overall effect onglobal food production by 2030 is likely to besmall: cereal yields, for example, are projectedto decline by about 0.5 percent by the 2020s.But there will be large regional variations: intemperate regions an increase in yields isthought possible; in East Asia, the Sahel andSouthern Africa the outcome could be eitherpositive or negative; in other developing regionsa decline in yields is thought more likely. In allof these cases the potential yield changes areup or down by 2.5 percent or less by 2030 andby 5 percent or less by 2050.

It is important to note that these are only thechanges that may result from global warming inthe absence of any other factors. In practice,changes in technology are likely to reduce oroutweigh the impact of climate change. Amongthe most important technological changes willbe improved crop varieties and cropping

80

practices, which will raise yields. Factors suchas the spread of NT/CA and the expansion ofirrigation will combine with the dissemination ofnew crop varieties to reduce the sensitivity ofsome systems to climate change.

Inequalities in food security may accentuateOverall, global warming seems likely to benefitagriculture in developed countries located in

temperate zones but to have an adverse effecton production in many developing countries intropical and subtropical zones. Hence climatechange could increase the dependency ofdeveloping countries on imports and accentuateexisting North-South differences in foodsecurity.

Some future trends will cushion the blow.Improved communications and roads will allow

Many of the measures needed to reduce or adapt to climate change are valuable in coping withexisting problems such as water and air pollution, soil erosion, and vulnerability to droughts orfloods.

Measures to reduce greenhouse gas emissions:• Removal of subsidies and introduction of environmental taxes on chemical fertilizers and energy

inputs• Improvement of fertilizer use efficiency• Development of rice varieties emitting less methane• Improved management of livestock waste• Restoration of degraded lands• Improvement of crop residue management• Expansion of agroforestry and reforestation

Measures to promote adaptation to climate change:• Development and distribution of crop varieties and livestock breeds resistant to drought, storms

and floods, higher temperatures and saline conditions• Improvement of water use efficiency through:

• No-till/conservation agriculture in rainfed areas• Appropriate water pricing, management and technology in irrigated areas

• Promotion of agroforestry to increase ecosystem resilience and maintain biodiversity• Maintenance of livestock mobility in pastoral areas subject to drought

Measures to reduce food insecurity:• Reduction of rural and urban poverty• Improvement of transport and communications in areas vulnerable to disasters• Development of early-warning and storm-forecasting systems• Preparedness plans for relief and rehabilitation• Introduction of flood- and storm-resistant and salt-tolerant crops• Introduction of land use systems to stabilize slopes and reduce the risk of soil erosion and

mudslides• Building of homes, livestock shelters and food stores above likely flood levels.

Technology and policy choices

81

food to be transported more quickly intodrought- or flood-affected areas. Economicgrowth and rising incomes will still allow mostpeople in most countries to improve theirnutrition levels. A continuing shift out ofagricultural occupations into industry andservices, and out of rural and marginal areasinto urban centres, will leave fewer countriesunable to pay for food imports, and fewer peoplevulnerable to local declines in food production.

But the food security of poor people andcountries could well be reduced by climatechange. Even by 2030 there will still behundreds of millions of such people, who will beeither undernourished or on the brink ofundernourishment. They will be especiallyvulnerable to disruption of their incomes or foodsupply by crop failure or by extreme events suchas drought and floods.

As long as agricultural trade is not entirelyfree and communications with marginal areasremain poor, differences between local, nationaland international prices will persist, with theresult that food prices in areas hit by extremeevents could rise steeply, even if onlytemporarily. For example, in the south ofMozambique maize prices in the spring of 2000increased rapidly following the floods, while inthe north they remained at half the level in the

south or even declined somewhat, becausetransport between the two zones was difficult.

The adverse impacts of climate change willfall disproportionately on the poor. Hardest hitwill be small-scale farmers and other low-income groups in areas prone to drought,flooding, salt water intrusion or sea surges, andfishers affected by falling catches caused byhigher sea temperatures and shifts in currents.The areas most likely to suffer increasedclimate variability and extreme events aremostly those that are already handicapped bythese same phenomena. Many of the areas atrisk from rising sea levels are currently poorand may not enjoy the economic developmentnecessary to pay for flood protection.

The problem of increased vulnerability tofood insecurity caused by climate change islikely to be most serious in some 30 to 40countries. Major concern centres on Africa.Some estimate that, even as early as 2020 or2030, climate change could depress cerealproduction in this region by 2 to 3 percent,enough to increase the numbers at risk ofhunger by 10 million. This is the projected effectin the absence of other changes and could beoffset by even a modest annual increase inyields, but it still represents an additional hurdlethat agriculture in Africa must leap.

83

Annex 1

Developing countries

Africa, sub-Saharan Latin America Near East and South Asia East Asiaand Caribbean North Africa

AngolaBeninBotswanaBurkina FasoBurundiCameroonCentral Afr. Rep.ChadCongoCôte d’IvoireDem. Rep. of CongoEritreaEthiopiaGabonGambiaGhanaGuineaKenyaLesothoLiberiaMadagascarMalawiMaliMauritaniaMauritiusMozambiqueNamibiaNigerNigeriaRwandaSenegalSierra LeoneSomaliaSudanSwazilandTogoUgandaUnited Rep. of TanzaniaZambiaZimbabweSub-Saharan Africa,other1

ArgentinaBoliviaBrazilChileColombiaCosta RicaCubaDominican Rep.EcuadorEl SalvadorGuatemalaGuyanaHaitiHondurasJamaicaMexicoNicaraguaPanamaParaguayPeruSurinameTrinidad and TobagoUruguayVenezuelaLatin America, other 2

AfghanistanAlgeriaEgyptIran, Islamic Rep.IraqJordanLebanonLibyan Arab Yam.MoroccoSaudi ArabiaSyrian Arab Rep.TunisiaTurkeyYemenNear East, other 3

BangladeshIndiaMaldivesNepalPakistanSri Lanka

CambodiaChinaDem. Rep. of KoreaIndonesiaLao People’s Dem. Rep.MalaysiaMongoliaMyanmarPhilippinesRep. of KoreaThailandViet NamEast Asia, other 4

Countries and commodities covered

1 Cape Verde, Comoros, Djibouti, Guinea Bissau, Sao Tomé and Principe, Seychelles.2 Antigua, Bahamas, Barbados, Belize, Bermuda, Dominica, Grenada, Netherland Antilles, Saint Kitts and Nevis, Saint Lucia, Saint Vincent/Grenadines.3 Cyprus, Kuwait, United Arab Emirates.4 Brunei, Macau, Solomon Islands, Fiji, French Polynesia, New Caledonia, Vanuatu, Papua New Guinea, Kiribati.

84

Industrial countries

European Union1 Other Western Europe North America Oceania Other

AustriaBelgiumDenmarkFinlandFranceGermanyGreeceIrelandItalyLuxembourgNetherlandsPortugalSpainSwedenUnited Kingdom

IcelandMaltaNorwaySwitzerland

CanadaUnited States

AustraliaNew Zealand

IsraelJapanSouth Africa

1 In the analysis the European Union was treated as one country group (EU-15).

Transition countries

Eastern Europe and the Commonwealth of Baltic Statesformer Yugoslavia SFR Independent States

AlbaniaBosnia and HerzegovinaBulgariaCroatiaCzech RepublicHungaryPolandRomaniaSlovakiaSloveniaThe Former Yugoslav Rep. of MacedoniaYugoslavia, Fed. Rep.

ArmeniaAzerbaijanBelarusGeorgiaKazakhstanKyrgyzstanMoldova RepublicRussian FederationTajikistanTurkmenistanUkraineUzbekistan

EstoniaLatviaLithuania

Countries and commodities covered (continued)

85

Countries and commodities covered (continued)

WheatRice (paddy)MaizeBarleyMilletSorghumOther cerealsPotatoSweet potato and yamsCassavaOther rootsPlantainsSugar, raw1

PulsesVegetablesBananasCitrus fruitOther fruitVegetable oil and oilseeds (in vegetable oil equivalent)2

Cocoa beanCoffeeTeaTobaccoCotton lintJute and hard fibresRubber

Commodities

Crops

Beef, veal and buffalo meatMutton, lamb and goat meatPig meatPoultry meatMilk and dairy products (in whole milk equivalent)Eggs

Livestock

1 Sugar production in the developing countries was analysed separately for sugar cane and sugar beet.2 Vegetable oil production in the developing countries was analysed separately for soybean, groundnut, sesame seed, coconut, sunflower seed, palm oil/palm-kernel

oil, rapeseed and all other oilseeds.

NoteAll commodity data and projections in this report areexpressed in terms of primary product equivalent unlessstated otherwise. Historical commodity balances (SupplyUtilization Accounts - SUAs) are available for about 160primary and 170 processed crop and livestock commodi-ties. To reduce this amount of information to manageableproportions, all the SUA data were converted to thecommodity specification given above in the list ofcommodities, applying appropriate conversion factors(and ignoring joint products to avoid double counting:e.g. wheat flour is converted back into wheat while wheatbran is ignored). In this way, one SUA in homogeneousunits is derived for each of the commodities of the study.

Meat production refers to indigenous meat production,i.e. production from slaughtered animals plus the meatequivalent of live animal exports minus the meatequivalent of all live animal imports. Cereals demandand trade data include the grain equivalent of beerconsumption and trade.

The commodities for which SUAs were constructedare the 26 crops and 6 livestock products given in the listabove. The production analysis for the developingcountries was, however, carried out for 34 crops becausesugar and vegetable oils were analysed separately (forproduction analysis only) for the 10 crops shown in thefootnote to the list.

86

Annex 2

Total population

Millions Annual increments (millions)

1979-81 1997-99 2015 2030 2050 1995 to 2010 to 2025 to 2045 to2000 2015 2030 2050

World (UN)1 4 430 5 900 7 207 8 270 9 322 79 76 67 43

World (FBS) 4 416 5 878 7 176 8 229 9 270 78 76 66 43Developing countries 3 245 4 573 5 827 6 869 7 935 74 74 66 45

Sub-Saharan Africa 345 574 883 1 229 1 704 15 20 24 23Near East and

North Africa 238 377 520 651 809 8 9 9 7Latin America and Car. 357 498 624 717 799 8 7 6 3South Asia 885 1 283 1 672 1 969 2 258 23 22 19 12East Asia 1 420 1 840 2 128 2 303 2 365 20 16 9 -1

Industrial countries 789 892 951 979 986 5 2 1 0Transition countries 382 413 398 381 349 0 -1 -1 -2

Growth rates (% per year)

Population Total GDP Per capita GDP

1969 to 1979 to 1989 to 1997-99 2015 to 1997-99 2015 to 1997-99 2015 to1999 1999 1999 to 2015 2030 to 2015 2030 to 2015 2030

World (FBS) 1.7 1.6 1.5 1.2 0.9 3.5 3.8 2.3 2.9Developing countries 2.0 1.9 1.7 1.4 1.1 5.1 5.5 3.7 4.4

Sub-Saharan Africa 2.9 2.9 2.7 2.6 2.2 4.4 4.5 1.8 2.3Near East and

North Africa 2.7 2.6 2.4 1.9 1.5 3.7 3.9 1.8 2.4Latin America and Car. 2.1 1.9 1.7 1.3 0.9 4.1 4.4 2.8 3.5South Asia 2.2 2.1 1.9 1.6 1.1 5.5 5.4 3.9 4.3East Asia 1.6 1.5 1.2 0.9 0.5 6.1 6.3 5.3 5.8

Industrial countries 0.7 0.7 0.7 0.4 0.2 3.0 3.0 2.6 2.8Transition countries 0.6 0.5 0.1 -0.2 -0.3 3.7 4.0 4.0 4.3

Table A1 Population and GDP, data and projections

Sources: Population: UN (2001)GDP to 2015: World Bank (2001b)

1 World (UN) covers all countries; World (FBS) covers all countries for which FAO Food Balance Sheet data are available.

87

1969 to 1999 1979 to 1999 1989 to 1999 1997-99 to 2015 2015 to 2030 1997-99 to 2030Demand

World 2.2 2.1 2.0 1.6 1.4 1.5Developing countries 3.7 3.7 4.0 2.2 1.7 2.0

Idem, excl. China 3.2 3.0 3.0 2.4 2.0 2.2Sub-Saharan Africa 2.8 3.1 3.2 2.9 2.8 2.9

Idem, excl. Nigeria 2.5 2.4 2.5 3.1 2.9 3.0Near East and North Africa 3.8 3.0 2.7 2.4 2.0 2.2Latin America and Caribbean 2.9 2.7 3.0 2.1 1.7 1.9

Idem, excl. Brazil 2.4 2.1 2.8 2.2 1.8 2.0South Asia 3.2 3.3 3.0 2.6 2.0 2.3East Asia 4.5 4.7 5.2 1.8 1.3 1.6

Idem, excl. China 3.5 3.2 2.8 2.0 1.7 1.9Industrial countries 1.1 1.0 1.0 0.7 0.6 0.7Transition countries -0.2 -1.7 -4.4 0.5 0.4 0.5

Production

World 2.2 2.1 2.0 1.6 1.3 1.5Developing countries 3.5 3.7 3.9 2.0 1.7 1.9

Idem, excl. China 3.0 3.0 2.9 2.3 2.0 2.1Sub-Saharan Africa 2.3 3.0 3.0 2.8 2.7 2.7

Idem, excl. Nigeria 2.0 2.2 2.4 2.9 2.7 2.8Near East and North Africa 3.1 3.0 2.9 2.1 1.9 2.0Latin America and Caribbean 2.8 2.6 3.1 2.1 1.7 1.9

Idem, excl. Brazil 2.3 2.1 2.8 2.1 1.8 2.0South Asia 3.1 3.4 2.9 2.5 1.9 2.2East Asia 4.4 4.6 5.0 1.7 1.3 1.5

Idem, excl. China 3.3 2.9 2.4 1.9 1.8 1.9Industrial countries 1.3 1.0 1.4 0.8 0.6 0.7Transition countries -0.4 -1.7 -4.7 0.6 0.6 0.6

Population

World 1.7 1.6 1.5 1.2 0.9 1.1Developing countries 2.0 1.9 1.7 1.4 1.1 1.3

Idem, excl. China 2.3 2.2 2.0 1.7 1.3 1.5Sub-Saharan Africa 2.9 2.9 2.7 2.6 2.2 2.4

Idem, excl. Nigeria 2.9 2.9 2.7 2.6 2.3 2.4Near East and North Africa 2.7 2.6 2.4 1.9 1.5 1.7Latin America and Caribbean 2.1 1.9 1.7 1.3 0.9 1.1

Idem, excl. Brazil 2.1 1.9 1.8 1.4 1.0 1.2South Asia 2.2 2.1 1.9 1.6 1.1 1.3East Asia 1.6 1.5 1.2 0.9 0.5 0.7

Idem, excl. China 2.0 1.8 1.6 1.2 0.9 1.0Industrial countries 0.7 0.7 0.7 0.4 0.2 0.3Transition countries 0.6 0.5 0.1 -0.2 -0.3 -0.2

Table A2 Growth rates of aggregate demand and production (percent per annum)

88

Food consumption (kcal/capita/day)

1964-66 1974-76 1984-86 1997-99 2015 2030

World 2 358 2 435 2 655 2 803 2 940 3 050Developing countries 2 054 2 152 2 450 2 681 2 850 2 980

Sub-Saharan Africa 2 058 2 079 2 057 2 195 2 360 2 540Idem, excl. Nigeria 2 037 2 076 2 057 2 052 2 230 2 420

Near East and North Africa 2 290 2 591 2 953 3 006 3 090 3 170Latin America and Caribbean 2 393 2 546 2 689 2 824 2 980 3 140South Asia 2 017 1 986 2 205 2 403 2 700 2 900East Asia 1 957 2 105 2 559 2 921 3 060 3 190

Industrial countries 2 947 3 065 3 206 3 380 3 440 3 500Transition countries 3 222 3 385 3 379 2 906 3 060 3 180

Incidence of undernourishment, developing countries

% population Million people1990-92 1997-99 2015 2030 1990-92 1997-99 2015 2030

Developing countries 20 17 11 6 815 776 610 443Sub-Saharan Africa 35 34 23 15 168 194 205 183

Idem, excl. Nigeria 40 40 28 18 156 186 197 178Near East and North Africa 8 9 7 5 25 32 37 34Latin America and Caribbean 13 11 6 4 59 54 40 25South Asia 26 24 12 6 289 303 195 119East Asia 16 11 6 4 275 193 135 82

Population living in countries with given per capita food consumption (million)

kcal/capita/day Population (million)1964-66 1974-76 1984-86 1997-99 2015 2030

Under 2200 1 8931 2 2811 558 571 462 1962200-2500 288 307 1 2902 1 4872 541 8372500-2700 154 141 1 3373 222 351 3522700-3000 302 256 306 1 134 2 3972 2 4512

Over 3000 688 1 069 1 318 2 4643 3 4253 4 3923

World total 3 325 4 053 4 810 5 878 7 176 8 229

1 Includes India and China.2 Includes India.3 Includes China.

Developing countries with a given percentage of undernourishment4

Population (million) kcal/capita/day % of population Million people1997-99 2015 2030 1997-99 2015 2030 1997-99 2015 2030 1997-99 2015 2030

Under 5% 349 1 158 5 129 3 187 3 130 3 150 2 3 3 8 37 1785-10% 1 989 2 162 524 2 999 3 066 2 758 8 6 7 167 134 3810-25% 1 632 1 939 948 2 434 2 644 2 411 21 13 16 349 250 155Over 25% 586 544 239 1 988 2 085 2 149 43 35 30 251 190 72

Total 4 555 5 804 6 840 2 681 2 850 2 980 17 11 6 776 611 443

4 Different countries form each group in the different years.

Table A3 Per capita food consumption and undernourishment

89

Cereals Roots and Sugar Pulses Vegetable oils, Meat Milk and dairy tubers (raw eq.) (dry) oilseeds (oil eq.) (carcass weight) (fresh milk eq.)

kg/capita/year

World1979-81 160 74 23.5 6.5 8.4 29.5 771997-99 171 69 24.0 5.9 11.4 36.4 782015 171 71 25.1 5.9 13.7 41.3 832030 171 74 26.3 6.1 15.8 45.3 90

Industrial countries1979-81 139 67 36.8 2.8 15.7 78.5 2021997-99 159 66 33.1 3.8 20.2 88.2 2122015 158 63 32.4 4.0 21.6 95.7 2172030 159 61 32.0 4.1 22.9 100.1 221

Transition countries1979-81 189 119 45.9 3.1 9.2 62.9 1811997-99 173 104 34.0 1.2 9.3 46.2 1592015 176 102 35.0 1.2 11.5 53.8 1692030 173 100 36.0 1.1 14.2 60.7 179

Developing countries1979-81 162 70 17.6 7.8 6.5 13.7 341997-99 173 67 21.3 6.8 9.9 25.5 452015 173 71 23.2 6.6 12.6 31.6 552030 172 75 25.0 6.6 14.9 36.7 66

Sub-Saharan Africa1979-81 115 172 9.9 9.8 8.5 10.6 341997-99 123 194 9.5 8.8 9.2 9.4 292015 131 199 11.3 9.8 10.7 10.9 312030 141 202 13.0 10.5 12.3 13.4 34

Near East and North Africa1979-81 199 26 28.2 6.4 11.1 17.4 851997-99 209 34 27.6 6.7 12.8 21.2 722015 206 33 28.7 6.9 14.4 28.6 812030 201 33 29.9 6.9 15.7 35.0 90

Latin America and Caribbean1979-81 130 74 48.5 12.6 10.2 40.6 971997-99 132 62 48.9 11.1 12.5 53.8 1102015 136 61 48.2 10.7 14.5 65.3 1252030 139 61 47.9 10.6 16.3 76.6 140

South Asia1979-81 151 20 20.7 11.2 5.8 4.0 421997-99 163 22 26.7 10.9 8.4 5.3 682015 177 27 29.5 9.1 11.6 7.6 882030 183 30 32.2 7.9 14.0 11.7 107

East Asia1979-81 181 83 8.1 4.3 4.7 13.0 51997-99 199 66 12.4 2.1 9.7 37.7 102015 190 64 14.6 2.0 13.1 50.0 142030 183 61 16.6 2.1 16.3 58.5 18

Table A4 Changes in commodity composition of food

90

Demand Production Net trade SSR1 Growth ratesmillion tonnes % % per year

Food Feed All uses Demand ProductionWorld

1979-81 706 575 1 437 1 442 3 100 1979-99 1.4 1.41997-99 1 003 657 1 864 1 889 9 101 1989-99 1.0 1.02015 1 227 911 2 379 2 387 8 100 1997-99 to 2015 1.4 1.42030 1 406 1 148 2 831 2 839 8 100 2015-2030 1.2 1.2

Industrial countries1979-81 110 286 428 551 111 129 1979-99 1.0 0.81997-99 142 331 525 652 111 124 1989-99 1.7 1.42015 150 387 599 785 187 131 1997-99 to 2015 0.8 1.12030 155 425 652 899 247 138 2015-2030 0.6 0.9

Transition countries1979-81 72 176 297 242 -41 81 1979-99 -1.9 -1.11997-99 72 105 211 210 1 100 1989-99 -4.9 -4.22015 70 127 237 247 10 104 1997-99 to 2015 0.7 1.02030 66 149 261 287 25 110 2015-2030 0.7 1.0

Developing countries1979-81 524 113 712 649 -66 91 1979-99 2.6 2.51997-99 790 222 1 129 1 026 -103 91 1989-99 2.2 2.12015 1 007 397 1 544 1 354 -190 88 1997-99 to 2015 1.9 1.62030 1 185 573 1 917 1 652 -265 86 2015-2030 1.5 1.3

Sub-Saharan Africa1979-81 40 2 48 41 -8 85 1979-99 3.4 3.41997-99 71 4 86 71 -14 82 1989-99 3.1 2.72015 116 8 139 114 -25 82 1997-99 to 2015 2.9 2.82030 173 15 208 168 -40 81 2015-2030 2.7 2.6

Near East and North Africa1979-81 47 19 80 58 -24 72 1979-99 2.7 2.41997-99 79 34 133 83 -49 63 1989-99 2.2 1.32015 107 62 192 107 -85 56 1997-99 to 2015 2.2 1.52030 131 93 249 133 -116 54 2015-2030 1.8 1.5

Latin America and Caribbean1979-81 46 37 94 87 -8 93 1979-99 2.3 1.81997-99 66 60 142 125 -14 88 1989-99 2.8 3.12015 85 98 204 188 -16 92 1997-99 to 2015 2.1 2.42030 99 135 257 244 -13 95 2015-2030 1.6 1.8

South Asia1979-81 134 2 151 147 -2 98 1979-99 2.6 2.71997-99 208 3 234 239 -3 102 1989-99 1.8 2.02015 295 11 335 323 -12 97 1997-99 to 2015 2.1 1.82030 360 22 416 393 -22 95 2015-2030 1.5 1.3

East Asia1979-81 257 53 339 316 -24 93 1979-99 2.5 2.51997-99 366 120 534 507 -23 95 1989-99 2.1 2.12015 404 218 675 622 -53 92 1997-99 to 2015 1.4 1.22030 422 309 787 714 -73 91 2015-2030 1.0 0.9

1 SSR = Self-sufficiency ratio = Production / total demand.

Table A5 Cereal balances

91

Tabl

e A

6B

alan

ces

for

sele

cted

com

mod

itie

s (m

illio

n to

nnes

)

Dev

elop

ing

coun

trie

sIn

dust

rial

cou

ntri

esTr

ansi

tion

cou

ntri

esD

eman

dP

rodu

ctio

nN

et tr

ade

Dem

and

Pro

duct

ion

Net

trad

eD

eman

dP

rodu

ctio

nN

et tr

ade

Food

1Fe

edA

ll us

esFo

od 1

Feed

All

uses

Food

1Fe

edA

ll us

es

Whe

at 1979

-81

171.

87.

520

5.1

156.

8-4

8.7

62.6

22.7

95.8

168.

165

.754

.360

.513

8.0

112.

5-1

5.7

1997

-99

289.

612

.933

8.4

280.

2-6

1.8

75.3

52.1

142.

321

5.9

66.0

56.1

28.1

101.

810

0.8

-0.3

2015

392.

327

.746

1.8

358.

1-1

03.7

76.1

66.1

158.

626

2.5

103.

954

.736

.110

9.9

113.

53.

620

3047

8.1

41.2

566.

042

4.9

-141

.276

.570

.816

5.2

298.

513

3.3

51.6

46.9

120.

013

1.5

11.6

Ric

e (p

addy

)19

79-8

133

3.1

5.8

370.

836

8.7

-2.6

18.0

0.2

19.6

22.7

4.2

3.7

0.0

4.0

2.6

-1.7

1997

-99

491.

217

.455

2.6

561.

93.

720

.80.

423

.324

.32.

12.

50.

02.

61.

2-1

.420

1559

8.4

32.2

679.

868

5.0

5.2

21.5

0.4

23.7

24.1

0.4

3.2

0.0

3.4

1.5

-1.9

2030

665.

951

.577

1.1

778.

06.

922

.00.

524

.323

.5-0

.83.

50.

03.

82.

0-1

.8C

oars

e gr

ains

1979

-81

130.

410

2.1

259.

524

6.5

-16.

035

.026

3.3

319.

136

7.4

42.3

15.6

115.

115

6.6

127.

8-2

4.5

1997

-99

172.

319

7.1

421.

837

1.0

-43.

252

.827

8.1

367.

142

0.3

43.4

13.8

76.5

107.

210

8.6

2.1

2015

215.

634

8.3

628.

853

9.4

-89.

459

.932

0.8

423.

950

6.7

82.8

13.1

90.4

124.

913

3.0

8.1

2030

262.

349

7.8

836.

970

8.6

-128

.264

.035

3.7

470.

758

5.3

114.

612

.110

2.5

139.

015

3.9

14.9

Roo

ts a

nd tu

bers

1979

-81

227.

162

.533

9.9

356.

416

.452

.430

.699

.683

.3-1

6.3

45.4

43.7

130.

212

8.6

-0.3

1997

-99

304.

497

.349

2.4

501.

98.

758

.815

.094

.983

.2-1

1.5

43.0

26.2

98.0

94.9

-0.5

2015

412.

414

2.6

653.

266

2.9

9.6

59.6

15.6

98.1

86.6

-11.

540

.627

.496

.797

.00.

420

3051

3.8

182.

280

8.2

817.

69.

459

.615

.799

.988

.3-1

1.6

38.2

28.9

94.8

95.4

0.7

Suga

r (r

aw e

q.)

1979

-81

57.2

2.3

67.8

74.0

7.3

29.1

0.2

30.5

28.9

-2.0

17.5

0.9

18.8

12.8

-4.8

1997

-99

97.2

3.9

121.

912

8.8

5.8

29.5

0.1

31.6

36.0

3.8

14.1

0.4

15.2

8.6

-5.9

2015

135.

46.

016

7.3

173.

05.

731

.50.

133

.836

.93.

113

.90.

515

.29.

9-5

.320

3017

1.4

8.2

208.

521

2.0

3.5

32.3

0.1

34.6

38.7

4.1

13.7

0.5

15.1

10.9

-4.3

Pul

ses

1979

-81

25.3

2.6

31.8

31.5

-0.2

2.2

1.1

3.7

3.8

0.1

1.2

3.6

5.8

5.5

0.1

1997

-99

31.0

4.7

40.5

39.3

-1.0

3.4

6.8

11.1

13.2

1.7

0.5

2.3

3.4

3.5

0.2

2015

38.3

7.4

51.9

51.0

-0.9

3.8

7.3

12.0

12.8

0.8

0.5

2.5

3.5

3.9

0.4

2030

45.4

10.5

63.5

62.4

-1.1

4.0

7.7

12.6

13.4

0.8

0.4

2.6

3.7

4.2

0.5

Vege

tabl

e oi

ls, o

ilsee

ds (o

il eq

.)19

79-8

121

.00.

526

.428

.61.

512

.30.

517

.017

.2-0

.43.

50.

35.

64.

7-0

.919

97-9

945

.11.

561

.867

.74.

018

.02.

130

.630

.2-0

.93.

80.

56.

05.

8-0

.220

1573

.22.

410

5.7

109.

13.

420

.62.

440

.940

.4-0

.54.

60.

67.

47.

2-0

.220

3010

2.3

3.3

152.

215

5.6

3.5

22.4

2.5

53.2

52.0

-1.3

5.4

0.6

9.1

9.2

0.0

Coc

oa 1979

-81

0.2

0.0

0.3

1.7

1.3

1.1

0.0

1.1

0.0

-1.1

0.2

0.0

0.2

0.0

-0.2

1997

-99

0.6

0.0

1.0

3.0

2.1

1.9

0.0

1.9

0.0

-1.8

0.4

0.0

0.4

0.0

-0.4

2015

1.0

0.0

1.4

4.1

2.7

2.2

0.0

2.3

0.0

-2.3

0.5

0.0

0.5

0.0

-0.5

2030

1.4

0.0

1.9

5.0

3.2

2.5

0.0

2.6

0.0

-2.6

0.6

0.0

0.6

0.0

-0.6

Cof

fee 1979

-81

1.5

0.0

1.6

5.3

3.5

3.3

0.0

3.3

0.0

-3.3

0.2

0.0

0.2

0.0

-0.2

1997

-99

1.6

0.0

1.8

6.5

4.6

3.9

0.0

4.0

0.0

-4.0

0.5

0.0

0.5

0.0

-0.5

2015

2.5

0.0

2.7

7.8

5.2

4.5

0.0

4.5

0.0

-4.6

0.6

0.0

0.6

0.0

-0.6

2030

3.4

0.0

3.6

9.4

5.8

4.9

0.0

4.9

0.0

-5.2

0.8

0.0

0.8

0.0

-0.8

92

Dev

elop

ing

coun

trie

sIn

dust

rial

cou

ntri

esTr

ansi

tion

cou

ntri

esD

eman

dP

rodu

ctio

nN

et tr

ade

Dem

and

Pro

duct

ion

Net

trad

eD

eman

dP

rodu

ctio

nN

et tr

ade

Food

1Fe

edA

ll us

esFo

od 1

Feed

All

uses

Food

1Fe

edA

ll us

es

Tea

1979

-81

1.4

0.0

1.4

1.9

0.5

0.5

0.0

0.5

0.1

-0.4

0.2

0.0

0.2

0.1

-0.1

1997

-99

2.8

0.0

2.9

3.7

0.8

0.5

0.0

0.5

0.1

-0.4

0.3

0.0

0.3

0.1

-0.2

2015

4.1

0.0

4.2

5.1

0.9

0.6

0.0

0.6

0.1

-0.5

0.4

0.0

0.3

0.1

-0.3

2030

5.4

0.0

5.5

6.5

1.9

0.6

0.0

0.6

0.1

-0.5

0.4

0.0

0.3

0.1

-0.5

Bee

f and

vea

l19

79-8

115

.10.

015

.215

.70.

522

.70.

022

.822

.80.

18.

80.

08.

98.

7-0

.219

97-9

927

.80.

028

.128

.0-0

.123

.30.

023

.425

.01.

56.

30.

06.

45.

7-0

.620

1541

.20.

041

.541

.2-0

.324

.60.

024

.726

.61.

97.

00.

07.

16.

3-0

.820

3055

.30.

055

.755

.0-0

.724

.70.

024

.726

.51.

87.

40.

07.

56.

9-0

.6M

utto

n an

d la

mb

1979

-81

3.9

0.0

4.0

3.8

-0.2

2.0

0.0

2.1

2.5

0.4

1.2

0.0

1.2

1.2

0.0

1997

-99

7.6

0.0

7.6

7.4

-0.3

2.2

0.0

2.3

2.7

0.4

0.7

0.0

0.8

0.8

0.0

2015

11.9

0.0

12.0

11.2

-0.7

2.2

0.0

2.3

3.1

0.8

0.9

0.0

1.0

0.9

0.0

2030

16.6

0.0

16.6

15.4

-1.2

2.2

0.0

2.3

3.5

1.3

1.1

0.0

1.1

1.1

0.0

Pig

mea

t19

79-8

117

.10.

017

.417

.50.

123

.70.

023

.823

.7-0

.110

.30.

010

.510

.60.

119

97-9

949

.40.

049

.549

.3-0

.228

.30.

028

.529

.30.

98.

30.

08.

47.

9-0

.520

1570

.00.

070

.169

.5-0

.631

.20.

031

.332

.30.

98.

50.

08.

58.

4-0

.120

3083

.50.

083

.682

.8-0

.831

.90.

032

.133

.11.

08.

40.

08.

58.

60.

1P

oult

ry m

eat

1979

-81

8.2

0.0

8.3

7.7

-0.6

13.6

0.0

13.7

14.3

0.6

3.8

0.0

3.8

3.9

0.0

1997

-99

31.5

0.0

31.9

31.3

-0.7

24.9

0.0

25.1

27.7

2.6

3.7

0.0

3.8

2.9

-0.9

2015

61.2

0.0

61.4

59.1

-2.3

33.0

0.0

33.1

37.5

4.3

5.0

0.0

5.1

4.1

-1.0

2030

96.5

0.0

96.7

93.5

-3.2

39.1

0.0

39.3

44.1

4.8

6.3

0.0

6.4

5.7

-0.6

Milk

and

dai

ry (f

resh

milk

eq.

)19

79-8

111

0.2

12.7

129.

511

2.3

-17.

615

8.9

40.1

207.

522

4.9

18.5

69.2

50.5

126.

512

7.3

0.8

1997

-99

203.

723

.223

9.1

219.

3-1

9.8

189.

423

.922

5.8

245.

819

.765

.726

.694

.596

.62.

220

1532

2.8

36.4

375.

834

6.2

-29.

620

6.1

24.8

240.

426

8.5

28.1

67.0

27.2

96.9

100.

43.

520

3045

1.8

50.8

523.

148

4.0

-39.

121

6.2

25.4

250.

528

6.3

35.8

68.1

27.7

98.6

103.

85.

2C

otto

n lin

t19

79-8

18.

30.

08.

38.

2-0

.13.

60.

03.

63.

4-0

.22.

60.

02.

62.

70.

119

97-9

914

.00.

014

.012

.1-1

.73.

90.

03.

94.

80.

90.

80.

00.

81.

50.

720

1520

.10.

020

.118

.0-2

.14.

50.

04.

56.

01.

51.

00.

01.

01.

70.

720

3025

.30.

025

.322

.7-2

.54.

80.

04.

86.

92.

01.

20.

01.

21.

80.

6R

ubbe

r19

79-8

11.

20.

01.

23.

82.

52.

10.

02.

10.

0-2

.10.

40.

00.

40.

0-0

.419

97-9

93.

30.

03.

26.

63.

43.

10.

03.

10.

0-3

.10.

20.

00.

20.

0-0

.220

154.

40.

04.

47.

83.

43.

20.

03.

20.

0-3

.20.

30.

00.

30.

0-0

.320

305.

40.

05.

48.

93.

53.

20.

03.

20.

0-3

.20.

40.

00.

40.

0-0

.4

1In

dust

rial

use

for

cott

on a

nd r

ubbe

r

Tabl

e A

6B

alan

ces

for

sele

cted

com

mod

itie

s (m

illio

n to

nnes

)(co

ntin

ued)

93

Arable land Harvested land Cropping intensitymillion ha million ha %

Total Rainfed Irrigated Total Rainfed Irrigated Total Rainfed Irrigated

Developing countries1997-99 956 754 202 885 628 257 93 83 1272015 1 017 796 221 977 671 306 96 84 1382030 1 076 834 242 1 063 722 341 99 87 141

Sub-Saharan Africa1997-99 228 223 5.3 154 150 4.5 68 67 862015 262 256 6.0 185 179 5.7 71 70 952030 288 281 6.8 217 210 7.0 76 75 102

Near East and North Africa1997-99 86 60 26 70 43 27 81 72 1022015 89 60 29 77 45 32 86 75 1102030 93 60 33 84 46 37 90 78 112

Latin America and Caribbean1997-99 203 185 18 127 112 16 63 60 862015 223 203 20 150 131 19 67 64 952030 244 222 22 172 150 22 71 68 100

South Asia1997-99 207 126 81 230 131 100 111 103 1242015 210 123 87 248 131 117 118 106 1342030 216 121 95 262 131 131 121 109 137

East Asia1997-99 232 161 71 303 193 110 130 120 1542015 233 155 78 317 186 131 136 120 1682030 237 151 85 328 184 144 139 122 169

Table A7 Land use

94

Table A8 Yield and harvested land for selected crops

Yield Harvested land(tonnes/ha) (million ha)

1979-81 1997-99 2015 2030 1979-81 1997-99 2015 2030

WheatDeveloping countries 1.64 2.53 3.11 3.53 95.6 110.7 113.3 118.4

Sub-Saharan Africa 1.30 1.62 2.03 2.44 1.0 1.6 2.2 2.8Near East and North Africa 1.35 1.83 2.21 2.56 25.3 27.2 27.9 29.0Latin America and Caribbean 1.50 2.53 2.84 3.17 10.1 8.9 9.5 10.5South Asia 1.55 2.46 3.12 3.77 30.0 36.3 40.2 43.8East Asia 2.04 3.15 3.99 4.30 29.1 36.7 33.5 32.2

Rice (paddy)Developing countries 2.65 3.57 4.21 4.73 138.0 156.7 162.1 163.9

Sub-Saharan Africa 1.36 1.63 2.19 2.79 4.5 7.1 8.6 10.1Near East and North Africa 4.01 5.63 6.17 6.72 1.2 1.6 1.9 2.2Latin America and Caribbean 1.94 3.47 4.35 4.91 8.0 5.9 6.4 6.9South Asia 1.91 2.92 3.80 4.32 54.5 59.3 62.2 63.9East Asia 3.36 4.17 4.67 5.23 70.0 82.8 83.0 80.8

MaizeDeveloping countries 1.96 2.78 3.44 3.96 75.5 96.5 117.8 136.2

Sub-Saharan Africa 1.14 1.25 1.61 1.97 12.1 20.7 27.2 33.9Near East and North Africa 2.39 4.66 5.29 6.39 2.3 2.2 2.6 3.2Latin America and Caribbean 1.84 2.79 3.59 4.18 25.2 26.8 32.3 36.6South Asia 1.14 1.68 2.25 2.72 7.1 8.0 8.5 8.8East Asia 2.59 3.70 4.50 5.12 28.7 38.8 47.1 53.6

BarleyDeveloping countries 1.29 1.42 1.74 2.05 16.6 16.9 18.2 19.6

Sub-Saharan Africa 1.21 1.06 1.34 1.65 0.9 1.0 1.2 1.4Near East and North Africa 1.11 1.31 1.61 1.86 10.9 11.6 12.6 13.4Latin America and Caribbean 1.36 1.87 2.51 3.04 0.9 1.0 1.3 1.8South Asia 1.07 1.75 1.95 2.08 2.0 1.0 0.9 0.9East Asia 2.70 1.79 2.18 2.64 1.8 2.3 2.2 2.2

Sugar caneDeveloping countries 54.9 61.8 77.4 88.1 12.4 18.7 20.5 22.0

Sub-Saharan Africa 56.8 49.5 62.8 75.0 0.6 0.9 1.2 1.5Near East and North Africa 79.1 103.9 105.1 108.4 0.1 0.2 0.2 0.3Latin America and Caribbean 57.3 64.7 76.0 82.8 6.2 8.5 8.9 9.3South Asia 48.6 63.0 84.8 100.2 3.7 5.4 6.1 6.7East Asia 56.9 54.7 71.5 83.6 1.8 3.8 4.0 4.2

PulsesDeveloping countries 0.61 0.67 0.85 1.09 51.8 60.0 59.7 57.1

Sub-Saharan Africa 0.55 0.44 0.66 0.93 7.8 15.8 17.4 18.4Near East and North Africa 0.92 0.89 1.12 1.26 2.3 3.8 4.5 5.0Latin America and Caribbean 0.58 0.84 0.98 1.06 8.3 7.3 7.3 7.8South Asia 0.47 0.62 0.81 1.05 25.9 25.7 23.0 19.4East Asia 1.07 1.04 1.15 1.54 7.6 7.4 7.5 6.6

95

Yield Harvested land(tonnes/ha) (million ha)

1979-81 1997-99 2015 2030 1979-81 1997-99 2015 2030

SoybeanDeveloping countries 1.37 1.84 2.24 2.63 21.2 40.8 56.5 71.5

Sub-Saharan Africa 0.56 0.85 1.11 1.40 0.4 0.8 1.2 1.7Near East and North Africa 1.94 1.84 2.66 3.19 0.1 0.1 0.2 0.3Latin America and Caribbean 1.66 2.33 2.74 3.15 11.2 21.6 30.7 39.7South Asia 0.68 1.09 1.40 1.70 0.5 6.1 9.1 11.8East Asia 1.08 1.41 1.83 2.21 9.0 12.1 15.2 18.0

GroundnutDeveloping countries 0.93 1.28 1.51 1.69 17.6 23.3 31.1 38.5

Sub-Saharan Africa 0.70 0.83 1.06 1.29 5.9 8.7 12.2 16.2Near East and North Africa 1.76 2.42 2.85 3.23 0.1 0.1 0.2 0.2Latin America and Caribbean 1.35 1.60 1.72 1.85 0.8 0.7 0.8 1.0South Asia 0.84 1.03 1.27 1.43 7.2 7.4 8.7 9.5East Asia 1.38 2.12 2.28 2.43 3.6 6.5 9.2 11.6

SeedcottonDeveloping countries 0.96 1.35 1.84 2.17 25.4 26.2 28.6 30.5

Sub-Saharan Africa 0.57 0.85 1.06 1.25 2.9 4.2 5.3 6.2Near East and North Africa 2.12 2.70 2.93 3.10 1.6 1.6 2.2 2.6Latin America and Caribbean 0.90 1.49 1.70 1.85 5.5 2.1 2.6 3.1South Asia 0.61 0.91 1.54 2.08 10.1 12.1 12.8 13.1East Asia 1.56 2.16 2.88 3.14 5.2 6.1 5.7 5.6

RubberDeveloping countries 0.69 0.91 1.07 1.18 5.5 7.2 7.3 7.5

Sub-Saharan Africa 0.69 0.71 0.95 1.19 0.3 0.5 0.6 0.6Near East and North Africa 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0Latin America and Caribbean 3.81 1.05 1.18 1.31 0.0 0.1 0.2 0.2South Asia 0.67 1.20 1.34 1.50 0.4 0.6 0.6 0.6East Asia 0.68 0.90 1.05 1.15 4.7 6.0 6.0 6.1

Table A8 Yield and harvested land for selected crops (continued)

96

Given below is a list of sources for tables and figures only. A complete list of sources may be foundin the main report.

Alexandratos, N. (ed.) 1988. World agriculture towards 2000, an FAO study. London: BelhavenPress, and New York, USA: New York University Press.

Alexandratos, N. (ed.) 1995. World agriculture towards 2010, an FAO Study. Chichester, UK: JohnWiley and Sons, and Rome: FAO.

Anderson, K., François, J., Hertel, T., Hoekman, B. & Martin, W. 2000. Potential gains from tradereform in the new millennium. Paper presented at the Third Annual Conference on GlobalEconomic Analysis, 27-30 June 2000, Monash University, Melbourne, Australia.

FAO. 1970. Provisional indicative world plan for agricultural development. Rome.FAO. 1981. Agriculture towards 2000. Rome.FAO. 2001. Global forest resources assessment: main report. FAO Forestry Paper 140. Rome.FAO, forthcoming. World agriculture towards 2015/30, an FAO study. Rome.Fischer, G., van Velthuizen, H. & Nachtergaele, F. 2000. Global agro-ecological zones assessment:

methodology and results. Interim report. Laxenburg, Austria: International Institute for SystemsAnalysis (IIASA), and Rome: FAO.

Gallup, J., Sachs, J. & Mellinger, A. 1999. Geography and economic development. CID WorkingPaper No. 1. Harvard, USA: Harvard University.

Huang, J., Rozell, S., Pray, C. & Wang, Q. 2002. Plant biotechnology in China. Science 295: 674–677.ISAAA. 2001. Global preview of commercialised transgenic crops. ISAAA Briefs Nos 21–24. Cornell,

USA: Cornell University.Mosier, A. & Kroeze, C. 1998. A new approach to estimating emissions of nitrous oxide from

agriculture and its implications for the global change N20 budget. IGBP Global ChangeNewsletter 34: 8–13.

Oakridge National Laboratory. 2000. Landscan global population density 2000 map. Oak Ridge, USA.Oldeman, L., Hakkeling, R. & Sombroek, W. 1991. World map of the status of human-induced soil

degradation. Wageningen, Netherlands: ISRIC, and Nairobi: UNEP.UN. 2001. World population prospects: the 2000 revision: highlights. New York, USA.Willer, H. & Yussefi, M. 2002. Organic agriculture worldwide 2001: statistics and future prospects.

Special publication. Stuttgart, Germany: Foundation for Ecology and Agriculture.World Bank. 2001a. World development indicators. Washington DC.World Bank. 2001b. Global economic prospects and the developing countries, 2002. Washington DC.

Sources

97

Acronyms

AIDS acquired immunodeficiency syndromeAoA Agreement on AgricultureBMI body mass indexBSE bovine spongiform encephalopathyBt Bacillus thuringensisEU European UnionGDP gross domestic productGLASOD Global Assessment of Land DegradationGM genetically modifiedha hectareHIV human immunodeficiency virusIPCC Intergovernmental Panel on Climate ChangeIPM integrated pest managementIUCN World Conservation UnionLDC least developed countryMNE multinational enterpriseNPK nitrogen, phosphorus and potassiumNT/CA no-till/conservation agricultureNWFP non-wood forest productOECD Organisation for Economic Co-operation and Developmentppm parts per millionSFM sustainable forest managementUN United NationsvCJD variant Creutzfeldt-Jakob disease