IMPACT ON FOOD SECURITY AND RURAL DEVELOPMENT OF REALLOCATING WATER FROM AGRICULTURE FOR OTHER USES MARK W. ROSEGRANT, CLAUDIA RINGLER INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE ABSTRACT Water in its multiple uses--for irrigation, household, industrial, and environmental uses--is essential to a healthy and productive life, as well as for the expansion of economic opportunity. The competition for limited water resources between agriculture and more highly valued domestic and industrial water demands is rapidly increasing, however, and will likely require a reallocation of water from agricultural uses to meet these demands. This paper examines the impact of the shift of water resources from agricultural to other uses on the household, on local, national and global food production scenarios, and on global food security. National and global impacts are explored using the IMPACT global food model to create alternative scenarios. The baseline scenario shows that effective food demand in 2020 will be met with slowly declining food prices and significantly increased food imports from developed to developing countries, as well as little improvement in food security in poorer regions. An alternative scenario, simulating a large reallocation of water from agricultural uses without countervailing gains in water use efficiency and productivity, demonstrates the potentially dramatic macro-level impacts of water reallocation on national and global food markets. Under this scenario--in contrast with the baseline scenario--prices of staple foods increase sharply, consumption is depressed in low-income regions, and malnutrition increases. The reallocation of water from agriculture can also negatively impact rural economies, causing loss of income, decreased food production and overall social disruption. On closer examination, however, the actual impacts are mixed. Reallocation tends to have negative impacts for rural communities in particular circumstances: when the transfers are so large that they eliminate farming or other economic opportunities in the area of origin, when farmers have no incentive to sell but water resources are transferred anyway, and when institutions and secure legislation to adequately compensate sellers or third parties are absent. Conversely, the reallocation of water from agricultural uses can have positive impacts. Such water transfers can stimulate economic growth in both rural and urban areas, especially under certain scenarios: when farmers sell only a portion of their water, when sellers are compensated for the water resources sold, when sellers have a stake in the economic development of the urban area, or when sellers' water rights are adequately protected by institutions and organizations. Appropriate policies to sustain food security and agricultural productivity growth, especially in local economies losing water resources through such reallocation, will vary from region to region. Important elements of policy reform will therefore include the establishment of secure water rights, the decentralization and privatization of water management functions to appropriate levels, the use of incentives for water conservation including reduction in water subsidies and establishment of markets in tradable property rights, and the introduction of appropriate water-saving technologies. In addition, specific compensatory measures should be taken for the poorest water users in the rural communities, those who would be the most harshly affected by water transfers. 1. Introduction Population and economic growth in developing countries will pose serious challenges for humanity in simultaneously meeting food requirements and water demands. Competition for limited water resources increasingly occurs between different stakeholders and at different levels: between farmers within an irrigation system; between irrigation systems in the same river basin; between the agricultural sector and other rural uses, such as fisheries or domestic water supply and drinking water; and more and more between agricultural and urban and industrial users and uses; and environmental uses (for example, instream flows and recreation). Agriculture still accounts for the majority of global water withdrawals, and is often responsible for 80 percent or more of total withdrawals for consumptive uses in developing countries, but, as this paper will show, it is likely that significant amounts of water will be reallocated from agricultural uses to higher valued domestic and industrial water demands. The impacts of the shift of water from agricultural to other uses on household, local, national, regional, and global food production and food security have not been studied in an integrated manner; this paper reviews and synthesizes the available evidence. The paper focuses on the implications of water reallocation at the sectoral level, in particular, on the transfer of water out of agriculture to meet urban and industrial demands, to provide an overview of the issues involved in water reallocation and to assess the potential local, sectoral, and global impacts. The following sections examine recent food supply and demand trends as well as projections of global food production based on IFPRI's global food model; describe the role of irrigation in global food production; and examine recent trends and projections in water demand. The paper then addresses the potential for meeting future water demands through expansion of water supplies; provides an account of the potential impacts of reallocation on global food production and on local and regional rural economies; and examines the potential for water policy reform and demand management to save water and minimize adverse impacts when water is reallocated from agriculture, followed by some concluding observations. 2. Recent Trends in and Projections of Global Food Supply and Demand 2.1. Recent Trends in Global Food Supply and Demand The world population is expected to grow to 7.7 billion in 2020, from 5.3 billion in 1993 (UN, 1996), raising serious concerns about how food demand will be met in the next decades. In addition, the global urban population is expected to increase to about 5.1 billion by 2025, from 1.5 billion in 1975, and 2.6 billion in 1995. The majority of the population is projected to live in urban areas by 2025 (61 percent), up from 38 percent in 1975 and 45 percent in 1995. Almost all urban population growth, about 90 percent, will occur in developing countries, where roughly 150,000 people are added to the urban population every day (WRI, 1996). These developments will have serious impacts on global food supply and on the structure of water demand. 2.2. Projections of Global Food Supply and Demand to 2020 Projections of global food supply and demand have been made using an updated version of IFPRI's International Model for Policy Analysis of Commodities and Trade (IMPACT). The model covers 37 countries and regions and 17 commodities, including cereals, roots and tubers, soybeans, and meats, and is specified as a set of country-level supply and demand equations, with each country model linked to the rest of the world through trade. Food demand is a function of prices, demand elasticities, income and population growth. Growth in commodity production in each country is determined by prices and the rate of productivity growth, which in turn is influenced by advancements in public and private agricultural research and development, extension and education, markets, infrastructure and irrigation. Irrigation expansion directly affects area harvested and yields. The world price of each commodity is determined as the price that clears world markets. A full description of the model is beyond the scope of this paper, but see Rosegrant, Agcaoili-Sombilla and Perez (1995) for the detailed model structure; and Rosegrant et al. (1997) for a detailed presentation of the baseline results summarized below. The baseline analysis using IMPACT projects that real world prices of food will decline, but more slowly than in the past two decades. Cereal prices on average are projected to drop by about 10 percent by 2020, and meat prices by 6 percent. Projected real prices of cereals will be nearly constant through 2010, but the continued slowdown in the population growth rate after 2010, together with declining income elasticities of demand for cereals, will reduce demand growth enough to cause cereal prices to fall. The tighter future price scenario implies that shortfalls in meeting water demand for agriculture could put serious upward pressure on food prices. This issue will be explored below. In developing countries, especially in Asia, rising incomes and rapid urbanization will change the composition of cereal demand. Per capita food consumption of maize and coarse grains will decline as consumers shift to wheat and rice, livestock products, fruits and vegetables, and processed foods. The projected strong growth in meat consumption, in turn, will substantially increase cereal consumption as animal feed, particularly maize. Growth in cereal and meat consumption will be much slower in developed countries. These trends will lead to an extraordinary increase in the importance of developing countries in global food markets: 82 percent of the projected increase in global cereal consumption, and nearly 90 percent of the increase in global meat demand between 1993 and 2020 will come from developing countries. Developing Asia will account for 48 percent of the increase in cereal consumption, and 63 percent of the increase in meat consumption. The composition of food demand growth across commodities will change dramatically. Total cereal demand is projected to grow by 717 million metric tons (mt), or by 40 percent, of which 35 percent will be maize; 31 percent, wheat; 18 percent, rice; and 16 percent, other coarse grains. How will the expanding cereal demand be met? Expansion in area will contribute very little to future production growth, with a total increase in cereal crop area of only 39 million hectares (ha) by 2020, from 700 million ha in 1993, 88 percent of which will originate in developing countries. The projected crop area growth represents the net effect of slow expansion in irrigated area (see section 3.3); slowly increasing crop intensity on existing irrigated areas; declining commodity prices that limit the profitability of investment in land expansion; and gradual loss of land to soli degradation and urbanization. The slow growth in crop area places the burden to meet future cereal demand on crop yield growth. Although yield growth will vary considerably by commodity and country, in the aggregate and in most countries it will continue to slow down. The global yield growth rate for all cereals is expected to decline from 1.5 percent per year in 1982-94 to 1.1 percent per year in 1993-2020; in developing countries, average crop yield growth will decline from 1.9 percent per year to 1.2 percent per year; and in developed countries from 1.3 percent per year to 0.9 percent per year. Even with these reduced growth rates, yield growth will account for 80 percent of growth in cereal production in developing countries, and for 94 percent in developed countries. 2.3. Food Demand and Supply Gaps and World Trade in Food Two types of food gaps can be identified. The most devastating is the gap between actual food consumption and the quantity and quality of food required to sustain a healthy and productive life. By this measure, there will be little improvement in food security for the poor in many regions. Sub-Saharan Africa will have only small increases in per capita calorie availability as income growth will be only slightly in excess of population growth, and the number of malnourished children is projected to increase by 12 million in 1993-2020. Thus, even with relatively abundant food in the world, there will not be enough growth in effective per capita demand for food in Sub-Saharan Africa to improve the food supply situation. More progress can be seen for South Asia, home to more than one-half of the world's malnourished children, but nearly 70 million children will still be malnourished in the region in 2020. The second type of food gap is the difference at the national level between food production and food demand as reflected in food imports. Growing imports are not a problem if they are the result of strong economic growth generating the necessary foreign exchange to pay for the food imports. In the case of some Middle Eastern countries facing extreme water scarcity and sharp population increases, the strategy of substituting food imports for irrigated agricultural production paid for by (water-based) urban and commercial growth has been called imports of "virtual water" (Allan, 1996). However, even when rapidly growing food imports are primarily a result of rapid income growth, they often act as a warning signal to national policymakers concerned with heavy reliance on world markets, and can induce pressures for trade restrictions that would damage growth and food security in the longer term. More serious food security problems arise when high food imports are the result of slow agricultural and economic development that fails to keep pace with basic food demand growth driven by population growth. Under these conditions, it may be impossible to finance the required imports on a continuing basis, causing a further deterioration in food in the ability to bridge the gap between food consumption and food required for basic livelihood. World trade in food is projected to increase rapidly, with trade in cereals projected to increase from 186 million mt in 1993 to 349 million mt in 2020, and trade in meat products to increase from 8 million mt to 23 million mt. Expanding trade will be driven by the increasing import demand from the developing world: net cereal imports in developing countries are projected to rise by nearly 150 percent, from 94 million mt in 1993 to 229 million mt in 2020, and net meat imports are expected to increase from less than 1 million mt in 1993 to 11 million mt in 2020. Trouble spots for food trade gaps are Sub-Saharan Africa, and potentially West Asia and North Africa (WANA). Cereal imports in Sub-Saharan Africa are projected to increase from 12 million mt in 1993 to 29 million mt in 2020. It is highly unlikely that this level of imports could be financed internally, but instead would require international financial or food aid. Failure to finance these imports would further increase malnourishment in this region. In WANA, cereal imports are projected to increase from 38 million mt in 1993 to 65 million mt in 2020, with most of this increase expected to occur in the nonoil-producing countries. 3. The Role of Irrigated Agriculture in Global Food Production 3.1. Contribution of Irrigation to Global Food Production Worldwide, the agriculture sector is the largest consumer of water. During the 1950s to the 1980s, irrigation expanded rapidly and currently accounts for about 72 percent of global water withdrawals, and about 90 percent of water use in the lowest-income developing countries. Such a major role for irrigation had been justified by the contribution of irrigation systems to stabilizing, then expanding national and world food supplies during the Green Revolution, especially in Asia (Svendsen and Rosegrant, 1994). Dramatic increases in yield during and after the Green Revolution were largely based on the introduction and successful adoption of high-yielding varieties of wheat and rice that depend heavily on timely nutrient and pest control management as well as irrigation applications to secure and control soil moisture (FAO, 1996). Irrigated agriculture was a major factor in achieving the yield growth rates described above. In the mid-1990s, irrigated agriculture contributed nearly 40 percent of world food production on 17 percent of the cultivated land. In India, for example, irrigated areas (one third of total cropped area) account for more than 60 percent of total production. Over the next 30 years, as much as 80 percent of the additional food supplies required to feed the world may depend on irrigation (IIMI, 1992). Irrigation also furthers stability through greater control over production and scope for crop diversification. In many developing countries, irrigation constitutes an important element of rural development policies, as it provides higher rural incomes and employment and allows for increased agricultural and rural diversification through secondary economic activities derived from extended and more varied agricultural production (as compared to rain-fed agriculture). In addition, in arid and semi- arid areas, alternatives to irrigated agriculture are rare, and water reallocation can lead to rural-urban migration and abandonment of plots (Fereres and Cen~a, 1997; Raskin, Hansen and Margolis, 1995; Wolter, 1997). Thus, irrigation plays a vital role in achieving food security and sustainable livelihoods in developing countries, both locally, through increased income and improved health and nutrition, and nationally, through bridging the gap between production and demand. 3.2. Recent Trends in Irrigated Area The development of new irrigation has slowed considerably since the late 1970s, due to escalating construction costs for dams and related infrastructure, low and declining prices of staple cereals, declining quality of land available for new irrigation, and increasing concerns over the environmental and negative social impacts of large-scale irrigation projects. Lending for large-scale irrigation projects from international donors declined sharply after the 1970s: loans from four major donors, the World Bank, the Asian Development Bank, the U.S. Agency for International Development (USAID), and the Japanese Overseas Economic Cooperation Fund (OECF) peaked in the late 1970s, but by the late 1980s were just over 50 percent of the 1977-79 level (Rosegrant, 1997). These declining expenditures are reflected in the declining growth in crop area under irrigation. Globally, the growth rate in irrigated area declined from 2.16 percent per year during 1967-82 to 1.46 percent in 1982-93. The decline was slower in developing countries, from 2.04 to 1.71 percent annually during the same periods, but the lagged effect of declining investment in irrigation will be increasingly felt through future slowdowns in expansion of irrigated area. Declining investment in irrigation has been accompanied by a decline in the quality and performance of existing irrigation systems. Although data are limited and definitions of damaged area vary between sources, estimates of annual global losses of agricultural land due to waterlogging and salinization range from 160,000-300,000 ha (Tolba, 1978; Barrow, 1991) to 1.5 million ha (Kovda, 1983). Most of the waterlogging and salinization have occurred in irrigated croplands with high production potential. Global estimates of the total area affected by salinity but still in production also vary considerably. El-Ashry (1991), Barrow (1991), Rhoades (1987), and Kayasseh and Schenck (1989) estimate that salinity seriously affects productivity in 20 to 46 million ha of irrigated land. However, with expansion of irrigation into new areas likely to be slow, the future contribution of irrigation to food production must come mainly from improvement in the productivity of the existing irrigated land base. This implies both the need to increase the efficiency of water use and the need to improve the quality of the resource base in irrigated areas, reversing the trends towards increased degradation through waterlogging and salinization of soil, as well as degradation of water quality and groundwater mining (Rosegrant and Pingali, 1994). 3.3. Projections of Irrigated Area to 2020 Rosegrant, Ringler, and Gerpacio (1997) assessed future expansion in irrigated area, consistent with the underlying assumptions in the global food projections. The projections indicate a continued decline in irrigated area growth. In developed countries, irrigated area is expected to increase by only 3 million ha between 1995 and 2020, at an annual rate of growth of just 0.2 percent, compared to 0.8 percent annually during 1982-93. In developing countries, an additional 37 million ha of irrigated area is projected by 2020, at an annual rate of increase of 0.7 percent, compared to 1.7 percent per year during 1982-93. For the world as a whole, irrigated area is projected to grow at 0.6 percent per year, compared to 1.5 percent during 1982-93. The largest increase is expected in India with 17.3 million ha by 2020, as public investment in irrigation has remained relatively strong and private investment in tubewells has been very rapid. However, even in India, the projected 1995 to 2020 rate of growth in irrigated area of 1.2 percent per year is well below the rate of 2.0 percent per year during 1982-93. Area under irrigation will remain very low in Sub-Saharan Africa, despite a potential increase of 50 percent to 7.4 million ha in 2020. Simulations suggest that increased investment in irrigation can make a significant contribution to food production growth in Sub-Saharan Africa, although the amount of land under irrigation and the potential area exploitable relative to total crop area may not be large enough to generate revolutionary increases in crop production (Rosegrant and Perez, 1997). 4. Recent Trends in and Projections of Water Demand 4.1. Recent Trends in Global Water Demand Given the current global use of water of around 3,700 billion cubic meters (BCM), the estimated 9,000-14,000 BCM of reliable annual freshwater runoff would be adequate to meet growth in demand in all sectors for the foreseeable future, if supplies were distributed equally across the world's population. But freshwater is distributed unevenly across the globe. Per capita water availability is highest in Latin America and North America, while Africa, Asia, and Europe have far less water per capita. However, these regional figures hide the huge variability in water availability. Freshwater is poorly distributed across countries (Canada is blessed with 120,000 cubic meters (cu m) per capita per year of renewable water resources; Kenya has 600 cu m; and Jordan, 300 cu m); across regions within countries (although India has adequate average water availability of 2,500 cu m per capita, the state of Rajasthan has access to only 550 cu m per capita per year); and across seasons (Bangladesh annually suffers from monsoon flooding followed by severe dry season water shortages) (Rosegrant, 1997). Moreover, with a fixed amount of renewable water resources supplying an increasing population, per capita water availability has declined from 9,600 cu m to 5,100 cu m in Asia, and from 20,000 cu m to 9,400 cu m in Africa between 1950 and 1980 (Ayibotele, 1992). Tightening supplies have been accompanied by rapid growth in the demand for water. Between 1950 and 1990, water use increased by more than 100 percent in North and Latin America, by more than 300 percent in Africa, and by almost 500 percent in Europe (Clarke, 1993). Global demand for water has grown rapidly, at a rate of 2.4 percent per year since 1970. In 1995, annual per capita domestic withdrawals ranged from a high of 240 cu m in the U.S. to only 11 cu m in Sub-Saharan Africa, a level that is just over one-half of the 20 cu m per capita estimated by Gleick (1996) as required to meet the most basic human needs. China, India, and other South Asian countries are all at or just above this basic human needs level. Southeast Asia, Latin America, and WANA cluster at 56 cu m per capita to 65 cu m per capita. For developing countries as a group, per capita water demand was 33 cu m in 1995, less than one- fourth the amount in developed countries. In addition to the basic water requirements for sanitary and other domestic uses, estimates of minimum water requirements for basic food needs range from 400 cu m per capita per year (Postel, 1996) to 1,000-2,000 cu m per capita annually (FAO, 1989). However, actual minimum requirements are often higher, especially in urban areas, due to increased living standards. Expansion of high quality freshwater supply to domestic users is essential to the development of improved health and well-being. Unsafe drinking water, combined with poor household and community sanitary conditions, is a major contributor to disease and malnutrition, particularly among children. The World Bank (1992) has estimated that access to safe water and adequate sanitation could result in 2 million fewer deaths from diarrhea among young children. However, it is estimated that 1 billion people in the developing world do not have access to potable water, and that 1.7 billion have inadequate sanitation facilities. Pollutants from disposal of untreated sewage and poor sanitation are becoming a very serious problem in domestic water supply, especially in and downstream of major cities. In addition, contaminated wastewater is often used for irrigation, creating significant risks for human health and well-being. Environmental demands also gain higher priority with rising incomes. In a growing number of developed countries, environmental uses are even becoming the first claimant on available water resources; in developing countries, these demands are increasingly acknowledged, but honored usually only if local economic development is not hindered. However, the latent demands are expected to be served as incomes grow (Burton and Chiza, 1997; Franks, Shahwahid and Lim, 1997; Grossman and Krueger, 1997). 4.2. Projections of Water Demand to 2020 Taking into account long-term growth in income, industrial expansion, and irrigation development, Rosegrant, Ringler, and Gerpacio (1997) project that global water withdrawals will increase by 35 percent by 2020, to 5,060 BCM, with growth in developing countries much faster than in developed countries. Developed countries as a group will increase water demand by 22 percent to 1,710 BCM, more than 80 percent of which will be for industrial uses. The demand pressure on water resources will be much higher in the developing world, where water withdrawals are projected to increase by 43 percent, from 2,347 BCM in 1995 to 3,350 BCM in 2020. In sharp contrast to past growth patterns in developing countries, the projected absolute increase in domestic and industrial water demand of 589 BCM from 1995 to 2020 will be greater than the increase in agricultural water demand of 415 BCM. With these differential rates of growth, the combined share of domestic and industrial water demand in total water demand in developing countries will more than double, from 13 percent to 27 percent. 5. Potential for Meeting Future Water Demands Through Supply Expansion Can the rapid growth in water demand, particularly in the domestic and industrial sectors, be met without massive transfers of water out of agriculture that could derail the projected growth in crop yield and area described above? This section examines the potential for expansion of water supplies through traditional and non-traditional means. 5.1. New Investment in Irrigation and Water Supply Development of irrigation and water supplies has become increasingly expensive. In India and Indonesia, for example, the real costs of new irrigation have more than doubled since the late 1960s and early 1970s; costs have increased by more than 50 percent in the Philippines; they have tripled in Sri Lanka; and increased by 40 percent in Thailand (Rosegrant and Svendsen, 1993). In China, Pakistan and Indonesia, irrigation has absorbed over half of all agricultural investment, and about 30 percent of all public investment in India. In addition, once established, irrigation projects become some of the most heavily subsidized economic activities in the world, both directly and indirectly. In the mid-1980s, it was estimated that average subsidies to irrigation in six Asian countries covered 90 percent or more of the total operating and maintenance costs (Repetto, 1986). The cost of supplying water for household and industrial uses is also increasing rapidly. In Shenyang, China, the cost of new water supplies will nearly triple from US$0.04 to US$0.11 per cu m between 1988 and 2000 because pollution of the current groundwater source will require a shift to water conveyed by gravity from a surface source 51 kilometers (km) from the city. In Mexico City, water is currently being pumped over an elevation of 1,000 m into the Mexico Valley from the Cutzamala River through a pipeline about 180 km long, at an average incremental water cost of US$0.82 per cu m, almost 55 percent more than the previous source, the Mexico Valley aquifer (World Bank, 1993). Because of the high costs and increasing concerns about economic, environmental, and social impacts, it will be difficult to justify construction of large-scale dams and water supply systems, despite the fact that a review of the World Bank's experience with irrigation shows that there are in fact economies of scale in irrigation projects: the rates of return to large projects have been higher than returns to small- scale projects (Jones, 1995). However, these estimates do not take into account the full range of negative externalities generated by these projects, and also do not account for the economic, environmental, and social consequences if the projects are not developed. The heightened national and international concern over the broad environmental and human effects of large irrigation projects will make it very difficult to proceed with many of these projects. Small-scale irrigation projects can have considerable advantages over large-scale projects. However, in many cases the bureaucratic mode of implementation has effectively eliminated the potential advantages, and big and small systems often share a number of common characteristics: high capital costs per ha and per farmer; bureaucratic, costly, and inefficient management; low technical efficiency, low settler incomes, and zero or negative returns (Adams, 1990). Farmer-owned and -controlled systems, on the other hand, have a better performance record. Experience indicates that it is not so much the size of the irrigation system that determines its success, but a host of institutional, physical, and technical factors. Every river basin is different, and the appropriate choice of system size and operational characteristics in any given basin is likely to be determined by conditions unique to that basin. A pragmatic approach to project design should be taken that ensures quantification of full benefits, including not only irrigation benefits, but also health, household water use, and catchment improvement benefits (Jones, 1995) and full assessment of, and compensation for, negative environmental and resettlement costs. Selective development of new surface water must still play a role in future water resource development. Sustainable development of groundwater resources offers significant opportunities for some countries. The massive expansion of private sector tube well irrigation in Bangladesh, India, and Pakistan is the most successful example of private sector irrigation development in the developing world. However, extensive investigation is required to determine the characteristics of aquifers (including geometry, continuity, boundaries, hydraulic characteristics, spatial and temporal variability), sustainable exploitation rates, and the potential adverse environmental and other impacts of these water sources. 5.2. Desalination The supply of freshwater through desalination is in essence infinite, but expensive. However, although desalination capacity increased 13-fold from 1970 to 1990 to more than 13 million cu m per day, desalinated water accounts for just one-tenth of one percent of freshwater use (Engelman and LeRoy, 1993; Gleick, 1993). Nearly 60 percent of the desalination capacity in the world is in the oil-rich, water-scarce Persian Gulf, and much of the rest is on island nations and in other arid countries (Postel, 1992). Although 'raw' production costs for desalination are comparable to the costs of new supplies in some of the most arid areas of the world, a wide diffusion of this technology is unlikely considering the often substantial transportation costs to pump the desalinated water inland, the high capital and energy costs, and the potential environmental damages from generated wastes. Growth of desalination capacity will likely be confined to coastal regions that are both very water scarce and relatively wealthy. 5.3. Recycling and Wastewater Reuse After being used once, freshwater can be used again in the same home or factory (usually called recycling) or collected from one or more sites, treated, and redistributed and used in another location (generally called wastewater reuse) (Postel, 1992). The greatest potential for water saving is likely to be industrial recycling, although wastewater reuse can offer significant and increasing savings as the scarcity value of water increases. Only a small fraction of industrial water used for cooling, processing, and other activities is actually consumed. Although the rest of the water may be heated or polluted, it can often be recycled within a factory or plant, thereby getting more output from each cu m delivered to that operation. In developed countries, pollution control laws have been a primary motivator for industrial water recycling. Japan, for example, produced industrial output of US$77 per cu m of water supplied to industries in 1989, compared with US$21 per cu m in 1965 (in real terms). In the U.S. between 1950 and 1990, total industrial water use fell 36 percent while industrial output increased nearly fourfold (Postel, 1992). Similar conservation efforts have also begun in water- scarce developing country cities. In Beijing, China, for example, the water recycling rate increased from 61.4 percent in 1980 to 72.3 percent in 1985; between 1977 and 1991, total industrial water use declined steadily while output increased by 44 percent in real terms (Nickum, 1994). The rate of expansion of wastewater reuse depends on the final quality of the wastewater and on the public's willingness to use these supplies. In California, which has the highest reuse of wastewater in the U.S., this water is reused for barriers against salt water intrusion, dust control, groundwater recharge, industrial cooling, wetlands, and irrigation of parks, golf courses, and certain types of crops. Even in California, however, wastewater reuse accounts for less than 1 percent of the state's developed water supplies (Frederick, 1993). Worldwide, about 500,000 ha of cropland is irrigated by treated municipal wastewater, amounting to only two-tenths of 1 percent of the world's irrigated area. In water-scarce developing countries, wastewater reuse for agricultural irrigation can provide a strong economic impetus because it can help to conserve resources (including water and soil nutrients) and protects the environment by preventing river pollution, protecting water quality, and preventing seawater intrusion in coastal areas (Shuval, 1990). Israel undertakes the largest wastewater reuse effort in the world, treating 70 percent of the nation's sewage to irrigate 19,000 ha of cropland. Reclaimed wastewater is projected to supply more than 16 percent of Israel's total water needs by the start of the next century. Most of this would be used in agriculture to replace freshwater reallocated to nonagricultural uses (Postel, 1992). Wastewater can be used for crops that tolerate low water quality and could potentially contribute much to reforestation and revegetation activities. However, given the relatively high cost of wastewater treatment and transport to agricultural areas, it is likely that wastewater can make up an important share of agricultural water supply only in arid regions where the cost of new water supplies has become very high. In order to generate substantial increases in wastewater use, major technological improvements in wastewater treatment and reuse would be required to reduce the unit cost of wastewater reuse. 5.4. Water Harvesting Water harvesting, the capture and diversion of rainfall or flood water to fields to irrigate crops, has been used for centuries in traditional agriculture. The improvement and expanded use of such techniques can increase production and farm income in some environments. In semi-arid areas of India and Pakistan, low earthen banks are constructed to hold back the monsoon floods and submerge and saturate the fields. Crops are then planted when the floods recede. In Bihar, India, as many as 800,000 ha of land are planted under this system (Clarke, 1993). Stone bunds, terracing, and vegetative barriers can also be used for water harvesting. Vetiver grass, native to India and known there as khus, has been used in both Africa and Asia. When densely planted along the contours of a sloping field, the grass forms a vegetative barrier that slows runoff, allowing rainfall to spread out and seep into the soil more easily (Postel, 1992). Water harvesting can provide farmers with improved water availability, increased soil fertility, and higher crop production in some local and regional ecosytems. Water harvesting can also provide broader environmental benefits through reduced soil erosion. However, given the limited areas where such methods appear feasible, and the small amounts of water that can be captured, water harvesting techniques are unlikely to have a significant impact on global food production and water scarcity. 5.5. Integrated Watershed Management The watershed, or river basin, is the logical hydrologic unit that includes the key interrelationships and interdependencies of concern for land and water management as represented, for example, in the linkages between upstream and downstream water users. Upland watersheds are source areas for surface and groundwater recharge while downstream agriculture and urban development are directly dependent on water supplies from the upper watershed. In many regions, poor management of watersheds through deforestation, the eradication of perennials, and other human interventions in upland areas often leads to soil erosion and decreases in agricultural productivity, siltation of reservoirs and irrigation systems, adverse impacts on fisheries, wildlife, river habitat and recreational water uses, water pollution, flooding of lowland areas, and reductions in water supply for irrigated agriculture, hydropower, industrial and urban uses. The magnitude of these negative on- and off- site effects and their interdependencies have yet to be estimated in a comprehensive manner, but they appear to be large in many regions. Integrated watershed management requires an interdisciplinary, intersectoral and watershed-wide approach to the identification of sustainable resource utilization and management practices that allow for a more effective and sustainable exploitation of water and other natural resources. Such management approaches can improve the food supply situation in the region. Measures to improve integrated watershed management include development and dissemination of appropriate technology, such as erosion control practices and fragile lands protection, active mountain, forest range and prairie management, and adaptation of farming systems to hillsides (Easter and Hufschmidt, 1985). Policies to counteract watershed degradation should be targeted towards zones of high risk and could include public investments in research, technology development, extension services, and rural infrastructure, in order to stabilize or reverse degradation. Above all, broad policy and institutional reform should address watershed degradation through, for example, the establishment of property rights to land and forests, utilization of market incentives for appropriate resource management, and reform of regulatory, tax and subsidy policies that often encourage excessive rates of exploitation of forests and adoption of environmentally damaging farming systems. 5.6. Interbasin Water Transfers Interbasin water transfers have often been proposed as the best solution to solve acute water shortages in adjacent basins or sub-basins, particularly in arid and semiarid regions and where a large shift of water from agricultural to urban and industrial users is necessary. Plans for interbasin transfers was widespread in the 1960s and 1970s: the Soviet Union planned to divert Siberian rivers to reduce water shortages and the shrinking of the Aral Sea, at least since 1973; Chile planned to divert water from other basins for the Maipo River, to compensate for increasing water uses by the capital, Santiago, in the 1980s; the Middle Eastern countries had plans of Nile water diversions to replenish the Jordan river as early as 1902; and the U.S. planned to transfer large water quantities from Canada to the semiarid southwestern states in the 1960s. However, most of the larger-scale proposals never materialized due to huge capital costs; substantial scope for less capital-intensive alternative water savings; and increasing concerns about negative economic, environmental, and social impacts in the exporting basin, such as the potential cutting off of future development opportunities, social disruption, irreparable environmental damage, and rural-urban migration. China is an exception in that it realized several large interbasin transfers (in 1980, for example, roughly 10 BCM were diverted from the Chang into the Huai basin, and 8 BCM from the lower Huang into the Hai and Huai [Nickum, 1997]), and recently decided to carry out the proposed middle route of the South-to-North Water-Transfer Project for agricultural development on the North China Plain and for the city of Beijing. Another noteworthy transfer example is Libya, which in 1996 inaugurated the first phase of the Great Man-made River, transferring water from the arid south to the coastal region. Once completed, the network will have 5,000 km of underground pipelines with a capacity of 6.1 million cu m, providing a 50-100 year's supply, at a cost of US$25 billion (Garay and Sugheiar, 1997). Micro-level basin transfers over short distances have proven to be viable options in some regions. Several states in the U.S. have drafted interbasin legislation in recent years (London and Miley, 1990) and Texas, for example, currently has about 80 active interbasin transfer permits, typically to serve the rapidly growing cities. However, as with large-scale transfers, the potential economic and social costs in the area of origin must be taken into account. A case where the constraints on future development in the exporting basin were not considered is the purchase of water rights by the city of Los Angeles in the Owens Valley of Eastern California. This purchase had a devastating impact on the Valley, one from which it has never recovered (U.S. Office of Technology Assessment, 1993). However, interbasin transfers do not always curtail production on irrigated lands: the Metropolitan Water District in California, for example, has a 35-year contract to pay for conservation projects in the Imperial Valley in exchange for temporary use of the conserved water. In this example, the exporting basin retains the water rights and suffers no reduction in levels of water use (Postel, 1992). In summary, a portion of the growing demand for water will be met through new investments in irrigation and water supply systems, and some potential exists for expansion of non-traditional sources of water. However, in many regions, neither of these sources will be sufficient to meet the rapidly growing nonagricultural demands for water or to mitigate the effects of water transfers out of agriculture. 6. Impacts of Water Reallocation from Agriculture on Food Production 6.1. Global Impacts of Water Reallocation from Agriculture on Food Production This section explores the possible impacts on global food production of a large transfer of water away from agriculture assuming no reforms in institutions, policies, and technologies to achieve water savings and mitigate the impact of the transfer. The possible ramifications of this scenario are examined using IMPACT. This scenario is not presented as a likely outcome, but rather as an exploration of the potential effects that significant transfers of water could have on agriculture, if water savings are not simultaneously achieved through policy reform. The transfer of water from agriculture is simulated using the following assumptions: (1) no increase in irrigated area to the year 2020, corresponding to a cutback in investments and loss of existing irrigated area due to degradation and urban encroachment to balance any current pipeline investment. Under this scenario, there would be 43 million ha less irrigated area compared with the baseline projection; (2) phased-in reductions in agricultural water use over the projections period for the 37 IMPACT countries and regions, consistent with the urban and industrial demand projections described above, assuming no improvements in water use efficiencies in agriculture and slow improvements in domestic and industrial efficiencies; (3) declines in crop area growth, in proportion to the reduction in agricultural water use; and (4) reduction in crop yield growth, in proportion to changes in relative water supply, based on the relative water supply/crop yield function approach (FAO, 1979). The projected reductions in agricultural water withdrawals by 2020 are substantial, compared with the baseline 2020 values: for example, China, nearly 24 percent; India, 21 percent; and WANA, 20 percent; reductions in other developing countries range from 10 to 35 percent. This reallocation of water out of agriculture scenario shows dramatic impacts on demand in global food markets. In developing countries, yield growth for all cereals will slow from 1.20 percent annually in the baseline scenario to 1.07 percent per year, and area growth from 0.29 to 0.23 percent annually during 1993-2020. Rice is hit hardest, because it relies most heavily on irrigation water: rice yield growth will decline from 1.08 percent to 0.89 percent. The adverse impacts on production would be much higher except that, as water is being removed from production, cereal prices begin to increase rapidly, thereby depressing consumption and, simultaneously, inducing production increases, that partially offset the water-induced shortfalls. The average rice price is projected to increase by 68 percent between 1993 and 2020, to US$480 per mt and would be 85 percent higher than the projected baseline rice price in 2020; the price for wheat would increase by 50 percent; maize, 31 percent, and other coarse grains, 40 percent, compared to the baseline projections. Rising food prices depress food demand and worsen food security through widening the food supply and demand gaps described above. At the local and regional level, price increases of this magnitude would cause a significant decline in the real income of poor food consumers. Malnutrition would increase substantially, given that many of the poorest people in low-income developing countries spend more than half their income on food. Higher international prices also hurt at the national level, as poor countries will have to spend increasing resources to import a large portion of their food. Sharp price increases can fuel inflation in these countries, place severe pressure on foreign exchange reserves, and can have adverse impacts on macroeconomic stability and investment. Developing country imports will increase significantly overall, putting greater pressure on foreign exchange. In China, projected wheat imports will increase from the baseline value of 22.4 million mt in 2020 to 36.1 million mt; the country would shift from an exporting position in rice to becoming a rice importer; and total cereal imports by 2020 will increase by 76 percent, from 41.3 million mt to 72.8 million mt. In WANA, total cereal imports would increase from 65.1 million mt to 74.8 million mt. An exception is Sub-Saharan Africa, where imports by 2020 would actually decrease, because high cereal prices will severely depress demand. Although these imports of -virtual waterş would help to fill the demand gap created by reduced production due to water transfers from agriculture, the general rise in food prices will slow demand growth. This shows that a strategy of virtual water imports will have limited success if there is a general cutback in water supply to agriculture worldwide without countervailing improvements in water use efficiency and productivity. 6.2. Micro-Level Impacts of Water Reallocation Many economic studies suggest that the negative local impacts of properly managed water transfers from agriculture will be minimal, but popular perceptions (such as "draining the lifeblood of farmers") are typically more pessimistic. Transferring water out of agriculture can have impacts on a wide range of stakeholders, particularly if effective institutions to manage water transfers are not in place. Reallocation can decrease agricultural productivity and irrigated area, and change cropping patterns. In addition to direct impacts on agricultural production, water transfers can negatively affect business activities, local government fiscal capacity, and the quality of public services in areas from which water is being transferred, because of the reduction in irrigated area or production and associated reductions in agriculturally linked economic activities and in the tax base. In addition, permanent transfers of water rights may limit future economic development in the area of origin and induce out-migration (Rosegrant, 1997). Whereas the buyer and seller of water presumably gain from the transfer if the seller holds secure water rights, other parties can be negatively affected (and not compensated) through reductions in water availability and quality, and instream flows. Furthermore, the water in irrigation systems is used for a wide variety of other purposes that are often not accounted for, such as hydropower generation, fishing, gardens, rural domestic water supplies, and livestock production, all activities that would be severely affected by reallocation (Howe, Lazo and Weber, 1990; Meinzen-Dick, 1997). Microeconomic and regional analyses suggest that the severity of economic impacts on the area of origin will differ according to (a) whether or not the destination of transferred water remains within the same area of economic activity; (b) whether or not transfer proceeds are reinvested in the area of origin; (c) the economic vitality of the area of origin; and (d) the strength of backward and forward linkages of the irrigated agriculture sector (Howe, Lazo and Weber, 1990). In this section, the available (but quite limited) case study evidence on potentially adverse micro-level impacts on the area of origin of water transfers is reviewed. Whereas regional or national impacts of water transfers are usually positive overall, it is the area of origin -- usually rural areas in semiarid regions -- that may face adverse income and livelihood effects, particularly if water transfers are not appropriately managed. However, the evidence shows that the impacts of water reallocation are mixed and highly complex, and with the limited evidence available, it is difficult to fully identify the underlying conditions that determine the direction and the magnitude of these impacts. Care must also be taken in sorting out the effects of water transfers from the broader effects of dynamic change in the rural and urban economies. In many cases negative effects may not be attributable to water transfers, but rather may be the result of declining competitiveness of agriculture in a given region, with water transfers occurring as a byproduct of long-term economic change. Urbanization and water reallocation to urban areas The rapid expansion of urban areas can affect irrigation and food production in a number of ways, both negative and positive. Evidence from Chile, Indonesia, Thailand, the western U.S. and elsewhere clearly indicates that cities often occupy highly productive (irrigated) farmland; draw off skilled, young farm labor; compete with irrigation for the water sources to supply residents, industry, and power; and damage water quality for agricultural production through municipal sewage and industrial effluents (Hearne and Easter, 1995; Christensen, 1994; Kurnia, Avianto, and Bruns, 1998; Howe, 1998). On the other hand, nearby cities provide farm households with markets and income that can be used to purchase more water-efficient irrigation technology and to diversify into higher-value crops. In the suburbs of Beijing, for example, both grain output and overall agricultural output value continued to increase at the same time that water had been diverted to the urban core and the overall irrigated area declined (Nickum, 1997). Hearne (1998) reports that one significant reason for the positive experience with agriculture-urban water transfers in Chile was that urban areas serve as service centers for the local agricultural areas, and that most large irrigators have houses and businesses in these communities and do not want them to be short of water. Impacts of water reallocation from agriculture on rural communities Reallocation of water out of agriculture can have negative effects on rural employment possibilities, not only directly in the irrigation sector, but even more through multiplier effects on agriculturally related activities. Idleness of forward and backward linkages of the agricultural sector can also substantially reduce the rural tax base. It is not realistic to assume that idle human and capital resources will move quickly and without cost to new uses of equal or higher productivity. Therefore, costs of water transfer out of agriculture attributable to the area of origin should be compensated and, in the case of large transfers, measures should be undertaken for human capital to adjust (Howe, 1998). On the other hand, it has also been shown that careful reallocation of water resources can favor economic growth in both urban and rural areas, and economically-induced water transfers can increase the overall living standard of the poor. Changes in rural employment possibilities and migration to urban areas are usually based on a wide array of factors, but abandonment of irrigated farming may catalyze developments. Hamilton, Whittlesey and Halverson (1989) evaluated the minimum compensation that farmers in the Snake-Columbia river system, Idaho, would be willing to accept in a long-term option contract with a hydropower station. Such an institutional arrangement would switch the use of water resources from farmers to the utility in dry years. Results indicate that estimated hydropower benefits are 10 times greater than losses in farm income, making these contracts economically valuable. In California, indirect economic effects from water transfers using the 1991 California State Emergency Drought Water Bank were relatively small. Farmers who sold water to the Bank reduced farm operating costs by US$17.7 million, or 11 percent, and crop sales by US$77.1 million, or 20 percent. These reductions adversely affected the suppliers of farm inputs and the handlers and processors of farm outputs, but the effects were not large when compared with the agricultural economy in the selling region or with the direct benefits to farmers from the sales. Operating costs, crop sales, and agribusiness revenues dropped 2 to 3 percent in selling counties because of the Bank (Dixon, Moore, and Schechter, 1993). Chang and Griffin (1992), in a study of water trading and reallocation in the very dynamic Lower Rio Grande river basin, Texas, find that water transfers have supported the growth in the value of agricultural production in the basin. Virtually all water transferred was from agricultural to nonagricultural uses, and 45 percent of all municipal rights had been obtained by transfer from the agricultural sector by 1990. Net benefits of average agriculture-to-urban transfers were estimated at around $12,000 per 1,000 cu m of water for the cities of Edinburg and Brownsville, indicating a sizeable aggregate benefit for the 94 BCM of water transferred from agricultural to municipal uses prior to 1991. Consultations with water sellers indicated that much of the agricultural water sold would otherwise have been unused by its owners, (sometimes due to prior conversion of agricultural land to other uses). Very rapid urban and economic growth in this area and reallocation of water over short distances likely helped prevent severe negative impacts on farm households. A study of the impact of drought-related water reallocation from agriculture to urban uses in 1987-92 on a rural farming community in Mendota, California, found that irrigated cropland declined by 14 percent, farms by 26 percent (small farms by 70 percent). Agricultural land values decreased by 30 percent. Increasing reliance on lower-quality groundwater reduced yields, for example, by 37 percent in melons, and by 5 percent in staple crops. Labor demand decreased over-proportionately as compared to cropland, and farm and packing salary income declined by 14 percent. Three out of 7 wholesale produce firm went out of business in the area. City tax revenues declined both as a result of depressed business conditions and declining property values (Villarejo, 1997). Palanisami (1994) finds that farmers in Tamil Nadu, India, view water transfers from rural to urban areas positively. He reports that farmers sell water to urban residents to alleviate diverse labor problems (34 percent); to achieve higher profits (44 percent); to sell surplus water (23 percent); and to sell supplies inadequate for irrigation (9 percent). Thobani (1998) reports new employment possibilities for farmers who sold their water rights in Chile and Mexico in water-intensive companies or in the larger, more profitable farms who bought the rights. Rosegrant and Gazmuri Schleyer (1994) also find evidence suggesting that area-of-origin impacts in Chile are small and that agricultural regions have benefitted substantially from water trading and sales. Farmers mostly sell small portions of their rights and maintain agricultural production with highly efficient on-farm irrigation technology for orchard or vegetable crops. However, Hearne (1998) documents that the sale of water rights by a few farmers still can have substantial negative impacts: when remaining farmers receive less canal water as seepage increases, or when canals cannot be maintained due to the decrease in members drawing water from the canal. Sadeque (1998) illustrates that it is not always the irrigation sector that suffers from water reallocation. He shows that in rural Bangladesh, competition for the scarce water resources during the dry season has favored a transfer of water from the domestic to the irrigation sector. The increasing use of deep water table extraction technologies for irrigation by relatively wealthy farmers outcompetes the shallow hand pumps used by the landless for domestic uses, disproportionately affecting women and children, who are the water carriers. With food production being a high priority of the Bangladesh government the development of deep tube wells for irrigation has been favored to the detriment of domestic water supply. Impacts of reallocation on water quality and environmental degradation There is substantial evidence on the adverse impacts of reallocation from irrigation water to industrial uses, and the pollution of water resources with industrial effluents, poorly treated or untreated domestic and industrial sewage, agricultural chemical runoff and mining wastes has become a growing environmental concern. In the Nam Siaw Basin in Northeast Thailand, for example, discharge and seepage of wastewater from rock salt mining made water unfit for human and animal consumption, and depressed rice yields in fields irrigated from the wastewater (Wongbandit, 1994). In China, about 80 percent of the population lives in areas surrounding seven major rivers and five large lakes. Untreated municipal and industrial wastewater of 35.56 BCM is discharged in these regions; 20-30 percent of the water is polluted, and the economic loss caused by water quality degradation has been estimated at US$4 billion. In the Yellow River and tributaries, wastewater discharge is 3 BCM, and water quality has fallen below the safe drinking water standard in 60 percent of the basin (Zhang and Zhang, 1995). However, the impacts of water reallocation from agriculture to industrial and other uses are often more complex. Kurnia, Avianto, and Bruns (1998) show some of these dynamics in the case of West Java, Indonesia. In this very productive agricultural region, water conflicts, which used to arise between farmers within or between irrigation systems, have shifted to the level of conflict between various sectors. A cluster of 31 textile firms in the Ciwalengke irrigation system in Bandung District, West Java, for example, has severely compromised the availability and value of surface and groundwater for irrigation purposes, fishing, and even domestic uses. Factories have increased their water abstraction beyond their permits through illegal installation of additional intakes or pumps in the permitted intakes. In the dry season, factories (illegally) buy or rent additional water from close upstream farmers who receive some benefits, whereas downstream farmers suffer. Yield decreases from 7 to 4 tons per ha have been reported in rice fields irrigated with polluted water, and some fields have ceased to be usable. This development speeds conversion of agricultural land to other uses. However, although many farmers lose out in agricultural production, some members of the farm household work in the factories, thereby increasing their living standards, and thus do not want factory activities to cease. Evidence of reduced instream flows due to water reallocation with impacts on river habitat, instream and out-of-stream recreation and other effects has been reported in several states of the western U.S., and environmental demands on water resources are increasingly being acknowledged. California, for example, has implemented a new regulation that reduces exports from the Sacramento/San Joaquin Delta in order to meet federal water quality standards and to protect endangered species (Livingston, 1998). Hearne and Easter (1995), in a comprehensive study on water markets and water transfers in Chile, find no evidence of increased environmental degradation related to active water trading. In fact, by inducing conservation, institutional arrangements in Chile seem to help prevent environmental degradation in river basins. In addition, they postpone the need of dam and other infrastructure construction projects and their inherent potential adverse effects, and decrease soil salinization, (a phenomenon which often stems from over-watering upstream), and waterlogging through increased water conservation. In summary, the evidence of the micro-level impacts on water reallocation indicates that the experience is negative for rural communities when the transfers are above the level allowing for continued farming or other opportunities in the area-of-origin; when farmers had no incentives to sell, but water was taken anyway; and when institutions and secure legislation to adequately compensate the sellers and third parties were absent. On the other hand, when sellers receive substantial benefits, sell only part of their water, have a stake in the economic development of the urban area, can rely on secure rights to their resource, are protected by adequate institutions and organizations, and have flexible tools (such as water leases or option contracts), the reallocation experience can be positive, providing economic growth in both rural and urban areas. 7. Water Policy Reforms to Save Water and Manage Reallocation The evidence presented here indicates that a shift in the future allocation of water among competing uses is inevitable, and that the global trend will be to reduce the share of water for agricultural use. Rapid nonagricultural demand growth is unlikely to be only met through the expansion of supplies, or through nontraditional sources. The key question will be how to accomplish the reallocation of water from agriculture in a rational and equitable manner that minimizes costs and avoids the potentially large negative impacts of the many ad hoc transfers today on both the rural economies from which the water is drawn and on the future growth of food supply and demand. The potentially negative implications of intersectoral water transfers can be mitigated through comprehensive policy reforms that save water in existing uses and improve the quality of water and soils through improved water demand management. In order to achieve this, greater attention must be placed on the institutions for water allocation and on the rights of water users and incentives for efficient use. The policy instruments available for demand management include: (1) enabling conditions, that facilitate changes in the institutional and legal environment in which water is supplied and used. Policies here include reform of water rights, the privatization of utilities, and laws pertaining to water user associations (WUAs); (2) market-based incentives, which directly influence the behavior of water users by providing incentives to conserve on water use, including pricing reform and reduced subsidies on urban water consumption, water markets, effluent or pollution charges and other targeted taxes or subsidies; (3) nonmarket instruments, including restrictions, quotas, licenses, and pollution controls; and (4) direct interventions, including conservation programs, leak detection and repair programs, and investment in improved infrastructure (Bhatia, Cestti and Winpenny, 1993). The precise nature of water policy reform and the policy instruments to be deployed will vary from country to country depending on the underlying conditions such as the level of economic development and institutional capability, the relative water scarcity, and the level of agricultural intensification. The mix of policy instruments will also vary from river basin to river basin, depending on the structural development of the different sectors in the region, prevailing rights to natural resources, relative water shortages, and other basin-specific characteristics. Therefore, no single recipe for water policy reform can be applied universally, and additional research is required to design specific policies within any given country, region, and basin. However, some key elements of a demand management strategy can be identified. The process of reallocating water from agriculture can be better managed through the reform of existing administrative water management organizations, through the use of incentive systems such as volumetric water prices and markets in tradable water rights, and through the development of innovative mixed systems of water allocation. 7.1. Reform of Administrative Management Institutional reform of public irrigation agencies holds considerable promise for long-term progress in improving system performance. Possible reforms include reorganization into a semi- independent or public utility mode, applying financial viability criteria to irrigation agencies, franchising rights to operate publicly constructed irrigation facilities, and strengthening accountability mechanisms such as providing for farmer oversight of operating agencies (Rosegrant and Svendsen, 1993). Devolution of irrigation infrastructure and management to WUAs has received particular attention in recent years. In the past, turnover of the infrastructure and management of systems has often failed because of a lack of financial resources at the local level, flaws in internal structural features or external factors that affect the viability and sustainability of WUAs in managing irrigation systems. A recent review has identified some of the characteristics that appear to be associated with successful WUAs. WUAs tend to be stronger if they build upon existing social capital or patterns of cooperation. Groups are likely to be stronger if they are homogeneous in background and assets. Such associations must demonstrably improve water control and farm profitability to ensure that the benefits to farmers outweigh the costs of participation. Particularly crucial to success is a supportive policy and legal environment that includes establishment and adjudication of secure water rights, monitoring and regulation of externalities and third-party effects of irrigation, and provision of technical and organizational training and support (Meinzen-Dick, et al., 1997). The goal should not be privatization per se, but to identify the lowest level in the system at which the devolved management is efficient (subsidiary principle). Local management organizations are expected to gain greater responsibility, decision-making authority, and control over physical and financial resources. These alternative organizational forms, which replace public agency management, include Irrigation Districts, as introduced recently in Mexico, Irrigation Associations, as are being rapidly introduced in Turkey, and Irrigation and Drainage Management Companies, as established recently in Viet Nam. The critical resource needed for these organizations to be effective is local managerial capacity, including skills in interaction and negotiation with government agencies, leadership, and financial management, in addition to the standard range of technical skills required for system operation and management (Svendsen and Meinzen-Dick, 1997). 7.2. Water Rights, Markets, and Prices The primary alternative to quantity-based allocation of water is incentive-based allocation, either through volumetric water prices or through markets in transferable water rights. The empirical evidence shows that farmers are price-responsive in their use of irrigation water. The main types of responses to higher water prices are use of less water on a given crop, adoption of water-conserving irrigation technology, shifting of water applications to more water-efficient crops, and change in crop mix to higher-value crops (Rosegrant, Gazmuri Schleyer and Yadav, 1995; Gardner, 1983). In urban areas, the use of incentive-based policy instruments, such as higher water prices, secure rights to water, and devolution of services, can achieve substantial water savings and improve the delivery of services for both households and industries (Bhatia and Falkenmark, 1993; Frederick, 1993; Gomez, 1987). Attempts to establish administered efficiency prices through increases in water charges have been met with strong opposition from established irrigators because this mechanism is perceived as an expropriation of existing water use rights, that would create income and wealth losses for established irrigated farms. This makes it difficult to institute and maintain an efficiency-oriented system of administered prices. The establishment of transferable property rights would formalize existing rights to water rather than expropriate these rights, and generate income for the water right holders rather than taxing them, and is therefore politically more feasible (Rosegrant and Binswanger, 1994). Devolution of water rights from centralized bureaucratic agencies to farmers and other water users has a number of advantages. The first is empowerment of the water user, by requiring user consent to any reallocation of water and compensating the user for any water transferred. The second is security of water rights tenure provided to the water user. If well-defined rights are established, the water user can benefit from investment in water-saving technology. Third, a system of marketable rights to water induces water users to consider the full opportunity cost of water, including its value in alternative uses, thus providing incentives to economize on the use of water and gain additional income through the sale of saved water. Fourth, a properly managed system of tradable water rights provides incentives for water users to internalize (or take account of) the external costs imposed by their water use, reducing the pressure to degrade resources (Rosegrant, 1997). Market allocation can provide flexibility in response to water demand, permitting the selling and purchasing of water across sectors, across districts, and across time by opening opportunities for exchange where they are needed. The outcomes of the exchange process reflect the water scarcity condition in the area with water flowing to the uses where its marginal value is highest. Markets also provide the foundation for water leasing and option contracts, which can quickly mitigate acute, short- term urban water shortages while maintaining the agricultural production base (Michelsen and Young, 1993). Establishment of markets in tradable property rights does not imply free markets in water. Rather, the system would be one of managed trade, with institutions in place to protect against third-party effects and potential negative environmental effects that are not eliminated by the change in incentives. The law forming the basis for the allocation of water through tradable rights should be simple and comprehensive; clearly define the characteristics of water rights and the conditions and regulations governing the trade of water rights; establish and implement water right registers; delineate the roles of the government, institutions, and individuals involved in water allocation and the ways of solving conflicts between them; and provide cost-effective protection against negative third-party and environmental effects which can arise from water trades (Rosegrant, 1997). 7.3. Mixed Systems of Water Management Centralized, public administrative management on the one hand and free market allocation of water on the other hand can be seen as the polar extremes for water allocation mechanisms. However, as could be seen even in the brief summaries in the preceding sections, water allocation systems in the real world will be much more complex and diverse. Systems will be mixed both in ownership (combining aspects of public and private ownership of water supply infrastructure and water rights) and in overriding water allocation principles (combining administrative/regulatory approaches with market/incentive-based approaches). Decentralization and privatization will increasingly create systems with public ownership and management down to a certain level in the distribution system and user-based ownership below that level. For water market systems to be efficient and equitable, judicious regulation will be required. The process of water policy reform should lead to mixed water allocation systems that are responsive to local institutions and conditions. The mixed management systems that have resulted from adjudication of groundwater rights in California offer a promising model for developing countries. These diverse and decentralized management systems developed in direct response to the depletion of groundwater resources and the degradation of the environment and have resulted in the elimination of overdrafts, the impoundment of surface and imported water for aquifer replenishment, and have stopped saltwater intrusion (Blomquist, 1995). The adjudication process has resulted in a governance structure for the water basin that establishes water rights, monitoring processes, means for sanctioning violations, financing mechanisms for the governance system, procedures for adapting to changing conditions, and includes representative associations of water users (Blomquist, 1992). Central to the governance structures is a water management program which employs a combination of instruments to influence water demand, including pumping quotas (usually based on historical use), pumping charges, and transferable rights to groundwater. Key elements for the success of these governance structures are that they are agreed upon and managed by the water users; are responsive to local conditions; operate with available information and data bases rather than requiring theoretically better but unavailable information; and are adaptive to the evolving environment. These attributes make mixed systems highly appropriate for developing country conditions. 7.4. Conservation Through Appropriate Technology If improved demand management introduces incentives for water conservation, the availability of appropriate technology will be essential to generating water savings and higher crop production per unit of water. As the value of water increases, the use of more advanced technologies such as drip irrigation (utilizing low-cost plastic pipes), sprinklers, and computerized control systems, used widely in developed countries, could have promising results for developing countries. If the scarcity value of water is high enough, appropriate use of new technologies appears to offer both real water savings and real economic gains to farmers. Field application efficiencies in flood irrigation in developing countries are typically in the range of 40-60 percent. High- pressure sprinklers save on drainage losses but may not reduce consumptive use because of the high evaporative losses. Modern low- pressure, downward-sprinkling systems, however, can reduce evaporation considerably (Seckler, 1996). Surge irrigation can reduce water applications significantly. Instead of releasing water continuously down field channels, surge irrigation alternates between rows at specific intervals. The initial wetting of the channel partially seals the soil and allows water to be distributed more uniformly, reducing percolation, runoff, and evaporation. Drip irrigation also has important applications in developing countries, but it is difficult to estimate the real potential for savings from this technology. On the one hand, by directing water applications directly to the root zones, drip irrigation can significantly reduce field evaporation losses and can increase the productivity of water in areas already affected by salinity. On the other hand, drip irrigation is usually not economically viable for use on cereals, which consume by far the largest share of irrigation water in developing countries, so the scope for real water savings from introduction of drip irrigation may be limited. Technological opportunities to reduce water withdrawals also exist at the irrigation system level. In Malaysia's Muda irrigation system, real-time management of water releases from the dam, keyed to telemetric monitoring of weather and streamflow conditions, has significantly improved water use efficiency and reduced drainage to the ocean. In North Africa, modern irrigation systems using hydraulically operated diversion and measuring devices were developed as early as the late 1940s, and were employed in irrigation schemes constructed in the 1950s. Modern schemes in this region deliver water on demand to individual farmers, allowing water users to be charged according to the volume of water delivered, encouraging conservation and efficient use of water. Some of these irrigation techniques have been transferred to the Middle East, and in pilot projects to other developing countries (World Bank, 1993). Continued increases in the value of water could make these capital-intensive irrigation distribution systems more widely feasible in other regions of the world. Adoption of high technology irrigation can have somewhat paradoxical impacts on water savings, and savings on a per ha basis may be limited. In the U.S., where detailed data is available, water withdrawals per ha of irrigated area increased by 35 percent between 1960 and 1975, declined nearly 15 percent from 1975 to 1980, increased again, and in 1990 were still higher than the 1975 level. In addition, reductions in water applications will likely be offset by increased water requirements for higher-yielding crops and increasing cropping intensities (Raskin, Hansen and Margolis, 1995). However, real water savings can be achieved with improved technologies through the increase in agricultural output per unit of water applied, or conversely, through reduction in the amount of water used per unit of output. The decrease of water (and land) per unit of production can also help to save on land resources under irrigated production, another major constraint for future global food production. 8. Conclusions Water demand is projected to grow rapidly, particularly in developing countries. The increase in demand will be higher for urban and industrial uses than for agriculture. A portion of the growing demand for water will be met through new investment in irrigation and water supply systems, and some potential exists for expansion of nontraditional sources of water supply. However, supply expansion will not be sufficient to meet increasing demands. Therefore, the rapidly growing urban and industrial water demands will need to be met increasingly from water transfers out of irrigated agriculture. The management of this reallocation could determine the world's ability to feed itself. If such transfers take place without mitigating policy reforms in demand management the prices of staple cereals in global food markets could increase sharply, resulting in broadly negative impacts on low-income developing countries and the poor consumers in these countries. The reallocation of water can also have substantial negative effects on rural economies, if supporting policy measures are not adopted. The evidence of the impact of transfers of irrigation water to urban and industrial uses on rural communities is mixed. In addition, interlinkages between urban and rural sectors and the importance of local, basin-level characteristics make it difficult to draw general conclusions about the impacts of transfers. However, some observations can be made: negative effects from water transfers can be mitigated through (1) the establishment of secure rights to water that are monitored and enforced by adequate institutions and organizations; (2) transfers of relatively small amounts from many irrigators, inducing conservation measures instead of plot abandonment; (3) reinvestment of gains-from-trade in the rural communities; and (4) adequate compensation of sellers and affected third parties. Flexible tools, in particular, markets in tradable water rights, when established in a participatory and rational manner, can facilitate and mitigate the potentially adverse impacts of water transfers, creating win-win situations for both rural and urban/industrial water users. Comprehensive reforms are required to improve the incentives at each level of the water allocation process in order to improve the efficiency of agricultural water use and sustain crop yield and output growth to meet rising food demands while allowing transfers of water out of agriculture. Institutional and legal environment reforms must empower water users to make their own decisions regarding resource use, while at the same time providing a structure that reveals the real scarcity value of water. Key policy reforms include the establishment of secure water rights to users; the decentralization and privatization of water management functions to appropriate levels; the use of incentives including pricing reform (especially in urban contexts) and markets in tradable property rights; and the introduction of appropriate water- saving technologies. Failure to implement these reforms could significantly slow the growth in crop production in developing countries and could have devastating impacts on the rural poor. References Adams, W.M. 1990. How beautiful is small? Scale, control, and success in Kenyan irrigation. World Development, Vol. 18, No. 10, 1309-1323. Allan, J.A. and J.H. Court (eds.). 1996. Water, peace and the Middle East: Negotiating resources in the Jordan basin. Tauris Academic Studies, London and New York. Ayibotele, N.B. 1992. The world's water: Assessing the resource. Keynote paper at the International Conference on Water and the Environment: Development Issues for the 21st Century, January 26-31, Dublin, Ireland. Barrow, C.J. 1991. Land degradation: Development and breakdown of terrestrial environments. Cambridge University Press, New York. Bhatia, R. and M. Falkenmark. 1993. Water resource policies and the urban poor: Innovative approaches and policy imperatives. Water and Sanitation Currents. United Nations Development Programme - World Bank Water and Sanitation Program, World Bank, Washington, D.C. Bhatia, R., R. Cestti and J. Winpenny. 1995. Water conservation and reallocation: Best practice cases in improving economic efficiency and environmental quality. A World Bank-Overseas Development Institute Joint Study. World Bank, Washington, D.C. Blomquist, W. 1995. Institutions for managing groundwater basins in Southern California. In Water quantity/quality management and conflict resolution: Institutions, processes, and economic analyses, ed. A. Dinar and E.T. Loehman. Praeger, Westport, Conn., U.S.A. Blomquist, W. 1992. Dividing the waters: Governing groundwater in Southern California. Institute for Contemporary Studies (ICS) Press, San Francisco, U.S.A. Burton, M.A. and C.K. Chiza. 1997. Water, conflict and the environment: A case study from Tanzania. In Water: Economics, Management and Demand, eds. M. Kay, T Franks, and L. Smith. E & FN Spoon, London. Chang, C. and R.C. Griffin. 1992. Water marketing as a reallocative institution in Texas. Water Resources Research, Vol. 28, No. 3, pp. 879-890. Christensen, S.R. 1994. Water allocation conflicts in Thailand: An analysis of government failure. In Water conflicts. Natural Resources and Environment Program, Thailand Development Research Institute and Queen's University, Canada. Clarke, R. 1993. Water: The international crisis. MIT Press, Cambridge, Mass., U.S.A. Dixon, L.S., N.Y. Moore and S.W. Schechter. 1993. California's 1991 drought water bank: Economic impacts on the selling regions. Rand, Santa Monica, Calif., U.S.A.. Easter, K. William, M. M. Hufschmidt. 1985. Integrated watershed management researchfor developing countries : workshop report. Honolulu, Hawaii : East-West Environment and Policy Institute, East-West Center. El-Ashry, M.T. 1991. Policies for water resource management in semiarid regions. International Journal of Water Resources Development, Vol. 7, No. 4, pp. 230-234. Engelman, R. and P. LeRoy. 1993. Sustaining water: Population and the future of renewable water supplies. Population Action International, Washington, D.C. FAO (Food and Agriculture Organization of the United Nations). 1979. Yield response to water. FAO Irrigation and Drainage Paper No. 33. FAO, Rome. FAO. 1996. Food production: The critical role of water, September. Advanced edition. FAO, Rome. FAO. 1989. Food and Agriculture Organization Production Yearbook. Vol. 43, FAO, Rome. Fereres, E. and F. Cen~a. 1997. Social benefits and environmental constraints of irrigation in an era of water scarcity. In Water: Economics, Management and Demand, eds. M. Kay, T. Franks and L. Smith. E & FN Spoon, London. Franks, T., Shahwahid, M. and Lim, H.F. 1997. The wise use of mangrove systems: The social and environmental value of water. In Water: Economics, Management and Demand, eds. M. Kay, T. Franks, and L. Smith. E & FN Spoon, London. Frederick, K.D. 1993. Balancing water demands with supplies: The role of management in a world of increasing scarcity. World Bank Technical Paper No. 189. World Bank, Washington, D.C. Garay, F.J. and M.A. Sugheiar. 1997. Libya's Great Man-made River. OPEC Bulletin, Vol. 28, No. 6, pp. 20-27. Gardner, B.D. 1983. Water pricing and rent seeking in California agriculture. In Water rights, scarce resource allocation, bureaucracy, and the environment, ed. T.L. Anderson. Pacific Studies in Public Policy, Pacific Institute for Public Policy Research. Ballinger, Cambridge, Mass., U.S.A. Gleick, P.H. 1996. Basic water requirements for human activities: Meeting basic needs. Water International, Vol. 21 No. 2, pp. 83-92. Gleick, P.H. 1993. About the data. In Water in crisis: A guide to the world's fresh water resources, ed. P.H. Gleick. Oxford University Press, New York. Gomez, C. 1987. Experiences in predicting willingness-to-pay on water projects in Latin America. Paper presented at the meeting Resource Mobilization for Drinking Water and Sanitation in Developing Nations, San Juan, Puerto Rico, May 26-29. Grossman, G.M. and A.B. Krueger. 1997. Economic growth and the environment. National Bureau of Economic Research Working Paper, National Bureau of Economic Research, Cambridge, MA. Hamilton, J.R., N.K. Whittlesey and P. Halverson. 1989. Interruptible water markets in the Pacific Northwest. American Journal of Agricultural Economics Vol. 71, No. 1, pp. 63-75. Hearne, R.R. 1998. Institutional and organizational arrangements for water markets in Chile. In Markets for water: Potential and performance, eds. K.W. Easter, M.W. Rosegrant and A. Dinar, forthcoming, Kluwer Academic Publishers. Hearne, R.R. and K.W. Easter. 1995. Water allocation and water markets: An analysis of gains-from-trade in Chile. World Bank Technical Paper No. 315, World Bank, Washington, D.C. Howe, C.W. 1998. Water markets in Colorado: Past performance and needed changes. In Markets for water: Potential and performance, eds. K.W. Easter, M.W. Rosegrant and A. Dinar, forthcoming, Kluwer Academic Publishers. Howe, C.W., J.K. Lazo and K.R. Weber. 1990. The economic impact of agriculture-to-urban water transfers on the area of origin: A case study of the Arkansas river valley in Colorado. American Journal of Agricultural Economics, Vol 72, No. 12, pp. 1200-1209. IIMI (International Irrigation Management Institute). 1992. Developing environmentally sound and lasting improvements in irrigation management: The role of international research, IIMI, Colombo, Sri Lanka. Jones, W.I. 1995. The World Bank and irrigation. World Bank, Washington, D.C. Kayasseh, M. and C. Schenck. 1989. Reclamation of saline soils using calcium sulphate from the titanium industry. Ambio, Vol. 18, No. 2, pp. 124-127. Kurnia, G., T. Avianto and B. Bruns. 1998. Farmers, factories and the dynamics of water allocation in West Java. In Negotiating Water Rights, eds. R. Meinzen-Dick and B. Bruns, forthcoming, Sage Publishers, India. Livingston, M.L. 1998. Institutional requisites for efficient water markets. In Markets for water: Potential and performance, eds. K.W. Easter, M.W. Rosegrant and A. Dinar, forthcoming, Kluwer Academic Publishers. London, J.B. and H.W. Miley. 1990. The interbasin transfer of water: An issue of efficiency and equity. Water International, Vol. 15, No. 12, pp. 231-235. Meinzen-Dick, R. 1997. Valuing the multiple uses of irrigation water. In Water: Economics, Management and Demand, eds. M. Kay, T. Franks and L. Smith. E & FN Spoon, London. Meinzen-Dick, R., M. Mendoza, L. Sadoulet, G. Abiad-Shields and A. Subramanian. 1997. Sustainable water user associations: Lessons from a literature review. In User organizations for sustainable water services, eds. A. Subramanian, N.V. Jagannathan and R. Meinzen-Dick, World Bank Technical Paper No. 354, World Bank, Washington, D.C. Michelsen, A.M. and R.A. Young. 1993. Optioning agricultural water rights for urban water supplies during drought. American Journal of Agricultural Economics, Vol. 75, No. 11, pp. 1010-1020. Nickum, J.E. 1997. Issue paper on water and irrigation. Paper prepared for the project "Strategy and Action for Chinese and Global Food Security," Millennium Institute, USDA, World Bank, World Watch Institute, September, Washington, D.C. Nickum, J.E. 1994. Beijing's maturing socialist water economy. In Metropolitan water use conflicts in Asia and the Pacific, eds. J.E. Nickum and K.W. Easter. East-West Center and United Nations Centre for Regional Development, Westview Press, Oxford. Palanisami, K. 1994. Evolution of agricultural and urban water markets in Tamil Nadu, India. Arlington, Virginia: Irrigation Support Project for Asia and the Near East (ISPAN), United States Agency for International Development. Postel, S. 1996. Dividing the waters: Food security, ecosystem health, and the new politics of scarcity. World Watch Paper No. 132, World Watch Institute, Washington, D.C. Postel, S. 1992. Last oasis: Facing water scarcity. The Worldwatch Environmental Alert Series. New York: W.W. Norten. Raskin, P., H. Evan and R. Margolis. 1995. Water and sustainability: A global outlook. Polestar Series Report No. 4. Stockholm Environment Institute, Stockholm, Sweden. Repetto, R. 1986. Skimming the water: rent-seeking and the performance of public irrigation systems. Research Report No. 4, World Resources Institute, Washington, D.C. Rhoades, J.D. 1987. The problem of salt in agriculture. In Yearbook of science and the future. Encyclopaedia Britannia, Chicago, Ill., U.S.A.. Rosegrant, M.W. 1997. Water resources in the twenty-first century: Challenges and implications for action. 2020 Vision for Food, Agriculture, and the Environment Discussion Paper 20, IFPRI, Washington, D.C. Rosegrant, M.W. and N.D. Perez. 1997. Water resources development in Africa: A review and synthesis of issues, potentials, and strategies for the future. Environment and Production Technology Division Paper 28, IFPRI, Washington, D.C. Rosegrant, M.W., C. Ringler and R.V. Gerpacio. 1997. Water and land resources and global food supply. Paper prepared for the 23rd International Conference of Agricultural Economists on "Food Security, Diversification, and Resource Management: Refocusing the Role of Agriculture," Sacramento, California, August 10-16. Rosegrant, M.W., M.A. Sombilla, R.V. Gerpacio, and C. Ringler. 1997. Global food markets and US exports in the twenty-first century. Paper presented at the Illinois World Food and Sustainable Agriculture Program Conference, May, Urbana-Champaign, Illinois. Rosegrant, M.W., M. Agcaoili-Sombilla and N.D. Perez. 1995. Global Food Projections to 2020: Implications for Investment, 2020 Vision for Food, Agriculture, and the Environment Discussion Paper No. 5, IFPRI, Washington, D.C. Rosegrant, M.W., R. Gazmuri Schleyer and S. Yadav. 1995. Water policy for efficient agricultural diversification: Market-based approaches. Food Policy, Vol. 20, No. 3, pp. 203-223. Rosegrant, M.W. and R. Gazmuri Schleyer. 1994. Tradable water rights: Experiences in reforming water allocation policy. Arlington, Virginia: Irrigation Support Project for Asia and the Near East (ISPAN), United States Agency for International Development. Rosegrant, M.W. and H.P. Bingswanger. 1994. Markets in tradable water rights: Potential for efficiency gains in developing country water resource allocation. World Development, Vol. 22, No. 11, pp. 1613- 1625. Rosegrant, M.W. and P.L. Pingali. 1994. Policy and technology for rice productivity growth in Asia. Journal of International Development, Vol. 6, No. 6, pp. 665-688. Rosegrant, M.W. and M. Svendsen. 1993. Asian food production in the 1990s: Irrigation investment and management policy. Food Policy, Vol. 18, No. 2, pp. 13-32. Sadeque, S.Z. 1998. Nature's bounty or scarce commodity: Competition and consensus over groundwater use in rural Bangladesh. In Negotiating Water Rights, eds. R. Meinzen-Dick and B. Bruns, forthcoming, Sage Publishers, India. Seckler, D. 1996. The new era of water resource management: From "dry" to "wet" water savings. Research Report No. 1. International Irrigation Management Institute, Colombo, Sri Lanka. Shuval, H.I. 1990. Wastewater irrigation in developing countries: Health effects and technical solutions. Water and Sanitation Discussion Paper Series No. 2, United Nations Development Programme - World Bank Water and Sanitation Program, World Bank, Washington, D.C. Svendsen, M. and R. Meinzen-Dick. 1997. Irrigation management institutions in transition: A look back, a look forward. Irrigation and Drainage Systems, 11, pp. 139-156. Svendsen, M. and M.W. Rosegrant. 1994. Irrigation development in Southeast Asia beyond 2000: Will the future be like the past? Water International, Vol. 19, No. 1, pp. 25-35. Thobani, M. 1998. How to meet water needs in developing countries: Resolving issues in the establishment of tradable water rights. In Markets for water: Potential and performance, eds. K.W. Easter, M.W. Rosegrant and A. Dinar, forthcoming, Kluwer Academic Publishers. Tolba, M.K. 1978. Welcome address. In Proceedings of water management for arid lands in developing countries workshop. Vol. 13. Pergamon Press, Oxford. UN (United Nations). 1996. World Population Prospects: The 1996 Revision, Population Division, Department for Economic and Social Information and Policy Analysis, United Nations, New York. U.S. Office of Technology Assessment. 1993. Preparing for an uncertain climate. Government Printing Office. OTA-O-567, Washington, D.C. Accessed at <http://enso.unl.edu/ ndmc/ mitigate/ policy/ota>. Villarejo, D. 1997. Mendota Executive Summary. Accessed at <http://www.whiteknight.com/ Alliance/mendota.htm>. Wolter, H.W. 1997. FAO's strategies for water management with special reference to intersectoral issues. In Strategies for Intersectoral Water Management in Developing Countries - Challenges and Consequences for Agriculture, eds. J. Richter, P. Wolff, H. Franzen, and F. Heim. Proceedings of the International Workshop held from 6th - 10th May, 1996, in Berlin, Germany, DSE. Wongbandit, A. 1994. Water use conflicts management in the northeast: case studies of Nam Siaw and Nam Pong, in Natural Resources and Environment Program Thailand Development Research Institute and Queen's University, Canada, Water Conflicts, 2nd ed. pp. 149-238. World Bank. 1993. Water resources management. A World Bank policy paper. World Bank, Washington, D.C. World Bank. 1992. World development report 1992, World Bank, Washington, D.C. WRI (World Resources Institute). 1996. World Resources 1996-97, Oxford University Press, New York. Zhang, Q. and X. Zhang. 1995. Water issues and sustainable social development in China. Water International, Vol. 20, No. 3, pp.122- 128.
This document has been posted online by the United Nations Department of Economic and Social Affairs (DESA). Reproduction and dissemination of the document - in electronic and/or printed format - is encouraged, provided acknowledgement is made of the role of the United Nations in making it available.
Date last posted: 8 December 1999 15:15:30