Global Change and Sustainable Development: Critical Trends

Report of the Secretary-General

                                                                  1997

Table of Contents

I. Development and Environment: from Stockholm to Rio

II. Trends in World Population

III. Energy and Materials Consumption

IV. Agriculture and Food Supply

V. Water: A Multifunction Resource

VI. Human Development

VII. Conclusions

I. DEVELOPMENT AND ENVIRONMENT: FROM STOCKHOLM TO RIO Introduction

1. The past 25 years have witnessed major changes in the way the issues of economic growth, human development and environmental protection are approached. Two international conferences serve as landmarks. The United Nations Conference on the Human Environment, held at Stockholm in 1972, was the first major discussion of environmental issues at the international level. The agenda was immense, touching on virtually all aspects of natural resource use, but the focus (reflecting the concerns of the developed countries which proposed the Conference) was on the threat to the natural environment posed by economic growth and industrial pollution. Developing countries, for whom these problems were still largely irrelevant, argued that poverty posed a greater threat to both human welfare and the environment, and that economic growth in their case was not the problem but the solution. Stockholm thus marked a polarization between the priorities of economic growth and environmental protection which has dominated the debate between rich and poor countries, and between interest groups within countries, for many years and is still not fully resolved.

2. During the 1980s, a new political and developmental paradigm emerged which appeared to reconcile these conflicting objectives. In 1987, the World Commission on Environment and Development published its report Our Common Future, better known as the Brundtland Report. The report set out the concept of "sustainable development", an integrated approach to policy- and decision-making in which environmental protection and long-term economic growth are seen not as incompatible but as complementary, indeed mutually dependent: solving environmental problems requires resources which only economic growth can provide, while economic growth will falter if human health and natural resources are damaged by environmental degradation.

3. Publication of the Brundtland Report set in motion a process which culminated in the United Nations Conference on Environment and Development (UNCED), held at Rio de Janeiro in 1992. A comparison of the action plans produced by the Stockholm and Rio Conferences illustrates a major shift in our understanding of, and approach to, the problems of long-term human development. Where Stockholm adopted an issue-oriented approach to pollution and non-renewable resource depletion, Rio emphasized integrated strategies to promote human development through economic growth based on sustainable management of the natural resource base. The UNCED action plan, Agenda 21, thus reaffirmed the Brundtland Report's central message: socio-economic development and environmental protection are intimately linked and effective policy-making must tackle them together.

4. This complex agenda, and the message of policy integration, have become widely accepted in the years since Rio, though integration remains a difficult concept to implement. Policy objectives in key areas can and do conflict and means of reconciling different interests and achieving acceptable trade-offs remain underdeveloped. The challenge facing policy makers over the coming years will be to identify critical issues in domestic and international development, and to prioritize measures according to the severity of the problem and the time required for policy to take effect.

Critical trends, past and future ...

5. The present report examines some critical issues of sustainable development, reviewing developments over the past 25 years and looking into the future with the help of model-based projections and scenarios (see box 1). The issues under consideration have been chosen to reflect the widely accepted "pressure-state-response" framework used in much integrated environmental analysis. Population growth (chap. II) in recent decades has increased the overall pressure on the natural resource base. However, greater environmental pressure appears to be exerted by the level and patterns of production and consumption of modern industrial economies. Agenda 21 states that energy and materials use (chap. III) in the developed world is the leading cause of global environmental degradation. The consequences of recent trends in population growth and consumption are potentially most serious when they lead to damage to the natural resources of land (chap. IV) and water (chap. V). Land supports the livelihood of over half the world's population and provides most of the global food supply. Fresh water is indispensable to all life forms and is an essential input to economic activity, especially agriculture. Society's response to trends in these key issues, and their interactions, may be measured in terms of human development (chap. VI) - that is, the extent to which people enjoy adequate income, health, education and other, more intangible, goods such as freedom of choice and personal dignity.

... and the role of policy

6. The focus throughout the report is on the role of policy. To what extent have trends in economic growth, consumption patterns and environmental degradation been influenced by policy intervention? Which policies have proved beneficial and which detrimental? Given the lessons of the past, and the likely shape of things to come (based on the best available projections), what appear to be the most urgent priorities in different regions of the world, and the most promising policy approaches?

7. Any evaluation of trends and the role of policy must take account of the constraints under which Governments operate. The world is becoming increasingly integrated in its commercial and financial activities, while modern means of travel and communications are breaking down many of the traditional physical and cultural barriers between States. National Governments have little control over the forces of globalization and it seems likely that new international bodies, or institutional reforms, will be required to deal with emerging global issues in the future. However, many of the most pressing developmental and environmental problems today are felt at local or national levels. While international action will often be part of the long-term solution, there remains great scope for short- and medium-term policy action at the national level. Accordingly, this report focuses primarily on the potential for effecting change through domestic policy measures.

8. An additional, and powerful, constraint on policy intervention to influence trends is the factor of time: different issues have their natural "pace of change" which becomes apparent when examined over a time-span of 50 years or more. Population programmes, by definition, take at least one generation to take effect. Action to combat pollution can sometimes improve local air or surface-water quality within a few years. Soil degradation and deforestation may be reversed only after decades of sustained effort, while contamination of groundwater reserves might never be corrected, necessitating the use of permanent and costly purification techniques. For these reasons, most problems of environmental degradation require long-term planning and consistent policy-making over many years. By contrast, economic and social problems may sometimes be remedied relatively quickly through targeted policy measures. Additionally, technologies and social behaviour patterns can change rapidly and, while these areas are not readily susceptible to government direction, they represent powerful agents of change and potential cause for optimism.

Global change and the concept of transitions

9. Economic growth, social development and natural resource use are interrelated in ways which, though not fixed, display certain patterns. One way of viewing these linkages over time is through the concept of transitions.

10. A transition may be defined as a gradual, continuous shift in society from one "mode of operation" to another - for example, from an agricultural to an industrial economic base. Four important phases in a transition can usually be described: (i) a pre-development phase of equilibrium and little change; (ii) a take-off phase, which may be hard to initiate through conscious policy effort; (iii) an acceleration phase, characterized by instability due to rapid technological, social and environmental change; societies and environmental quality are highly vulnerable to damage during this stage; and (iv) a stabilization phase, in which the pace of change slows and a new form of equilibrium is reached (the best-known example of this process is the demographic transition, which is summarized in chap. II).

11. This report describes trends in population growth, socio-economic development and the environment as a family of transitions. Clearly there are limitations and shortcomings in this approach but it may represent a promising and useful way of visualizing global change. It is important to note that transitions are not a law of nature; they do not determine what must inevitably happen. Rather, they represent development pathways which have already been experienced by a number of countries and which provide insight into a range of likely futures, dependent on economic, social and environmental circumstances. From the policy maker's point of view, the importance of transitions is that their magnitude, and rate of change, can be significantly influenced by policy intervention.

Taking stock, moving forward

12. Looking back over developments of the past 25 years, one clear lesson to be learned from projections made in the 1970s (most famously, in the 1972 report of the Club of Rome, Limits to Growth) is that dogmatic predictions regarding the Earth's future are unreliable and can be politically counter-productive. With hindsight, world developments since then can be characterized as more complicated, more surprising and, generally, more positive than anticipated. Many problems identified in earlier "doom scenarios" persist but they have not overwhelmed the planet. Some threats - nuclear war, fossil fuel exhaustion - have receded; others - population pressure, industrial pollution - have shown themselves susceptible to determined policy intervention. Less happily, new and unexpected threats have emerged - life-threatening damage to the stratospheric ozone layer, the resurgence of infectious diseases, the rise of AIDS, anticipated changes in the global climate. In sum, while millions of people enjoy lives of safety and comfort unimaginable a few generations ago, hundreds of millions more live in conditions as bad as any endured in the past.

13. The current outlook is uncertain, but prudent policy-making must involve some anticipation of events and the future is not entirely a black box. We have improved our understanding of the interactions between economic, social and environmental systems and have a greater appreciation of uncertainty and risk management. Forecasting is therefore a less deterministic science than 25 years ago. Scenarios no longer predict; they paint pictures of possible futures and explore the different outcomes that might result if basic assumptions are changed, for example, regarding policy interventions.

14. Currently, some trends appear positive: the growth in world population is slowing, food production is still rising, the majority of people are living longer and healthier lives, environmental quality in some regions is improving. But it is impossible to ignore other trends which have the potential to undermine these gains or even bring about catastrophic collapse of local economies. They include the growing scarcity of fresh water, loss of productive agricultural land and the downward spiral of impoverishment affecting a significant minority of the world's population. These threats are real and near-term; they already affect millions of people.

15. The present thus emerges as a time of tension between positive and negative forces - with the balance liable to tip in different directions in different regions of the world. Global catastrophe does not appear to be imminent. But projections cited in this report clearly indicate that pursuit of business-as-usual development patterns is most unlikely to result in sustainable development in the near future. That is, on current trends, we appear unlikely to achieve a desirable balance of economic growth, equitable human development and healthy, productive ecosystems.

16. Economic growth remains the primary policy objective in most countries, on the assumption that other desirable objectives - social and environmental improvement - will follow in its wake. Historically, this has been the case in the industrialized countries - given sufficient time, and making a major exception for their irreversible losses of natural habitat and biodiversity. These historical economic, social and environmental transitions will not be replayed exactly in developing countries. Today, the pace and scale of change are greater, geopolitical, macroeconomic, geographical and cultural circumstances are different. Some developing countries appear to have weathered a half-century of extraordinarily rapid change and emerged with strong economies and thriving populations. The environmental cost has often been high but may be repairable over time. A number of poorer developing countries have yet to undergo the transformations which are essential to function in the modern global economy. More critically, some face the prospect of losing much of their natural capital before alternative sources of employment and wealth generation are sufficiently developed.

17. The following chapters outline past trends and future prospects in some of these critical issues and seek to identify some of the key challenges and options for action which will confront policy makers over the next decade and more. History shows that many negative trends can be reversed given agreement on clear objectives.

Box 1

II. TRENDS IN WORLD POPULATION

Introduction

18. The 1972 Stockholm Conference took place at the end of two decades in which the world's population soared from 2.5 billion to 3.7 billion - the most rapid growth rate in human numbers ever experienced. Population doubling times for some developing countries were falling below 20 years and some scientists and policy makers in developed countries expressed fears that "the population explosion", if not controlled through vigorous population programmes, would lead to mass starvation and societal breakdown. This view was largely rejected by the Governments of developing countries, and the population issue proved too sensitive even to be included in the Stockholm agenda.

19. Subsequent international population conferences (Bucharest 1974, Mexico 1984 and Cairo 1994) charted the gradual emergence of a more consensual approach to the population issue. By the early 1980s, the "North" had largely accepted the "Southern" argument that population policies should be set more broadly in the context of socio-economic development. Concerns over possible environmental limits to population growth (focusing successively on deforestation, energy, water and climate change) were increasingly debated in scientific and policy communities. These linkages between population and environment slowly became more prominent at the international level: Our Common Future (1987) and Agenda 21 (1992) both explicitly discussed population issues in relation to sustainable development.

20. In 25 years, the population debate has evolved dramatically from a narrow focus on population size and growth rates to a more integrated agenda embracing demographic structures, distribution patterns and urbanization, levels of exploitation of natural resources, and creation of viable agricultural and industrial infrastructures. Policy makers increasingly acknowledge these interlinkages between demographics, environment and economy. At the same time, there is still heated controversy over whether the world's population is on an unsustainable trajectory or not; population data invite conflicting interpretations. The following section explores some key aspects of the global demographic transition and their implications for future policy intervention.

The demographic transition

21. The demographic transition is the most widely observed and documented example of the transition concept. Initially, during the pre-development phase, birth and death rates are high and in equilibrium with each other, resulting in slow or no population growth. In the take-off and acceleration phases, the average death rate falls, primarily due to improved health care, the average birth rate remains high and the population grows rapidly. In the stabilization phase, the dominant determinant is a decline in birth rates. In a complete transition, birth rates fall to match the reduction in death rates and a new stable, though much higher, population size is achieved. However, if death rates decrease but birth rates fail to decline to the same extent, the transition "stalls" and total population size continues to increase.

22. As of 1997, the demographic transition has become a historical fact in approximately 30 countries of the world, including all the larger industrialized countries of Europe and Japan. With high life expectancies and growth rates fluctuating around 0.4 per cent a year, their populations are effectively stable. A small number of European countries have negative growth rates and population dynamics in Eastern Europe and the former Soviet Union have been strongly affected by the political and economic transformations experienced since the late 1980s. Fertility rates in the region between 1990 and 1995 were 23 per cent lower than in the previous five-year period. At the same time, mortality rates rose; for example, the average life expectancy of males in the Russian Federation fell sharply, from 65 in 1987 to 57 in 1994.1/ A more varied picture emerges in other regions of the world where population levels are generally rising. However, falling fertility rates are already powerfully evident in Latin America and some parts of Asia. Sub-Saharan Africa has yet to complete the fertility transition, though fertility rates appear to have begun to decline in a number of countries, for example, Madagascar, the United Republic of Tanzania, Namibia, South Africa and Mauritania. For developing countries as a whole, fertility rates are falling rapidly (figure II.1).

23. Overall, the annual growth rate in the world's population has declined from a high of just over 2 per cent in the 1960s to 1.48 per cent in the decade beginning in 1990. The number of people added to the world's population each year (the annual increment) has risen steadily from 47 million in 1950, but is now thought to have peaked at around 81 million in 1995, with population stabilization (an annual increment of 0) expected in 2050. 2/ The population growth rate has fallen more swiftly than demographers expected, because of a faster fertility decline than was previously anticipated. The most recent (1996) United Nations population projections show a significant downward revision of both estimated historical population growth and the size of the world's future population (figure II.2).

24. The combination of declining fertility and accelerating population growth is, in part, the result of previously accumulated "demographic potential", that is, high numbers of young people who are now entering reproductive age. High fertility levels also persist in some parts of the world. Even if fertility rates everywhere were to fall instantly to replacement level (2.06 children per woman), the high proportion of young people "in the pipeline" ensures that world population growth will not be halted for another two generations. 3/

Looking ahead

25. The Conventional Development Scenario (CDS) assumes that developing countries will complete the demographic transition. CDS uses median range projections produced by the United Nations and World Bank which depend primarily on assumed future fertility and mortality rates. By 2050, CDS projects a world population of 10 billion people, with 95 per cent of growth occurring in the developing world. 4/ Total fertility rate in developing countries is expected to reach replacement level in the mid-twenty-first century. The world population is projected to continue increasing slowly to about 11 billion. However, even slight variations in fertility rates could have enormous consequences. The high and low United Nations population estimates differ by 4 billion people, a huge difference, but one that is accounted for by a difference in average fertility rates of just one child per woman (figure II.3).

Urbanization

26. The world has been steadily urbanizing for centuries. Waves of rapid urbanization have followed periods of economic growth in northern Europe and subsequently in the United States of America, Japan and industrializing countries in Asia and Latin America. Almost half the world's population is now urban and by about 2015 the majority - over 5 billion people - will live in urban settlements (see figure II.4).

27. Since 1970, most urban growth has taken place in developing countries, fuelled by both rural-urban migration and natural population increase. While developing countries' rates of urban growth are not unprecedented, their higher population base means that the scale of urbanization in developing countries today often dwarfs the experience of Europe or North America. Approximately 55 million people are now added to the urban population of developing countries every year; since 1970, the number of "million" cities (those with populations between 1 million and 10 million) in Africa, Asia and Latin America has more than doubled. 1/ Even relatively slow rates of urban growth can translate into enormous increases in absolute numbers (see figure II.5).

28. The rural-urban transition in some developing countries has departed from the path followed by the industrialized countries in another important respect: urbanization is occurring even in the absence of broad-based economic growth. The least developed countries are currently experiencing some of the highest urban growth rates; Africa has the highest urban growth rate of all world regions, at over 4 per cent per year.

29. Rapid urbanization in the late twentieth century thus appears to be characteristic both of the acceleration phase of economic transition (characterized by rising income and employment opportunities) and of failure to take off, that is, persistent poverty and social hardship in rural areas. In both cases, urban authorities will face mounting problems in providing adequate infrastructure, especially for impoverished slums and squatter settlements - which are now home to an estimated 25-30 per cent of urban inhabitants in developing countries. 5/

Changing age distribution

30. All current projections agree that an inevitable by-product of the demographic transition (to low birth rates and long life expectancy) is an ageing population. The proportion of the elderly (over 65) in developed countries rose from just nearly 8 per cent in 1950 to over 13 per cent today and, on current trends, will exceed 17 per cent by 2020. 2/ This proportion of elderly people will contribute to a greatly increased dependency ratio (the ratio of the labour force to children and pensioners). As a consequence, difficult policy reforms are already being implemented in developed countries to address the shifting financial dependency burden and the need for changes in infrastructure (for example, more geriatric nursing facilities). Consumption patterns, employment and even cultural attitudes are also likely to be affected.

31. This phenomenon will soon affect much of the developing world, where the number of elderly people is growing twice as fast as the number of people of all ages. If present trends continue, the number of elderly people in the world will exceed the number of the young (below age 15) before the end of the twenty-first century (figure II.6).

32. A comparison of current long-range projections with those undertaken a decade ago reveals some notable differences. CDS yields higher population numbers because it assumes a higher life expectancy; also, the population structure differs because the projected population aged 65 and over is about 25 per cent greater. A critical assumption in generally quoted future population scenarios is that developing countries will complete their demographic transitions. A very different picture emerges if, for example, the transitions in developing countries, especially Africa and South and East Asia, were to stagnate. Projections then show an ongoing increase in world population, reaching about 20 billion people in 2100 (see figure II.7).

The role of policy

33. The demographic transition in industrialized countries has clearly been the result not of policy intervention but of development and modernization, which first brought about falling death rates (through improved education and health care) and then encouraged a change in public preferences regarding family size (as urbanization, increasing economic opportunity and security reduced the need for many children). It has often been pointed out that the demographic transition in many industrialized countries occurred in the context of rich resources and opportunities for expansion and colonization, allowing ample time and space for their populations to stabilize.

34. Given this history, many people in developing countries were at first hostile to calls for population control. However, a marked shift in attitude occurred following the 1974 World Population Conference. The number of developing countries regarding their fertility rates as too high rose from around 40 per cent to 67 per cent in the early 1990s and there was a corresponding rise in the number of Governments prepared to intervene to reduce fertility. By 1993, 53 per cent of developing countries had policies in place aimed at reducing their population growth rates. 6/ This change was promoted, in part, by the difficulties of meeting the economic and social needs of escalating numbers of rural poor and the infrastructural problems presented by rapid, unplanned urbanization. Equally important, socio-economic development - rising levels of education, health programmes and community participation - fostered an environment more favourable to implementation of population programmes.

Steering the demographic transition

35. The causal mechanisms of demographic transitions are well, though not completely, understood. There is a general consensus among demographic experts that they include economic growth and rising per capita incomes, social "modernization" - a complex of factors including education, health care, rural-urban shifts, family structures and employment patterns (especially female labour force participation rates) - and population programmes focusing on family planning. 7/

36. The most rapid fertility declines have occurred in developing countries which have achieved major improvements in child survival rates and educational levels and have implemented family planning programmes. Within groups of countries at similar levels of development, fertility declines have generally been greatest in those with strong family planning programmes. In the absence of such programmes, economic development appears to induce fertility declines first among the more economically advanced population groups and only later across all groups. 8/

Demographic transitions and development pathways

37. The rate and manner in which countries pass through the demographic transition can be strongly influenced by development policies pursued by Governments. Some scholars have distinguished two main routes of development, one based on the introduction of advanced technologies and concentration on an urban, industrialized core, the other based on maximizing productivity of whatever capital is available, usually beginning with the agriculture sector. 9/ In the former case, modernization influences only a minority of the population and their falling fertility levels have a limited impact on the national birth rate. In the second case, broader-based economic development results in a majority of the population sharing in rising levels of income, education and health care, and a more rapid decline in total fertility rate. The speed and character of the demographic transition, in turn, has profound implications for social and economic development and environmental quality.

Policy lessons and priorities

38. There is increasing evidence that successful policy interventions to influence long-term demographic trends involve simultaneous action targeting a range of determinants: income distribution, improvement in the status of women, basic education, information and communication, primary health care (including family planning) and employment opportunities. 10/ A recent study of the links between the educational level of girls and fertility rates in 14 African countries showed that primary education reduced fertility in half the countries, and that secondary education did so in every case. The greatest success in reducing fertility was evident in the countries with the highest levels of female schooling, the lowest child mortality rates and the most vigorous family planning programmes. 11/

39. In the short term, providing contraceptives to close the "fertility gap" between the number of children born and the number of children desired remains a powerful and logical policy priority. The World Fertility Survey indicates that total fertility rates and contraceptive use are closely correlated; in many countries with high fertility rates, contraceptive use remains below 20 per cent (see figure II.8).

40. While recent downward trends in population growth and fertility rates are very encouraging, they will require additional policy efforts if they are to be sustained. As more women enter their reproductive years (a consequence of "demographic potential"), the need for family planning services will increase rapidly. During the 1990s, approximately 100 million more couples will need family planning services just to maintain current rates of contraceptive use. In order to realize the declining fertility rates projected in the United Nations medium population variant, about 75 million more couples will need access to family planning by the year 2000.

41. Regional differences in fertility rates, and the provision of family planning, should be a particular cause for concern, given that the highest population growth rates are, generally, occurring in the poorest and environmentally most fragile parts of the world.

Notes and references

1/ World Resources Institute, World Resources Report, 1996-97 (New York and Oxford, Oxford University Press, 1996).

2/ World Population Prospects: 1996 Revision (United Nations publication, forthcoming), annex I, Demographic indicators.

3/ According to the United Nations instant replacement fertility projection, where the total fertility rate is assumed to stabilize immediately at replacement level (2.06 children per woman), the world's population would still increase to 7.1 billion by 2025. See Long-range World Population Projections: Two Centuries of Population Growth, 1950-2150 (United Nations publication, Sales No. E.92.XIII.3).

4/ The most recent medium projection produced by the United Nations estimates a lower population of 9.4 billion by 2050. See World Population Prospects ....

5/ Report of the Secretary-General on the fourth review and appraisal of the World Population Plan of Action (A/CONF.171/PC/3).

6/ United Nations Population Policy Data Bank, cited in World Resources Report, 1996-97 ....

7/ According to a World Bank study of 64 countries, when the income of the poor rises by 1 per cent, general fertility rates drop by 3 per cent. However, this conclusion should be qualified by recognition of the great social and cultural heterogeneity among the poor in different parts of the world, which is highly relevant to the way in which fertility reacts to improving living conditions.

8/ Fertility Behaviour in the Context of Development: Evidence from the World Fertility Survey (United Nations publication, Sales No. E.86.XIII.5).

9/ H. Oshima, "Impacts of economic development on labor markets, education and population in Asia", Ambio 21, 1992.

10/ Review and Appraisal of the World Population Plan of Action: 1984 Report (United Nations publication, Sales No. E.86.XIII.2), chap. I; World Population Prospects ....

11/ United Nations Development Programme, Human Development Report 1996 (New York and Oxford, Oxford University Press, 1996).

III. ENERGY AND MATERIALS CONSUMPTION

Introduction

42. Energy and raw materials are fundamental to economic activity and human well-being. As the size of the world economy has grown (approximately fivefold since the Second World War), so resource consumption has accelerated at a rate unprecedented in human history. During the 1960s, energy and steel use in the industrialized countries increased at rates that would have doubled total consumption by 1987 and quadrupled it by 2000. This pace and scale of growth gave rise to fears that such exploitation rates of non-renewable resources could not be sustained. During the 1970s, predictions abounded that the world would shortly "run out" of fossil fuels and other essential raw materials.

43. Since then, however, rising demand has generally been matched by discoveries of new reserves and substitution between resources, in response to the operation of market forces and technological advance. Concerns have shifted away from resource depletion to a wider and more complicated package of issues relating to the provision of adequate energy supplies in developing countries and the environmental and health impacts resulting from conventional patterns of energy and materials use.

44. The Brundtland Report of 1987 pointed to the tensions between increased levels of energy and material use required for economic growth in developing countries and the environmental (and hence financial and social) costs that could be expected from business-as-usual growth in the developed and developing world. It called for a major reorientation of policies towards efficient technologies and conservation efforts but anticipated that even this solution would not prevent increased levels of global and regional environmental degradation. The challenge today remains the same: to meet an enormous projected global demand for resources, especially energy, through the use of new products, technologies and consumption patterns which will provide rising standards of living for all while minimizing economic costs and risks to human health and the environment.

Transitions in resource use

45. Economic growth and social modernization in the industrialized countries has been characterized by major changes in resource use. Energy and material consumption rose rapidly during the nineteenth and first half of the twentieth centuries to meet demands from intensified agriculture, construction and industrialization. This phase was followed by a decrease in the growth rate of resource use as economies matured and became more efficient. Over the same period, primary energy sources shifted from traditional renewables (wood, other biomass) to fossil fuels, with an increasing dependence on electricity as the end uses of energy diversified. Materials use is still changing, from an emphasis on heavy, bulk commodities to lighter, higher-value metals and composites. Some higher-income developing countries are now undergoing elements of this transition, but significant differences between the industrialized and developing countries are evident in both levels and patterns of resource consumption.

Trends in resource consumption ...

46. World commercial energy use accelerated sharply after 1950, growing at an average annual rate of 5 per cent till 1970, when the growth rate slowed. By 1993, world energy consumption was almost 50 per cent greater than in 1973. (figure III.1) 1/ The industrialized countries still account for over 60 per cent of total consumption, though their share is declining as the rest of the world develops. Developing countries experienced exceptionally fast growth rates during the 1960s and 1970s as a result of economic development and the rapid replacement of traditional by commercial (fossil) energy sources. However, this growth took place from a very low consumption base in absolute terms. This fact, together with high population growth, means that per capita consumption in developing countries remains very low by the standards of industrialized countries (see chap. VI, table). 47. World demand for metals and minerals rose by 120 per cent between 1961 and 1990. Growth rates were highest among low- and middle-income countries, where infrastructure and industries are rapidly being established. Demand in the mature economies of the Organisation for Economic Cooperation and Development (OECD) has slowed considerably since the 1970s; global demand growth rates have slowed from about 6 per cent in the 1960s to under 2 per cent in the 1990s, though this represents a considerable increase in absolute terms (figure III.2).

48. Despite rising levels of energy and material consumption, there is no short-term prospect of scarcity. Estimates of world energy reserves have increased significantly over the past 20 years and energy prices in recent years have remained low, indicating no perceived or anticipated scarcities in the near future. Proven reserves of the most important metals and minerals have also risen since 1970. Consumption as a proportion of reserves has declined and long-term prices for most raw materials have trended steadily downward. 2/ Concerns in recent years have turned from exhaustion of non-renewable resources to the degradation of renewable resources - soil, water, forests - and risks to human health which follow from current resource use patterns.

... and their social and environmental impacts

49. Despite the massive investments in energy supply over the past 30 years, over 2 billion people, mostly in rural areas of developing countries, still have little access to commercial energy supplies. Poverty, remote location, or both, leave communities dependent on animal or human energy for labour and fuelwood or animal dung for cooking and heating. 3/ Wood still provides up to 50 per cent of national energy needs in a number of Asian and sub-Saharan countries. 4/

50. Lack of access to commercial energy is a severe constraint on social and economic development. The lowest-income countries are also those with lowest per capita energy consumption; social indicators such as literacy rate, infant mortality, life expectancy and total fertility rate all improve sharply with increased per capita income and energy consumption. 5/ In addition, the use of traditional fuels is now understood to be damaging to human health; many studies document a correlation between indoor air pollution and the incidence of respiratory ailments and congestive heart failure. 6/

51. Population growth over recent decades has increased the demand for traditional fuels, especially in poor regions. Historical data on consumption of fuelwood are scarce but it is estimated that today some 1.8 billion m3, more than half of the total volume of wood produced, is consumed as fuel. 7/ Fuelwood consumption is leading in some areas to deforestation, soil impoverishment and further hardship for the poor, who are dependent on the natural resource base for their day-to-day survival.

52. Quite different problems have been created by the intensive use of fossil fuels in the developed countries. Poor urban air quality and transboundary air pollution from acidifying compounds ("acid rain") have been addressed with some success in the past 30 years. But the threat of global warming, caused by rising concentrations of greenhouse gases, has emerged more recently and has yet to be tackled effectively. Atmospheric concentrations of carbon dioxide (CO2), a leading greenhouse gas, have increased steadily since the industrial age began (figure III.3). Evidence is strong that most of the increase is attributable to human activity, in particular, to fossil fuel combustion. 8/ Continuation of these emission trends poses serious, though uncertain, risks of global and regional climate change, leading to unpredictable rises in sea level, inundation of low-lying coastal areas, migration of ecosystems and disruption of agriculture.

53. The industrialized countries today account for about 70 per cent of carbon dioxide emissions; historically, about 84 per cent of fossil energy carbon dioxide emitted since 1800, which still remains in the atmosphere, is the result of past emissions in the industrialized countries. 1/

54. Atmospheric pollution from fossil fuel combustion has also grown rapidly in developing countries, however. Current per capita primary energy use in some of the higher-income economies of Asia already exceeds that of some poorer OECD countries. Industrial and transport-related emissions are seriously undermining human health in many developing country cities; evidence is also accumulating that acid precipitation is reducing yields of some agricultural crops. Transport is now the fastest growing end-use of energy in developing countries; the sector grew at an average annual rate of nearly 6 per cent between 1970 and 1990, compared with just over 2 per cent in the developed countries. 1/

The role of technology

55. Technologies enable humans to expand their range of activities and transform the Earth's resources. Technological advances, in principle, enable more productive use of resources, thereby delivering equivalent or improved services while greatly reducing health and environmental burdens. Two long-term trends offer particular hope for mitigation of these adverse impacts: improvements in energy and material efficiency and decarbonization of energy sources.

Energy and materials intensity

56. Resource intensity (the energy and materials required for constant economic output) is declining in the industrialized countries. Energy intensity has fallen by about 1 per cent per year since 1800; it declined even faster during the 1970s and 1980s, at about 2 per cent per year, but has barely fallen since 1990. 9/ Materials intensity has fallen rapidly, at nearly 2 per cent per year since 1971. These improvements are due to more efficient technologies, structural economic changes (for example, shifts from resource-intensive industries towards light manufacturing and services) and technical advances which reduce the resource inputs required to manufacture a given product (energy efficiency and "dematerialization").

57. The phenomenon of partial "delinking", whereby growth in gross domestic product (GDP) is achieved with slower growth in energy consumption, is most advanced in the developed countries; resource efficiency improvements are also evident in the newly industrializing economies. Energy efficiency is now improving in some low-income countries but their material intensities are still high, reflecting the ongoing development of economic infrastructure (figure III.4).

Cleaner energy and materials

58. At the global level, decarbonization (the decreasing ratio of average carbon emissions per unit of primary energy) has occurred at a slow rate of about 0.3 per cent per year. The energy mix in industrialized countries has gradually shifted from high-carbon content fuels like coal to lower-carbon fuels such as oil and natural gas and carbon-free hydro and nuclear power (figure III.5). The carbon intensity of the major coal-dependent Asian economies is currently high - comparable to the industrialized countries in the nineteenth century - but the evidence so far indicates that once the trend to decarbonization begins countries advance at roughly comparable rates.

59. Information on material flows is weak for most countries but it appears that economic development is characterized by a decline in the intensity of use of heavy commodities - lumber, concrete, lead - and a shift towards "value added", sophisticated materials such as aluminium, plastics and composites. The number of materials in circulation has multiplied dramatically with technological progress and rising consumer demand for new products and services; for example, an estimated 90,000 chemicals are now in commercial use. It is the volume and heterogeneity of materials consumed, and the uncertainty surrounding their potential health and environmental impacts (rather than rates of use and possible depletion), that have now emerged as key policy issues.

Looking ahead

60. A principal anxiety of the 1970s, the exhaustion of non-renewable resources, has receded, if not permanently, then to a more distant future. The capacity of technology and markets to improve resource efficiency, substitute energy sources and materials and side-step perceived problems has been a recurring surprise, and forecasts of future demand and supply now allow for a wider range of outcomes. This section focuses on energy: the "energy challenge" over the next half century involves providing a growing world population with sufficient energy, without further damage to human health or disruption of critical environmental functions.

61. The Conventional Development Scenario (CDS) assumes that energy demand and supply will follow historical growth paths, which are mainly driven by demographic, economic and behavioural (cultural) determinants. The scenario assumes that (i) consumption continues to rise in the household and commercial, industry and transportation sectors, with demand rising fastest in the developing countries; (ii) energy intensities in all world regions decline, though not sufficiently to offset rising demand; and (iii) the global energy mix continues to be dominated by fossil fuels.

62. The CDS forecast may be unduly pessimistic. A "Middle Course" scenario developed by the World Energy Council (WEC) and the International Institute for Applied Systems Analysis (IIASA) 10/ assumes modest estimates of economic growth and technological development and greater improvements in energy intensity, which lead to lower future energy demand. Fossil fuels still dominate the primary energy mix but a gradual transition to renewable energy is seen as feasible after 2020 (figure III.6).

63. The Middle Course scenario could still could be improved upon. The "Low CO2-Emitting Energy Supply System" (LESS) scenario, developed for the Intergovernmental Panel on Climate Change (IPCC) 11/ assumes significant improvements in energy efficiency which cause primary energy consumption to rise much more slowly than GDP. The scenario also posits shifts in the fuel mix in favour of renewable energy sources, primarily modern biomass.

64. The composition of the global energy mix assumed in the CDS, Middle Course and LESS scenarios are shown in figure III.7, and the levels of CO2 emissions that would result from each of them are shown in figure III.8.

65. More pessimistic scenarios are also possible. Most conventional projections assume that developing countries will follow an energy development path similar to that experienced in developed countries. In developing countries, however, the energy transition from traditional to modern fuels may slow down or even stagnate if coal reserves continue to prove more economically attractive (in industrializing regions) or socio-economic development is insufficient to generate the income necessary to buy commercial fuel supplies.

66. Such a situation (a stagnating energy transition and continued poverty) would imply that traditional biomass fuels continue to play an important role in many developing regions, in particular in rural and low-income urban areas. Due to increasing population, over-exploitation of scarce traditional fuels, especially fuelwood and dung, would increase local and regional environmental stress and adversely affect food production (chap. IV). Health problems from indoor combustion of such fuels would continue.

67. Energy and materials are essential to development and all scenarios agree that consumption will increase substantially to meet projected demand, especially in developing countries where per capita consumption is currently too low to support even basic living standards. The scenarios suggest that meeting future demand could be achieved with greatly reduced social and environmental harm if the historical transitions from traditional to commercial fuels, towards improved efficiency (declining intensity of use) and towards safer sources (low-carbon energy and non-hazardous materials) could be greatly accelerated.

The role of policy

68. Major efforts have been made to improve access to commercial energy. Public expenditures on power sector infrastructure in developing countries rose from about 0.6 per cent of GDP in the 1960s to over 1.7 per cent in the 1980s. However, a key constraint to future energy development in developing countries is lack of capital. The World Energy Council has estimated that energy investment requirements in developing countries between 1990 and 2020 will be US$ 3 to US$ 7 trillion (in 1990 dollars) if the conventional energy transition is pursued. 12/ The costs of extending electricity grids to meet the needs of the rural poor in developing countries are beyond all foreseeable domestic budgets and development aid.

69. In order to improve competitiveness and reduce dependence on energy imports, many industrialized countries have deregulated domestic energy markets and introduced incentives for energy saving. Since the late 1980s, fossil fuel prices in OECD countries - except the United States - have been higher than market clearing levels. Efficiency standards established for cars, buildings and a range of consumer appliances in developed and developing countries have also produced measurable energy savings. By contrast, social and political considerations have led to energy being heavily subsidized in developing countries and, until the 1990s, in the countries of Eastern Europe and the former Soviet Union. In 1992, government subsidies worldwide for conventional energy amounted to over US$ 200 billion, more than official development assistance from all sources (taking debt repayment into account). 3/ Artificially low energy prices and restrictions on market-based competition in energy markets have encouraged inefficiency and wasteful use patterns and have discouraged energy conservation measures.

70. Policy measures aimed at encouraging materials efficiency have been stimulated in developed countries by the sheer volume of materials used and the economic, environmental and political problems associated with disposal. A new agenda of "eco-efficiency" is developing, focused on delivering equivalent or higher standards of goods and services with greatly reduced material throughput. Measures include material and product taxes, recycling targets, mandatory producer take-back requirements and consumer information schemes. However, it remains too early for significant changes to be observed at the national level.

71. The trend towards lower-carbon fuels has, on balance, been favoured by market forces, though the rapid penetration of the energy market by (carbon-free) nuclear power has been due almost entirely to government support. The transition is less clear in some developing countries where, for example, rapidly rising demand for electricity and transportation has encouraged high consumption of oil alongside continued use of traditional renewables. Renewable energy sources, such as wind and solar power and hydrogen fuel cells, remain under-exploited in both developed and developing countries. Uncompetitive costs, relative inefficiency and the problem of variable supply remain barriers to their more widespread use. This, in part, reflects prevailing R&D spending priorities. The developed countries currently spend over 50 per cent of their annual US$ 8 billion energy research budgets on (civil and military) nuclear programmes; less than 10 per cent is spent on renewables. 13/

Policy lessons and priorities

72. Market liberalization and privatization in the energy sectors of developing countries may be enough to meet the needs of enterprises but they have done little to improve energy services in poor rural areas. The most urgent situations are in low-income developing countries which are dependent on imports for their commercial energy supplies and where shortages of fuelwood are becoming acute. Ensuring adequate, affordable and secure energy supplies which sustain development in the short term will require accelerated development of indigenous energy resources, for example through tree planting. In the longer term, increased capacity for producing commercial primary energy - both fossil and renewable - is essential.

73. Energy efficiency will be encouraged, over the long term, in signatory countries to the United Nations Framework Convention on Climate Change, which requires reductions in national emissions of carbon dioxide. However, energy-related CO2 emissions in OECD countries rose by about 4 per cent between 1990 and 1995, and very few countries are currently on target to meet their reduction goals. 14/ The most potent policy instrument - "environmental" energy taxes - remains politically difficult to implement. However, some developed countries are exploring the idea of "ecological tax reform" in which increased taxation on fossil fuels (or pollution) is offset by reduced taxes on, for example, labour. This approach is reportedly meeting with a more favourable reception from energy users.

74. In developing countries, it is clear that, whatever the promise of energy-efficient technologies and advanced materials science, significant increases in commercial energy consumption will be essential for real economic growth and social development. Nevertheless, improved efficiency is receiving increased attention from many policy makers and energy providers under pressure from budgetary constraints. The marginal costs of meeting at least some of the existing and projected demand through efficiency savings and demand management approaches are estimated to be far lower than providing additional capacity by extending electricity grids. However, this approach is likely to make little headway as long as artificial incentives for energy use persist. 75. The switch to cleaner fuels in the industrialized countries has been encouraged by international agreements on air quality such as the Vienna Treaty protocols setting reduction targets for SOx and NOx. There is a clear need for similar regional agreements between industrializing countries, which are now beginning to suffer the full effects of transboundary pollution.

76. Policy measures targeting specific hazardous emissions or substances have proved highly effective, where evidence of damage to human health or the environment is strong and substitutes exist. For example, reductions in lead emissions in North America were due almost entirely to the legislated phase-out of leaded gasoline (figure III.9).

77. In conclusion, structural changes in the global energy system are relatively slow, reflecting the long lifetime of facilities and infrastructure. On average, major transitions (for example, the replacement of coal by crude oil) take about 50 years. The total capital stock of the energy sector can be expected to be replaced at least twice by the end of the next century, offering numerous opportunities for efficiency improvements and shifts away from carbon-intensive fuels. Changes in materials use may be much faster, offering even greater opportunities for policy intervention. Major historical transitions in resource use to date have been largely the result not of policy intervention but of technological advance responding to market forces. Judging by recent experience, price rises are not likely to occur soon enough, or with the necessary consistency, to deliver a smooth transition. This underlines the fact that policies affecting the evolution of energy supply and demand will be of crucial importance.

Notes and references

1/ Nebojŝa Nakiŝenoviŝ and Arnulf Gru"bler, "Energy and the protection of the atmosphere", paper prepared for the Department for Policy Coordination and Sustainable Development of the United Nations Secretariat, February 1996.

2/ World Bank, World Development Report 1992: Development and the Environment (Washington, D.C., 1992).

3/ United Nations Development Programme, "UNDP initiative for sustainable energy", June 1996.

4/ Department for Policy Coordination and Sustainable Development of the United Nations Secretariat, based on 1992 Energy Balances and Electricity Profiles (United Nations publication, Sales No. E/F.94.XVII.14).

5/ See, for example, Fundacio'n Bariloche, Catastrophe or New Society: A Latin American World Model, IDRC-064e (Ottawa, International Development Research Centre, 1976). Cited in "UNDP initiative for sustainable energy", June 1996.

6/ See, for example, B. H. Chen and others, "Indoor air pollution in developing countries", World Health Statistics Quarterly, 1990, vol. 43, No. 3.

7/ European Forest Institute and Norwegian Forest Research Institute, "Long-term trends and prospects in world supply and demand for wood and implications for sustainable forest management", report prepared for the Ad Hoc Intergovernmental Panel on Forests of the United Nations Commission on Sustainable Development, July 1996.

8/ The body of statistical evidence now points towards a discernible human influence on global climate. (See Second Assessment Report of the Intergovernmental Panel on Climate Change (IPCC), "Climate change 1995: the science of climate change", 1996.

9/ Andrew Glyn, "Northern growth and environmental constraints", in V. Bhaskar and Andrew Glyn, eds., The North the South: Ecological Constraints and the Global Economy (London, Earthscan Publications Ltd., 1995).

10/ World Energy Council and International Institute for Applied Systems Analysis, Global Energy Perspectives to 2050 and Beyond (1995).

11/ R. H. Williams, "Variants of a Low CO2-Emitting Energy Supply System (LESS) for the world", report prepared for the IPCC Second Assessment Report, Working Group IIa, Energy Supply Mitigation Options, 1995.

12/ A more recent (1995) study by the International Institute for Applied Systems Analysis and the World Energy Council revises these requirements upwards to between US$ 6 and US$ 9 trillion by 2020 and between US$ 11 and US$ 18 trillion by 2050.

13/ International Energy Agency, Annual Report, 1995.

14/ World Energy Council, Climate Change Negotiations: COP-2 and Beyond, Report No. 6, September 1996.

IV. AGRICULTURE AND FOOD SUPPLY Introduction

78. The Malthusian spectre of mass starvation was revived in the late 1960s and early 1970s when rapid population growth was seen by many environmentalists as a primary cause of the mass famines experienced in Asia and Africa. The 1974 World Food Conference was held at the height of a perceived world food crisis, the result of poor harvests in the world's major food producing regions and soaring food commodity prices, aggravated by the 1973 oil shock. The Conference established the goal that hunger, food insecurity and malnutrition should be eradicated within a decade. Raising food production became an international policy priority for the remainder of the decade and investment in agricultural research and development increased significantly.

79. Over the next decade, world food production grew rapidly and a decline in the incidence of mass famines generated optimism over the ability of farmers to feed ever-growing numbers of people. Rising global harvests, food surpluses in industrialized countries and steadily falling prices for most food commodities encouraged a perception among many donor Governments that food supplies were secure for the foreseeable future, despite the fact that many millions of people continued to go hungry. During the 1980s, the intensity of efforts to boost production declined and concerns shifted to the environmental and social consequences of intensified farming methods, especially land degradation, water pollution, rural unemployment and health impacts associated with pesticide use.

80. More recently, a number of factors have combined to revive doubts about food security in the longer term. The growth rate in world harvests has slowed markedly, many marine fish stocks have become depleted, population growth continues to be most rapid in those areas which are already food insecure - and land degradation is now measurably undermining the agricultural resource base. In late 1996, the World Food Summit was convened to examine these developments and called for a halving of the numbers of hungry people by 2015. Policy approaches recommended at the Summit demonstrate a new understanding that food security must be addressed within the wider context of poverty eradication, economic development and environmental sustainability, as well as improved agricultural technology. 1/

The agricultural transition

81. The past two centuries have witnessed a worldwide, but still incomplete, shift from low-input, low-output agriculture to more intensive farming systems. In extensive agricultural systems, production tends to be raised by cultivating additional land as necessary. The "take off" phase leads to raised productivity (yields per hectare), achieved through the use of new crop varieties, improved cultivation techniques and increased inputs of agrochemicals and fossil fuel energy. The mature phase of the transition is characterized by slower yield growth rates; productivity may even decline locally, if the adoption of more intensive techniques has resulted in damage to soil and water resources.

Food supply

82. World agricultural production has risen impressively over the past 40 years (figure IV.1). Since 1961, the industrialized countries have raised cereal production by an average of 1.7 per cent each year. The annual rate of increase in developing countries averaged 3 per cent and they now account for well over half the world cereal harvest. 2/

83. The Green Revolution of the 1950s and 1960s enabled some developing countries, notably India and China, to increase their food production dramatically through the use of irrigation, fertilizer and scientifically bred new strains of rice and wheat. Other developing regions, including North Africa, much of South America and South Asia, achieved major though less spectacular increases. Sub-Saharan Africa so far has continued to raise food production more through the cultivation of new land than through intensified production. Fertilizer application rates and average yields in much of the region remain well below levels elsewhere.

84. Marine and inland fisheries are also a vital part of the world's food supply. Approximately 950 million people, mostly in developing countries, rely on fish for their primary source of protein. The marine catch (representing about 78 per cent of the total) rose nearly fivefold between 1950 and 1989; it has dropped slightly since then but the total catch has continued to rise thanks to expanded aquaculture production. This steady increase masks a more complicated picture in which new species of fish, and new fishing grounds, have been successively exploited and depleted. A recent study indicates that an increasing proportion of the world's major ocean fisheries are now at, or nearing, the point of over-exploitation (figure IV.2). 3/ The greatest potential for raising future fish harvests appears to lie in conservation of depleted stocks, improved management of rich fishing grounds such as the Indian Ocean, and in continued expansion of aquaculture which already contributes nearly one quarter of the total fish supply in Asia. 4/

85. The overall growth in world food supply conceals a number of trends which, if not corrected, will have disturbing implications for future food security.

86. Agricultural intensification in recent decades has taken a heavy toll on the environment. Poor cultivation and irrigation techniques and excessive use of pesticides and herbicides have led to widespread soil degradation and water contamination. Approximately 300 million hectares worldwide are now severely degraded and local farming systems abandoned; a further 1.2 billion hectares - 10 per cent of the earth's vegetated surface - are at least moderately degraded. 5/

87. Competition for land is increasing. Development and population growth claim land for housing, industry and infrastructure. Worldwide data are not available, but economic growth in Asia is estimated to have reduced cropland area in some countries by 1 per cent per year; 6/ slower rates of land loss are still under way in the industrialized countries.

88. Dietary preferences are changing with increasing wealth in favour of meat and dairy products. Direct consumption of grain by humans is the most efficient use of available food supplies but more land in developing countries is now used for growing grain feed, fodder and forage for livestock in order to supply export feed to industrialized countries and to meet an increasing demand for meat and dairy products in developing countries.

89. Agriculture in parts of North America and Europe has reached a stage where public interest in "healthy food", recreational use of the countryside (made possible by mass ownership of private vehicles) and protection of wildlife and habitats are encouraging less intensive techniques. Policy incentives are being implemented to remove some cropland from production altogether.

90. The net effect of these trends (a static or declining productive land base and less efficient use of cereals) has been offset for many years by increasing yields. However, while yields and total world food production are still rising in absolute terms, the rates of growth have been declining since the early 1980s. 7/ This situation, combined with rising population, means that per capita food production has grown relatively slowly (figure IV.3). Since 1984, the per capita grain harvest has actually declined by an average of more than 1 per cent per year (figure IV.3).

Hunger amidst plenty

91. Globally, there is still abundant food. Between 1961 and 1994, per capita food supply in developing countries increased by 32 per cent and mass famines on the scale seen in the nineteenth and mid-twentieth century have not recurred. The proportion of the world's population who are hungry or chronically undernourished has fallen from 35 to 21 per cent. Rising food production has not delivered universal food security, however. Nearly 840 million people remain hungry worldwide and the number of chronically undernourished in sub-Saharan Africa has more than doubled since 1969 (figure IV.5). 8/ 92. During the 1970s, food availability increased significantly in many countries in East and West Asia, Latin America and North Africa, despite rapidly growing populations. These gains were due not only to improved agricultural productivity but also to economic development (or in some cases borrowing), which enabled rapid growth in food imports. Net imports of cereals in developing countries more than tripled between 1969/71 and 1979/81. 9/ But in South Asia, per capita food supplies stagnated and they declined in sub-Saharan Africa, where domestic production and imports could not match population growth.

93. The 1995 aggregate food security index, compiled by the Committee on Food Security of the Food and Agriculture Organization of the United Nations (FAO), indicates that a number of countries in sub-Saharan Africa and South Asia and a limited number in Latin America and the Caribbean have experienced serious declines in per capita food availability since the mid-1980s. 10/ These countries, with "low" or "critical" food security, have made little progress in raising per capita food production, or have proved unable to maintain earlier gains because of economic or political instability. It is generally agreed that food insecurity at present is largely a distributional problem; people have inadequate physical and/or economic access to food as a result of poverty, political instability, economic inefficiency and social inequity.

Food for the future

94. The challenge in providing adequate nutrition to a population of 9.4 billion in 2050 is threefold: (i) food production must be doubled; given current regional population growth trends, a nearly threefold increase in supply will be required to feed the developing world and a fivefold increase for Africa and the Middle East; (ii) regional self-sufficiency will not be possible everywhere and food imports must increase; since food supply traditionally responds to demand (which is a function of income) and not to human need, the focus must be on ensuring that people have income to purchase the food they need; (iii) increased production must be achieved without further damage to the productive base (soils and water) or to human health.

95. Projections of future global food production and food security depend on underlying assumptions regarding population growth, available cropland, yield gains and diet (lifestyles). Recent long-term forecasts involve a wide range of outcomes because even small changes in underlying assumptions can lead to huge differences in future agricultural supply and demand (figure IV.4).

96. The Conventional Development Scenario assumes a global growth in food production of about 1.5 per cent per year, falling to around 1 per cent, resulting in doubled production by 2050. The potential for cropland expansion appears greatest in Africa and Latin America, but limited in developing Asia; cropland areas could decrease in industrialized regions. Irrigated areas are likely to increase only modestly. The greatest productivity gains are likely to come in Africa, where yields are currently well below average; yield gains could also be strong in China and Latin America. In industrialized countries, yield increases are expected to continue, though at a lower rate.

97. The International Food Policy Research Institute (IFPRI) has developed a number of scenarios of future food production, including two variants based on differing assumptions regarding investment in agricultural research. 11/ Further cutbacks in public investment are expected to reduce production by about 10 per cent in developing countries (relative to a baseline scenario). Higher investment, by contrast, is projected to raise production and to create additional benefits in non-agricultural income growth, to increase public expenditures on social services and to improve access to water, sanitation services and education.

98. FAO has produced a forecast for 2010 which also assumes rapid technological improvement and projects a growth rate of 1.8 per cent per year. 12/

Food production and the natural resource base

99. Limiting factors in expanding cultivated land area include the scarcity of high quality agricultural land, the risk of environmental degradation of marginal cultivated lands and competition from alternative land uses, particularly urban growth and development in developing countries and "counter-urbanization" (urban-rural migration), tourism and leisure uses in developed countries. Competition for land between agriculture and forest cover will sharpen; recent estimates suggest that nearly two thirds of tropical deforestation - some 12 million hectares per year - is due to farmers clearing land for agriculture. 13/ Scenarios developed by the Finnish Forest Research Institute suggest that tropical forest cover could decline from 1,757 million hectares (1990) to between 1,164 million hectares and 1,360 million hectares in 2025. 14/

100. While significant areas of Africa and Latin America remain to be opened up for agriculture, the costs to indigenous forest dwellers, forest and savannah vegetation and biological diversity are likely to be high. Additionally, the degradation and deforestation of watershed forests have an important impact on water quality, quantity and periodicity of flow (see chap. V, section on water and ecosystem functions). These facts suggest the need for a twin focus on increasing food production in more fertile areas (intensification) thereby reducing the need for further expansion into marginal and easily degraded non-agricultural lands, and developing technologies appropriate to less fertile areas which can improve farmers' opportunities to increase food production while minimizing the risk of environmental damage.

101. Water is likely to prove the most important constraint on raising food production (chap. V). Irrigated cropland area grew by 2-4 per cent annually during the 1960s and 1970s but expansion has since slowed to less than 1 per cent per year. 15/ This trend is not expected to be reversed in the short term, given low projected world food prices and the increasing economic costs and environmental and social impacts associated with major new irrigation schemes. Yields from existing irrigated land, especially in parts of China, India, Pakistan and the United States, are also expected to decline where aquifers become exhausted and/or urban populations and industry claim a growing share of limited water resources. Nevertheless, more than half of additional crop production by 2010 is expected to come from irrigated land, according to FAO projections. This implies the need for vigorous, renewed investment in small-scale irrigation schemes which can raise yields while minimizing economic costs and environmental damage. 16/

Agricultural productivity and food security

102. The CDS, IFPRI and FAO forecasts all suggest that future production gains will be sufficient to cope with increased population and rising demand at the global level. At the regional level, however, they all indicate worsening food insecurity in sub-Saharan Africa and only marginal improvement in South Asia, even under a variety of assumptions about future growth, investment and trade liberalization. It seems unavoidable that food security in some developing countries will (as now) depend on imports from food surplus to food deficit countries. On the basis of current policies, nearly 700 million people are expected to go hungry in 2010 (figure IV.5).

103. In theory, there is scope for further immense yield gains and raised production in the future. Wide disparities still exist between the yields achieved in agricultural research stations and by farmers. Typically, dryland farmers in Asia, Latin America and Africa obtain between one tenth and two thirds of research station yields; most farmers achieve less than half. In addition to yield gaps in improved crops, many plant and animal breeds have not yet been subject to much scientific improvement. However, a major challenge will be to ensure that future improvements do not exclude tropical farmers - where potential production gains are high and food insecurity is often greatest.

104. The CDS and IFPRI scenarios represent a "mid-way" point between technologically based projections which recognize few limits to food production and more pessimistic forecasts founded on the belief that world food production is already on a declining trajectory. If historical improvements in agricultural productivity can be maintained, there will be enough food to meet expanding demand. Future improvements will depend critically on continuing investment in improved crop varieties, greater efficiency in water use, improved soil management and socio-economic development which enables farmers to take advantage of new techniques.

The role of policy

105. Agriculture is a heavily regulated economic sector. National Governments in both developed and developing countries intervene to influence what is grown and sold, where and how. Trade policies determine market access and can be used to protect the interests of domestic producers. The potential for government action to improve, or worsen, food security at every level is therefore enormous.

Positive impacts ...

106. Countries which have followed broad-based economic development strategies including high levels of investment in agriculture have often experienced a number of positive benefits: increased employment and income through agriculturally led economic growth, decreased incidence of rural poverty and greatly improved food security. The World Bank has noted the clear relationship between rising gross national product (GNP) and declining prevalence of underweight children between the 1970s and 1980s; the relationship is especially strong in countries that have done well in agriculture, such as Indonesia and Thailand. 17/

107. More specifically, the extraordinary yield increases of the Green Revolution were made possible by a technological package (comprising genetically improved plant varieties, irrigation, fertilizer and pesticide application and management skills) developed by scientists working in State and donor-sponsored research institutes and field stations.

108. Public demand for nature conservation, together with budgetary constraints, have encouraged "environment friendly" policy measures to reduce agricultural surpluses and improve environmental protection in some developed countries. Strict regulation and pricing incentives have been introduced in parts of Scandinavia and Northern Europe, for example, which limit the use of agrochemicals and encourage preservation of habitat "enclaves" in farmland.

... and negative impacts

109. It has been argued that market failures in the agriculture sector and government interventions which distort product and input prices (and tend to reinforce market failure) have a greater effect on soil, water, human health and ecosystems than equivalent failures in any other sector. 18/ Examples abound in both developed and developing countries of tax incentives which have encouraged inappropriate changes in land use. Subsidies in many developing countries for pesticides and herbicides, combined with inadequate education and product labelling, have encouraged excessive use, leading to soil and water contamination and thousands of pesticide-related deaths per year. The Common Agricultural Policy (CAP) of the European Union at one time involved a system of guaranteed prices for unlimited production. These guarantees improved regional self-sufficiency and turned Europe from a net importer to a net exporter of cereals. At the same time, they encouraged intensive farming methods which have led to soil and water pollution, landscape degradation and loss of biodiversity.

110. At the global level, agricultural production and trade patterns have been seriously distorted by national and international policy responses to the macroeconomic events of the 1970s and 1980s: the oil shocks, the boom and bust cycle in food commodity prices, developing country indebtedness and world recession. One notable impact was that agricultural policies adopted by OECD countries during the "boom" years of the 1970s protected their farmers from fluctuations in world markets. When commodity prices plunged in the 1980s, domestic price supports, particularly in the United States and Europe, were generally maintained, leading to massive over-production and export subsidy programmes. This further depressed both world prices and the incentives for developing countries to produce for either domestic consumption or export. 19/

Policy lessons and priorities

111. The fundamental challenge for sustainable agriculture and future food security is to make better use of available physical and human resources. Dramatic gains in agricultural productivity have proven possible to achieve but some regions remain at least partially "locked" in the low-input, low-output phase of the agricultural transition, unable to capitalize fully on available, superior crop varieties and management techniques. Sub-Saharan Africa especially has not benefited as expected, or has not sustained early gains. The experience of the Green Revolution has shown a number of elements to be critical to more widespread success.

112. Maintaining, or improving on, current trends in yield gains will depend on continued investment in agricultural research, preservation of biological diversity, both in situ and in gene banks, and substantial increases in fertilizer use in many developing countries. Donor funding remains critical for the foreseeable future: the IFPRI scenarios estimate that if international donors were to eliminate all funding of national and international agricultural research, food grain production would drop by 10 per cent and the number of malnourished children in developing countries would increase by 32 per cent. Conversely, if donor funding were to increase by 50 per cent, food grain production would increase by 40 per cent and the number of malnourished children would drop by 30 per cent. 11/

113. Technological improvements will not be sufficient, however. The continuing gap between yields achieved by many farmers and at experimental stations suggests that more emphasis must be put on creating thriving rural economies which provide enabling environments for small farmers. Success factors include improved land tenure, remunerative prices, access to credit and markets that, together, encourage farmers to adopt new agricultural crops and techniques. 20/

114. While market reforms are important, the experience of structural adjustment programmes undertaken by many developing countries as a condition of internationally funded assistance has served to demonstrate that the pursuit of more market-oriented policies does not automatically cause agricultural production to rise in the short term, or the numbers of hungry to fall. More attention is now being paid to the need for complementary policy measures such as the provision of agricultural extension services and facilitating greater participation by programme beneficiaries, especially women.

115. Degradation of the agricultural resource base through soil erosion, loss of cropland to development and contamination of soil and water by agrochemicals must be slowed and, where feasible, reversed. Food supplies in the future will come largely from higher yields on existing agricultural land, and losses of productive land must be compensated by even greater increases in yields. The more land lost, the greater the technological and economic challenge will be. Better management of irrigated land is a particular priority in this regard (chap. V). Policy-making in all countries has too often paid insufficient attention to the protection of agricultural land as a natural resource, suggesting the need for re-evaluation of the economic costs of soil degradation due to poor agricultural practices or land loss from unplanned development.

116. Agricultural productivity gains in many poor and food insecure countries have often been undermined by high losses caused by pest damage in the field and post-harvest losses from inadequate transport and packaging systems. This, in turn, has led to heavy reliance on pesticides, which have compromised human health and ecosystems in many regions. The most feasible alternative remains intensified development and use of integrated pest management (IPM) systems, including both biological and chemical controls. Cooperation between public authorities and the private sector, especially newly emergent biotechnology companies, will be required.

117. The prospect of future regional imbalances between food supply and demand suggests that the developed countries will increase production again in order to meet global needs, but this will occur only in response to rising prices on international markets. If food subsidies to urban consumers in developing countries were reduced, and food prices allowed to rise, the necessary economic incentives might be supplied.

Notes and references

1/ Rome Declaration on World Food Security and World Food Summit Plan of Action (WFS 96/3), adopted at the World Food Summit, Rome, 13- 17 November 1996.

2/ Food and Agriculture Organization of the United Nations, The State of Food and Agriculture, 1995 (Rome, 1995).

3/ Food and Agriculture Organization of the United Nations, Grainger and Garcia, FAO Fisheries Technical Paper 259 (in press).

4/ World Resources Institute, World Resources Report, 1996-97 (New York and Oxford, Oxford University Press, 1996).

5/ L. R. Oldeman, Global Extent of Soil Degradation (Glasod Survey), International Soil Reference and Information Centre (ISRIC), 1992.

6/ Lester R. Brown and Hal Kane, Full House: Reassessing the Earth's Population Carrying Capacity (London, Earthscan Publications Ltd., 1995).

7/ The situation has been worsened by plunging agricultural productivity in the former Soviet Union since 1989, though this trend may be expected to reverse with time.

8/ Food and Agriculture Organization of the United Nations, "The World Food Summit technical background documents" (WFS 1996/Tech/0), Rome, 1996.

9/ Food and Agriculture Organization of the United Nations, "Food, agriculture and food security: developments since the World Food Conference and prospects", technical background document of the World Food Summit (WFS/96/Tech/1), Rome, 1996.

10/ Food and Agriculture Organization of the United Nations, "Assessment of the current world food security situation and medium-term review", item II of the provisional agenda of the twentieth session of the Committee on World Food Security, Rome, 25-28 April 1995.

11/ M. Rosegrant, M. C. Agcaoili and N. Perez, Global Food Projection to 2020: Implications for Investment, Food, Agriculture and the Environment, Discussion Paper 5 (Washington, D.C., International Food Policy Research Institute, 1995)

12/ Nikos Alexandratos, ed., World Agriculture: Towards 2010, An FAO Study (Chichester, United Kingdom, John Wiley and Sons, and Food and Agriculture Organization of the United Nations, Rome, 1995).

13/ R. Rowe, N. P. Shama and J. Browder, "Deforestation: problems, causes and concerns", in Shama, ed., Managing the World's Forests: Looking for Balance Between Conservation and Development (Dubuque, Iowa, 1992). Cited in State of the World's Forests (Rome, Food and Agriculture Organization of the United Nations, 1995).

14/ Finnish Forest Research Institute, personal communication, 8 November 1996.

15/ Sandra Postel, "Water and agriculture", in Peter H. Gleick, ed., Water in Crisis: A Guide to the World's Fresh Water Resources (New York and Oxford, Oxford University Press, 1993).

16/ Food and Agriculture Organization of the United Nations, "Food production: the critical role of water", technical background document of the World Food Summit (WFS/96/Tech/2), Rome, 1996.

17/ P. Binswanger and P. Landell-Mills, The World Bank's Strategy for Reducing Poverty and Hunger: A Report to the Development Community, Environmentally Sustainable Development Studies and Monographs Series No. 4 (Washington, D.C., 1995).

18/ See, for example, C. Ford Runge, "The environmental effects of trade in the agriculture sector", in The Environmental Effects of Trade (Paris, Organisation for Economic Cooperation and Development, 1994).

19/ Food and Agriculture Organization of the United Nations, "Socio-political and economic environment for food security", technical background document of the World Food Summit (WFS/96/Tech/5), Rome, 1996.

20/ Food and Agriculture Organization of the United Nations, "Lessons from the Green Revolution: towards a new Green Revolution", technical background document of the World Food Summit (WFS/96/Tech/6), Rome, 1996.

V. WATER: A MULTIFUNCTION RESOURCE

Introduction

118. The United Nations Water Conference of 1977 covered a wide range of water management issues but is now chiefly remembered for its call to provide safe drinking water and adequate sanitation for all. The Mar del Plata Action Plan, which stemmed from the Conference, and the subsequent International Drinking Water Supply and Sanitation Decade (1981-1990), set out to implement this goal. The provision of services increased substantially over the decade but progress was slowed by the macroeconomic difficulties experienced by many developing countries in the 1980s and by population growth which offset many of the gains.

119. Throughout the 1980s and 1990s, concerns over water quality came to be matched, and in some countries dominated, by the issue of water supply; in particular, the increasing incidence of regional and local scarcity and conflicts over access to limited water resources. Faced with steadily rising consumption, many Governments struggled to meet the demand by raising supplies to match - whatever the cost. An era of gigantic water engineering projects was initiated in the 1950s but peaked by the late 1980s, when protests over their environmental and social impacts, and soaring economic costs, slowed the rate of construction.

120. The International Conference on Water and the Environment, held in Dublin in 1992, and Agenda 21 demonstrate international acceptance of a broader and more complicated water agenda. Water is increasingly recognized as a finite and vulnerable resource and one which is likely to be the principal constraint on development in some countries. Long-term solutions will need to focus more on managing water as an economic good, with an emphasis on increased efficiency of use and more rational allocation among users.

121. Equally, water quality issues have taken on renewed urgency. To humanitarian concern over the suffering caused by inadequate water supply and sanitation services, has been added the realization that pollution of water resources is reducing usable supplies and aggravating the scarcity problem. Thus quality and quantity issues have merged in a way that reinforces the need for a more integrated approach to water policy.

The water transition

122. Socio-economic development is characterized by more active exploitation of water resources. Water is a critical input to both industry and agriculture and economic growth usually requires transfers of water from moist to dry areas and construction of dams and reservoirs for hydropower, irrigation, flood control and seasonal storage. Water pollution from industrialization and urbanization also tends to increase in the early stages of development. Historically, as countries industrialize, the share of water used for agriculture declines and an increasing proportion is taken by the industrial, commercial and household sectors - domestic consumption, in particular, increases greatly with more affluent lifestyles (figure V.1). Worldwide, agriculture still consumes about 70 per cent of available water supplies, with a range from under 5 per cent in some northern European countries to over 90 per cent in parts of Africa, South America and central Asia.

123. There is, however, no clear relationship between per capita water use and national wealth. Higher per capita household and industrial use in rich countries tends to be offset by lower use rates in agriculture. The exception is found in the very poorest countries, which also have the lowest per capita use rates. One encouraging development is that newly industrializing countries appear to be moving towards relatively more efficient patterns of water use more quickly than did the old industrial economies. As with energy use (chap. III), rising per capita income in some higher-income developing countries is characterized by lower per capita water consumption in the industrial and household sectors than was the case in the past. 1/

Water: the question of supply and demand

124. Fresh water supplies are renewable (through precipitation, river inflow and groundwater recharge) but finite; the world's water is a fixed stock. The total human demand for fresh water has risen steadily with increasing population and economic activity. Since 1940, global water withdrawals have risen by an average of 2.5 per cent annually - faster than the rate of population growth (figure V.2). Humanity now uses, directly or indirectly, more than half of the world's water supply that is accessible and per capita availability of fresh water worldwide fell from 17,000 m3 in 1950 to 7,300 m3 in 1995. 2/

125. Globally, fresh water is abundant but it is very unevenly distributed among and within countries. Chronic water shortages already exist in many areas where precipitation is low or unreliable and/or where withdrawals have been significantly increased to meet additional demand from expanding irrigation, industry or urban populations. In addition to pressure on water resources from economic development and changes in social consumption patterns, water supply is increasingly constrained by land use changes (for example forest clearance, which tends to increase run-off and reduce water availability) and contamination from human settlements, industry and agriculture.

126. In most recent analyses, countries are considered likely to experience chronic scarcity problems when water availability falls below a "benchmark" of approximately 1,000 m3 per person per year. However, a recent major assessment of the world's water resources (the Global Water Assessment) 2/ has refined this measure and defined water stress as the ratio of water withdrawal to water availability on an annual basis. A ratio of less than 10 per cent indicates few water resource management problems; a range of 10 to 20 per cent indicates that water availability is becoming a limiting factor and significant investments will be needed in the future; water withdrawals exceeding about 20 per cent of available water indicate that management of both supply and demand will be needed and that resolution of competing uses will be necessary to assure sustainability.

127. Viewed in this way, issues of water demand, and actual or potential water scarcity, are intimately connected not only to population growth but also to the structure of an economy. Decision-making that favours or inhibits alternative uses of water critically affects whether or not a country will be water-scarce at any given level of water availability. Countries where current water use patterns are creating stress or scarcity in at least some part of their territory are identified in figure V.3.

128. Many water-stressed countries have been forced to turn to their groundwater reserves, which are frequently pumped faster than they can be recharged. Such water "overdrafting" is widespread in parts of India, China, Mexico, the United States and the former Soviet Union. Parts of Northern Africa and the Middle East are dependent on withdrawals from "fossil aquifers" which are not recharged at all. Given the rising costs and declining availability of new sources of water supply, some regions may need, over time, to reorient their economies towards less water-intensive end uses.

Competition and conflict

129. A notable trend in the past decade has been sharpening competition between rural and urban water users, over both surface and groundwater resources. When shortages become acute, it is usually (though not always) farmers who lose, since their economic and political lobbying power tends to be less than that of urban and industrial constituents. A number of developed countries are attempting to manage an orderly shift from rural to urban use through reallocation of water use rights, water trading and even schemes to buy out farmers and redistribute their water rights. 3/ However, there could be serious implications for food production if such solutions are adopted more widely (chap. IV).

130. At the international level, conflicts are focused at the river basin level. They have arisen with increasing frequency over projects to dam or divert water by countries in a powerful position upstream of their neighbours. Current flash points include cross-border river basins in South America, North Africa and the Middle East, though an encouraging 30-year cooperation agreement has recently been signed in South Asia. Future conflicts are a potent risk where countries sharing river basins combine circumstances of low water availability, rapidly growing populations, urbanization and industrialization and continued absence of regulation.

Water and human health

131. In developed countries, drinking water and sanitation services were extended to most major urban centres by the early twentieth century, bringing immediate and dramatic improvements in life expectancy. The importance of high quality water to human health and the healthy functioning of society is evident in current environmental spending: water treatment represents the largest item in pollution abatement and control expenditures in the OECD countries. 4/ Infectious diseases have steadily declined in most cities in the developed world. However, despite near universal access to sewerage systems, sewage is not necessarily treated before discharge; an estimated 30 per cent of wastewater in developed countries is still dumped raw into local rivers, lakes or marine waters, posing increasing health risks. 5/

132. There is evidence that developing Asian countries are successfully providing adequate sanitation services at lower per capita income levels than was the case in the developed countries. 6/ However, in many other developing countries, worsening water quality represents one of the most serious health hazards and constraints on socio-economic development. The International Drinking Water Supply and Sanitation Decade (the "Water Decade" 1981-1990) was a coordinated international effort to "speed up" the water quality transition and accelerate the introduction of water services in poor regions. Nearly US$ 100 billion was invested but the results were mixed. Impressive gains in the numbers of people served were offset by population growth, especially in urban areas. In 1994, approximately 1.2 billion people in developing countries lacked safe water supplies and nearly 3 billion lacked access to sanitation services, representing a decline in the percentage of the population served (figure V.4). 2/

133. The difficulties in providing people with adequate drinking water and sanitation services are being compounded by the pollution which now extends for miles around cities in many developing countries: rising volumes of industrial and household effluent contaminate surface and groundwater supplies and often overwhelm municipal treatment capacity. In this context, the importance of protecting groundwater as a resource is only now coming to be fully appreciated. In the Asian and Pacific region, more than 1 billion people depend on groundwater for their drinking water; in many areas, reserves are increasingly threatened by contamination, especially from intensive agriculture. 7/ New irrigation schemes, which create large bodies of standing water favourable to parasites, are responsible for significant increases in water-related diseases. For example, schistosomiasis has spread quickly through river regions following construction of major dams in North and West Africa, reaching infection rates of 90 to 100 per cent, compared with 5 to 10 per cent before construction. 8/

134. The World Health Organization estimates that almost half the world's population is suffering from debilitating water-borne or water-related diseases which account for an estimated 5 million deaths each year. Good progress has been made in reducing the incidence of guinea-worm disease, which may be eradicated in the near future, but epidemics of other infectious diseases, notably diarrhoeal diseases, cholera and schistosomiasis, remain frequent.

Water and ecosystem functions

135. Worldwide efforts to increase available water supplies have involved major trade-offs: natural capital has been lost, in return for gains in energy generation, food production and socio-economic development. The past half century has witnessed an unprecedented period of construction of dams, canals, reservoirs and pipelines. The number of major dams (over 15 metres in height) in the world rose from just over 5,000 in 1950 to nearly 38,000 today, with over 60 per cent in Asia. 9/ This "plumbing" of the world's river systems, though essential in some cases, has led to multiple environmental and social impacts, some of which are now recognized as serious obstacles to future development.

136. In many of the world's great rivers, the volume and timing of water flow is almost entirely controlled, with virtually no water reaching the sea. This has resulted in major disruptions to aquatic habitats, declining fish stocks and significant losses in biodiversity. Another important objective of water engineering is flood control. Ironically, there is evidence that flooding is worsening in some regions due to excessive canalization of rivers and loss of wetlands which act as natural sponges. Successful flood prevention, on the other hand, is preventing replenishment of deltas and fertilization of flood plains worldwide because silt tends to be trapped behind dams in reservoirs. Global loss of reservoir capacity due to sedimentation is estimated at 10 per cent per decade. 2/ Loss of productive land as deltas retreat is a problem which could be compounded by climate change and rising sea levels. For example, it has been estimated that Egypt could lose up to 19 per cent of its habitable land, displacing some 16 per cent of the population, over the next 60 years. 10/

137. Major concerns arise over waterlogging and salination of soils caused by poor irrigation techniques. Irrigation has accounted for more than half of the increase in global food production since the mid-1960s and is expected to contribute strongly to meeting future additional demand (chap. IV). However, about 20 per cent (50 million hectares) of the world's irrigated land is affected, to some degree, by soil degradation caused by faulty irrigation practices. 2/ In many areas, salination has reached the point where food production is significantly reduced; studies have found that yields of major crops in North Africa and parts of Asia are being cut by 30 per cent. 10/

138. The trend to ever bigger water engineering projects has slowed in the 1990s due to low food prices (which discourage the expansion of irrigation), rising real costs of construction and growing recognition of the environmental and social consequences. Much of the damage due to misuse of water resources could be reversed, given time, political will and finance. The more fundamental problem is that, despite all the effort put into storing and transporting fresh water, supply side solutions have not kept pace with rising demand in many water stressed countries. This suggests the need for alternative approaches if future demands are to be met.

Looking ahead

139. Water management over the next decades will be a development issue (ensuring water supplies are sufficient for economic activity without undermining the natural resource base of soil and water), a political issue (avoiding domestic and international conflicts over a scarce resource) and a welfare issue (providing people with the water services necessary for good health and quality of life). Projections regarding future water demand and availability are highly uncertain, being dependent on assumptions about population and economic growth, investment in additional water supply, relative demand from different consuming sectors, the development and adoption of new technologies and the extent to which existing supplies become contaminated.

140. Projections made in the recent Comprehensive Freshwater Assessment (see note 2 below) are based on the United Nations medium population growth forecast and assume no major changes in policy or technology; they represent a business-as-usual continuation of present trends. Estimated withdrawals reach a global total of over 5,000 km3 per year by 2025 (figure V.5). 141. The Conventional Development Scenario projects a much lower rate of increase, based on greater shifts worldwide towards less water-intensive economic activities and higher levels of water efficiency. Its estimate of total withdrawals of 4,300 km3 by 2050 is considered very conservative by most experts. At the opposite extreme, another estimate has been produced as a "backcasting" exercise, based on FAO projections of future food demand; it represents the amount of water that would be required to produce sufficient food for the world's population in 2025. Depending on water efficiency in agriculture, and other factors, estimates of this variant range from a 50 to 100 per cent increase in water demand over the next 30 years.

142. Future water supplies in many developing countries could be critically determined by contamination of rivers, lakes and groundwater reserves by industrial emissions and agricultural and urban run-off. Agricultural use of fertilizers and pesticides is expected to increase rapidly in developing countries in order to meet rising food demand and, if inadequately managed, industrial growth and urbanization will produce dramatic increases in pollution. The extent to which societies will choose to limit water withdrawals for economic purposes in order to protect terrestrial and aquatic ecosystem functions is a further uncertainty, especially in industrializing countries where standards of living and interest in environmental quality are rising rapidly.

143. Despite their wide range, most forecasts of water use suggest that demand from all economic sectors will continue to grow. On current trends, almost two thirds of the world population in 2025 will be subject to moderate to high water management difficulties, and almost half the world will have difficulties in coping due to inadequate financial resources. Figure V.6 illustrates the projected water stress index for countries in 2025; it is a conservative estimate, which assumes that per capita water withdrawals remain at 1995 levels.

144. It appears most likely that water will become a critical issue in development. Worldwide, water use cannot rise to match projected demand without a substantial increase in available supply, much greater use of the existing supply and major efforts to prevent pollution. Where increasing available supply to match demand is economically and/or environmentally infeasible, the emphasis must be on water conservation and pollution prevention. Conservation - especially water reuse in industry and greater efficiency in household and agricultural use - represents the greatest "resource" still to be exploited.

The role of policy

145. Water has traditionally been a State responsibility in most countries. Viewed as a strategic resource but also a public good, water resources have tended to be centrally planned and protected from market forces. Politically, major projects to increase water supply have often proved more attractive than measures aimed at water conservation and efficiency.

146. Political and social considerations have encouraged Governments worldwide to provide large subsidies to insulate water users from the true costs of provision. The World Bank estimates that, on average, municipal water users in developing countries pay about 35 per cent of the average cost of water supply. Developed country Governments also subsidize municipal users though generally at a lower rate than agricultural users. In both developed and developing countries, most urban water utilities further distort prices by charging flat or declining rates per unit of water use, rather than rates that rise to reflect increasing marginal costs of supply.

147. Worldwide, water subsidies benefit farmers above all other users; farmers typically pay between 10 and 20 per cent of the construction and operating costs of irrigation projects. 11/ Artificially low water prices largely underwrote the worldwide boom in irrigation and thereby enabled essential increases in agricultural production. However, they also encouraged the development of wasteful and inefficient practices; on average, 55 per cent of irrigation water never reaches the crop. 12/ Generous allocation of subsidized water to agriculture - while helpful for food security, rural poverty alleviation and increased agricultural exports - may not be the most economically beneficial option overall. Planners in China calculate that a given amount of water used in industry generates more than 60 times the value of the same amount used in agriculture. 10/

148. Water management at the national level is characterized by fragmentation of responsibility, for example, among ministries of agriculture, energy, health, transport, environment, economic affairs and local authorities. Despite some evidence of a move towards integration - linking water legislation to economic and social issues - weak and divided administrative structures persist in many countries. Isolated decision-making in one sector can create problems - or foreclose future options - in another. This has especially been the case in decisions relating to major irrigation schemes: the over-ambitious attempt to irrigate much of the Aral Sea Basin has led to economic, environmental and human health damage totalling an estimated 37 billion roubles; wide areas have been economically ruined. 13/

Policy lessons and priorities

149. There is a growing consensus that increasing water supplies to match demand will, in many areas, be economically prohibitive. Costs of tapping new water supplies are projected to rise to between two and three times the cost of current investments. 6/ A prime lesson of recent years, therefore, is recognition that improved management and greater efficiency are the keys to stretching water resources. Two broad policy approaches hold particular promise: (i) institutional change to encourage more integrated and effective water management and (ii) technical innovation and pricing reform to improve efficiency of use and manage demand.

150. Some countries are beginning to decentralize their water supply services and give a greater role to the private sector. Municipal authorities in most regions of the world are experimenting with some transfer of water management to private companies, autonomous utilities or water user associations in the hope of improving cost and water efficiencies and standards of service. 14/ Top priority needs to be given to providing safe drinking water and sanitation in Africa, Latin America, Asia and the Pacific region. However, a prime lesson of the Water Decade has been the inappropriateness of transferring developed country standards wholesale to developing countries. Western infrastructures have developed over more than a century, in step with economic and population growth. The size of population and budgetary constraints in many developing countries mean that providing all residents with, for example, piped water and flush toilets in their houses is not a realistic option in the near term. Low-cost solutions, tailored to local conditions and needs, are proving themselves effective but are commonly resisted on the grounds that they represent a low-technology, second-rate development route.

151. In rural areas, watershed management is being promoted as a means to more integrated management of water, forest and agricultural land resources. A number of Governments and international aid agencies are supporting innovative attempts to promote joint management of watersheds by local residents and government agencies. In northern Thailand, for example, an approach has been developed involving community mapping of watersheds, a physical model of the watershed area that serves as a basis for a joint zoning exercise and the development of a network of participating communities within the watershed to deal with transboundary issues. 15/

152. The growing number of cross-border conflicts over water supplies clearly requires shared water resources to be developed on a more cooperative basis in the future. International agreements regulating water use in shared river basins are com