How the world is fed
Between the early 1960s and the late 1990s, while world population almost doubled, the productive potential of global agriculture met the growth of effective demand. Figure 1 shows that total investment in irrigation and drainage tends to correspond to food prices. But even with the observed decline in food prices, the nutritional status of the world’s population continued to improve. Clearly, some of the early investment in agriculture paid off and productivity gains were being made. Irrigation played an important role in ensuring the needed growth in food production. Today, as the food production issue becomes less critical, concern arises over the future for large-scale irrigation in terms of its overall performance and the political and institutional viability of transferring the management of public irrigation schemes to users. The significance of non-structural irrigation and water management reform will grow as world agriculture in general is becoming more responsive to demand. These issues are taken up later in this report.
Figure 1 Food prices and investment in irrigation and drainage
This figure shows that total investment in irrigation and drainage tends to correspond with food prices. Lending does not include lending by commercial banks to private farmers; it only includes lending by the World Bank.
Per capita food consumption, expressed in kcal/person/day, is used as the indicator of food intake. The evolution of per capita food consumption in 1965 and 2030 is given in Table 2, which is based on historical data and on FAO projections for the years 2015 and 2030.
Table 2 and Figure 2 show a global food security situation that is consistently improving, with a steady increase in per capita food consumption at the global level as well as at the level of developing countries. Demand for food tends to saturate at the level of 3500 kcal/person/day. They also clearly show that per capita food consumption in sub-Saharan Africa remained disappointingly low over the last forty-five years although recent improvement trends are expected to continue. It should be noted that gains in overall food consumption are not necessarily translated into commensurate declines in the absolute numbers of undernourished people, in particular when there is high population growth.
The main source of food for the population of the world is agriculture. The term agriculture, as broadly used here, also includes livestock husbandry, managed fisheries (aquaculture) and forestry. The composition of meals changes gradually as demand for food strengthens and lifestyles change. For those that can afford it, many products that are grown out of season or are exotic now appear on their local market. What agriculture produces is driven by consumer demand, and changes in consumer preferences have an influence on the water needed for food production.
It has been estimated that unmanaged natural systems could provide food for 600 million people, one tenth of the current world population (Mazoyer and Roudart, 1998). Thus, about 90 percent of the present world population could not be sustained without agriculture. Yet, while few people live from only fishing, gathering and hunting, these unmanaged, or only loosely managed, natural food systems provide a strategically important contribution to the nutrition of indigenous people as well as to the existence and survival of many displaced, poor and marginal people. Except for marine fisheries, which are monitored, the diffuse reality of food resources directly obtained from natural ecosystems escapes most data collection and is usually not reflected in agricultural and economic statistics.
Table 2 Per capita food consumption from 1965 to 2030 (Kcal/person/day)
There is a global food security situation that is steadily improving, with a consistently increasing global level of food consumption per capita.
Figure 2 Per capita food consumption from 1965 to 2030
This figure shows a global food security situation that is consistently improving, at both global and developing country levels. The threshold of 2,700 kcal is taken as an indicator of the level of satisfaction of food security requirements.
Therefore, the bulk of global food production (cereals, oils, livestock and fish) is dependent upon a whole range of agricultural systems in which water is a critical factor of production (FAO/World Bank, 2001).
Cereals are by far the most important source of total food consumption as measured in calories. In developing countries, consumption of cereals thirty years ago was 141 kg/person/year, representing 61 percent of total calories. At present it is 173 kg/person/year and provides 56 percent of calories. Thus, cereal use has increased, but less than other components of food intake. The fact that the growth of global demand for cereals is declining reflects diet diversification, as more countries achieve higher levels of nutrition. However, it is expected that cereals will continue to still supply more than 50 percent of the food consumed in the foreseeable future.
To satisfy the cereal demand for a growing population using more cereals per capita, the annual world production of cereals grew by almost a billion tons from 0.94 billion tons in the mid-1960s to 1.89 billion tons in 1998. In the late 1990s, a slowdown in the growth of world consumption of cereals was recorded. It was, however, not caused by production constraints but by slowing demand. The annual world production of cereals is projected by FAO to increase by another billion tons from the 1998 level, to 2.8 billion tons. Within the cereal group, the relative importance of rice is expected to decline slightly, while consumption of wheat will continue to grow in per capita terms.
A large proportion of cereals are not produced for direct human consumption. Thus, of the future increment in cereal production projected by FAO, just under half will be for food, about 44 percent for animal feed, with the balance going to other uses, such as seed, industrial non-food, and waste.
One out of every five calories added to food consumption in developing countries in the last two decades originated in the group of oil crops, which includes palm oil, soybean, sunflower, groundnut, sesame and coconut. In projections towards the future, it is expected that 44 percent of additional calories may come from these products. This projection reflects the prospect, in the majority of developing countries, of only modest further growth in the direct consumption of staples such as cereals, roots and tubers, in favour of non-staples like vegetable oils. The major driving force of the world oil crops economy has been the growth of food demand in the developing countries, but additional demand growth has been experienced in the non-food industrial uses of oil and also in feed for the livestock sector. The future growth of aggregate world demand for, and production of oil crops, is expected to remain well above that of total agriculture. In terms of trade, developing countries have so far been net exporters of oil crops, but this position is likely to change as local consumption increases.
There are three sources of growth in crop production:
Since the early 1960s, land in agricultural use (arable land and land under permanent crops) in the world has increased by 12 percent to about 1.5 billion ha. This amounts to 11 percent of the globe’s land surface. During the same period, the world population nearly doubled from 3.1 billion to over 5.9 billion in 1998. By implication, arable land per person declined by 40 percent, from 0.43 ha in 1962 to 0.26 ha in 1998. As the world food system kept providing enough food for a growing population, a secular decline in the real price of food took place and the global situation of nutrition improved, both in relative terms and in absolute numbers. In the context of low food prices and consequent relatively low value of agricultural land, prime agricultural land is being converted to higher value urban and industrial uses. Also, irrigated land with inadequate or non-existing drainage infrastructure is being gradually lost to salinization that results in lowering yields. Yield increase and intensification have more than compensated for the reduction in per capita land availability.
As an example of growing crop yields, the world average grain yields doubled from 1.4 ton/ha/crop in 1962 to 2.8 ton/ha/crop in 1996. The average cropping intensity probably increased by some 5 percentage points, so that the arable land required to produce a given amount of grain declined by some 56 percent. It is expected that in future 80 percent of increased crop production in developing countries will come from intensification through higher yields, increased multiple cropping and shorter fallow periods. The remaining 20 percent would come from expansion of agricultural land in those developing countries and regions where the potential for expansion exists and where the prevailing farming systems and general demographic and socio-economic conditions favour it.
From 1998 to 2030, arable land in developing countries is projected to increase by 13 percent (120 million ha). The bulk of the projected expansion is expected to take place in sub-Saharan Africa and Latin America, with a smaller part in East Asia. The slowdown in the expansion of arable land is mainly a consequence of the projected slowdown in the growth of crop production.
Food production from the livestock sector includes meat (beef, pork, poultry, etc.), dairy products and eggs. In the last few decades, consumption of meat in developing countries has been growing at a rate of about 5 to 6 percent per year, that of milk and dairy products at 3 to 4 percent per year. Much of the growth has been occurring in a small number of countries, including such populous countries as Brazil and China. Many developing nations and whole regions, including sub-Saharan Africa and parts of the Near East/North Africa, where the need to increase protein consumption is the greatest, have not been participating in the buoyancy of the world meat sector. Worldwide, the poultry sector has been expanding fastest, and its share in the total meat output went from 13 percent in the mid-1960s to 28 percent currently. The increasing share of poultry in meat production is expected to continue in the future. The forces that made for the rapid growth of the meat sector in the past are, however, expected to weaken in the future owing to lower population growth and the deceleration of growth that follows the attainment of a fairly high level of consumption. Intensive forms of livestock production have led to a strong demand for cereals used as animal feed and production is rising steadily to meet this demand.
Average world per capita consumption of fish was at about 16.3 kg per year in 1999, with large differences ranging from countries with virtually no fish consumption to countries that reach over 100 kg per year. Per capita consumption could grow to 19 to 20 kg by 2030, raising total fish use to 150-160 million tons. Of the total world fisheries and aquaculture production in 2000 (130 million tons), some 74 percent (97 million tons) was available for direct human consumption as food fish. The remainder was reduced into fishmeal and fish oil for use in animal feeding (livestock and aquaculture) or for industrial purposes. Marine capture fisheries production, excluding aquaculture was in the range of 80 to 85 million tons per year in the 1990s. The long-term yearly sustainable yield of marine capture fisheries is estimated at no more than around 100 million tons per year; over-fishing of some species in certain parts of the worlds threatens the resource base. Achieving and sustaining these levels assumes more efficient utilization of stocks, healthier ecosystems and better conservation of critical habitats. Inland catches (excluding aquaculture) were recorded at about 7 to 8 million tons per year. However, a recent in-depth study of inland fisheries revealed that actual catches may be double this amount. It is important to note that fishery resources in many inland water bodies such as rivers and lakes are under increasing environmental threats resulting from continued trends of increasing aquatic pollution, habitat degradation, water abstraction, and other human-made pressures.
The bulk in the future increase of fish supply will have to come from aquaculture, which has been growing at a rate of 10 percent per year during the 1990s and increased its share in world fish supplies to about 27 percent. Most aquaculture development was in Asia (some 70 percent of world aquaculture production is in China). At present, aquaculture production amounts to 35 million tons, of which 21 million inland and 14 million marine. Over 90 percent of total aquaculture food fish production in 1995 came from developing countries, compared with 51 percent of terrestrial animal meat production. Fish exports from developing countries have been growing rapidly and now far exceed earnings from commodities such as coffee, cocoa, bananas or rubber. Strong growth may continue for some time, but constraints such as lack of feed stuffs and suitable sites, diseases and environmental constraints are becoming more binding. Major factors affecting both the sustainability of capture fisheries and the expansion of aquaculture will be improved management in the sector and a better understanding of aquatic ecosystems, as well as prevention and better management of environmental impacts affecting fishery resources and aquatic biodiversity.
At the global level, food production equals consumption. For individual countries and clusters of countries, production and consumption differ depending on agricultural trade. In general, the growth rates of food production in the developing countries have been below growth rates of demand, and food imports of these countries have been growing faster than their agricultural exports. For example, the net cereal imports of developing countries increased from 39 million tons in the mid-1970s to 103 million tons in 1998. Notwithstanding lower growth in the demand for cereals in the future, the dependence of developing countries on cereal imports is expected to continue to grow owing to limited potential in these countries to increase production. One production constraint is scarcity of water resources for irrigation, but inadequate access to credit and markets, and poor agricultural policy and management have also hampered production increases. The course towards a widening net trade deficit of the developing countries continues in the projections to the future: net food imports are expected to rise fairly rapidly to 198 million tons in 2015 and 265 million tons in 2030. This compares to a projected cereal production in developing countries of the order of 1,650 million tons in 2030.
Few countries pursue a policy of 100 percent food self-sufficiency, and few countries depend on imports for more than 20 percent of their food demand. A number of countries with a chronic trade balance deficit and high population growth already have difficulty in raising the foreign exchange needed to satisfy the growing demand for food imports. While in the past such a foreign exchange situation would have called for an increase in import taxes, and encouragement for local food production to supply the local market, the structural adjustment programmes and market liberalization policies implemented in the 1980s and 1990s have precluded the adoption of national policies leading in the direction of food self-sufficiency (Stiglitz, 2002). Yet, farmers in many developing countries with weak infrastructure and no access to capital and technology cannot face competition from the international market. This is particularly the case when their production competes with that of the heavily subsidized agriculture of industrial countries where the productivity of labour can be 1 000 times higher than theirs (Mazoyer and Roudart, 1997).
The term “virtual water” was coined in the 1990s in support of a trade and water policy point: for the food security of arid countries, where water is needed for domestic use and in support of the services and industrial sector, it is not necessary to use water for local food production, because the easier and economically more attractive alternative is to import food, in particular the inexpensive cereal base of the national diet. Thus, using a hydrologic perspective, trade in food was called trade in virtual water that is the water consumed to produce an agricultural commodity. For example, a crop such as wheat consumes about 1 000 to 1 500 litres of water to produce one kilogram of cereal. For poultry with a feed/meat conversion factor of 4:1, the virtual water content would be 6 000 litres per kg of poultry meat. For cattle, with a conversion factor of 10:1, the virtual water content of one kg of beef would be 15 000 litres. Table 3 in the following section gives examples of specific water needs for the production of a unit of a selection of agricultural commodities. The amount of virtual water imported by a country is a measure of the degree to which the country depends on the international market for its food supply.
Manipulation of the virtual water concept is subject to some caveats, one of which is that the water actually used by a crop may have stemmed partially or totally from rain, which is free of cost, whereas piped water definitely has a cost. In the case of meat, one has to keep in mind that free-roaming animals are efficient collectors of virtual water: in arid areas, the pasture they consumed grew on rainfall that usually would have no other use.