Alternative Futures

Sub-Saharan Africa: A Troubled Region

In projections of the future, Sub-Saharan Africa’s problems stand out in stark and troubling fashion. As already noted, Sub-Saharan Africa is the only region in which the number and percentage of children who are malnourished is expected to rise, rather than fall, over the next 20 years. Most disturbing, however, is that even this baseline projection rests on assumptions that may prove optimistic. One key assumption is that production of most crops will increase at healthy rates over the next two decades. African farmers have, in fact, increased production at even more rapid rates during the past decade. They have done so mainly by cultivating more land and not, for the most part, by using more fertilizer, better practices, or improved varieties of crops. As a result, although crop yields in Sub-Saharan Africa were nearly equal to those in South Asia in the 1960s, they are now far lower, and the gap is even greater between Sub-Saharan Africa and other developing regions (Figure 13).

Figure 13. Cereal yields by region, 1967, 1997, and 2020

SOURCES: IFPRI IMPACT projections, June 2001, and FAOSTAT (www.fao.org) for historical data.

Figure 14. Malnourished children in Sub-Saharan Africa, 1997 and 2020

SOURCE: IFPRI IMPACT projections, June 2001.

Future increases will have to come increasingly from more intensive production on existing agricultural land. More intensive agricultural production will have to be accompanied by measures to maintain and replenish the soil to prevent agricultural areas from becoming degraded and better rural infrastructure to allow increased production to reach consumers in larger towns and cities. All of this will require more enlightened policies and substantial investment in agricultural inputs, such as fertilizer and irrigation, as well as in roads, clean water, and education. What if such efforts falter?

What if, perhaps because of political turmoil or the toll exacted by the AIDS epidemic or simple lack of political will to make needed investments and policy reforms, the projected increases in both crop area and yields were cut in half and social indicators such as female life expectancy and access to education and sanitation fell significantly? The results under such a scenario would be tragic. Because agriculture is such an important part of the region’s economy, per capita income would fall, instead of rising slightly as in the baseline projection. Sub-Saharan Africa’s net food import bill in 2020 would jump from US$ 6.5 billion in the baseline to US$ 11 billion. Per capita calorie consumption would decline by 11 percent compared with the 2020 baseline level, and the blow would fall with greatest impact on children. The number of malnourished children would increase to 49 million, a 50 percent increase over the level in 1997 (Figure 14). To make matters worse, such high levels of food imports may be economically and politically unsustainable. If African nations lack the foreign exchange to pay for these imports, food shortages and malnutrition would be even more catastrophic.

What kinds of transformations - in economic and agricultural growth, education, and health - will be necessary to allow Sub-Saharan Africa to battle childhood malnutrition as effectively as the rest of the developing world is projected to do? Reversing the tide in Sub-Saharan Africa enough to reduce the number of malnourished children in 2020 by one-third, from 33 to 22 million, would require an increase in total investments on roads, irrigation, clean water, education, and agricultural research of US$ 76 billion, to US$ 183 billion, between 1997 and 2020 compared with the baseline. Crop yields would have to grow at an annual rate of around 3 percent. Even more significantly, total gross domestic product (GDP) would have to grow at an annual rate of 8 to 10 percent.

Such rates of growth, while spectacular, are not unprecedented. They occurred in Asia during the Green Revolution. It is unlikely, however, that Sub-Saharan Africa will follow the same path as Asia toward rapid agricultural growth, because Sub-Saharan Africa faces different constraints. Agroclimatic constraints to production are more difficult than in much of Asia, the cost of exploiting water is higher, and transportation and communications infrastructure is far more limited than it was when the Green Revolution began in Asia. Still, the costs of not making the necessary investments in Sub-Saharan Africa will be tremendous, not only to the region but also to the rest of the world. Sub-Saharan Africa is one of the last two bastions of hunger and malnutrition (South Asia is the other), and without progress here, the world will make only a small dent in the global burden of malnutrition.

Asian Giants Not Likely to Disrupt World Food Markets

It is often suggested that India and China, home to more than a third of the globe’s people, hold the fate of global food security in their hands. Shifts in food production or consumer behavior in these countries, it is alleged, could produce enormous shifts in demand for food on global markets, with effects reverberating throughout the world.

What would happen, for instance, if rapid growth in incomes caused Indians to consume more meat, as income growth has done in other parts of the world? A simulation of these events, however, proves surprisingly reassuring. If, for instance, the average Indian responded to growing prosperity by eating more meat, per capita meat consumption in India could reach 18 kilograms in 2020, similar to levels in Indonesia and Pakistan and more than twice the projections for India in the baseline scenario. Consumption of meat in China has grown at similar rates during the past decade. This meteoric rise in meat consumption, from very low levels, would place enormous demands on Indian agriculture and might, in fact, prove self-limiting. Rising demand for meat would have to be met both by increased domestic production and by expanded imports of meat. The demand for cereal for livestock feed would put additional pressure on both domestic cereal production and cereal imports. Meat imports in 2020 would rise to 1.8 million metric tons (compared with 0.2 million metric tons in the baseline), and cereal imports to 26 million metric tons instead of 6 million metric tons in the baseline.

The surprising aspect of this scenario, however, is how little it would affect international prices for cereals or meat. International food markets appear resilient enough to accommodate India’s increasing demands. Under this scenario, beef prices on international markets would decline only by 2 percent between 1997 and 2020, rather than by 4 percent in the baseline scenario; wheat prices would decline by 3 percent, rather than by 8 percent; and maize prices would rise by 5 percent, instead of falling by 1 percent.

But other vulnerabilities also exist in India and the other Asian giant, China. Some observers believe that both countries will have great difficulty expanding agricultural production further, because of degraded land, water shortages, and growing competition with cities and industry for land and water. They believe that environmental degradation, unsustainable groundwater extraction for irrigation, and slowing agricultural investments could result in a sharper decline in agricultural growth than envisioned in the baseline scenario. What would be the impact on world food markets if yields and cultivated area in India and China increased only half as rapidly as expected? Some of the effects of this slowdown in production would tend to offset each other. Food shortages would produce higher prices, bringing on a partial rebound in food production. The agricultural slowdown also would depress economic growth and with it demand for food, particularly meat.

Even so, both India and China would be forced to turn increasingly to food imports. China’s cereal trade deficit in 2020 would nearly double from 48 million metric tons under the baseline scenario to 89 million metric tons, while India would shift from near self-sufficiency under the baseline scenario to imports of 30 million. India’s small agricultural trade surplus in 1997 would give way to a deficit of US$ 9.1 billion in 2020, while China’s agricultural trade deficit would soar to US$ 33.5 billion. Yet even rising imports of this magnitude would not throw international cereal or meat markets into disarray or provoke devastating price increases. Wheat and maize prices would be higher by about 9 percent each compared with the 2020 baseline projection, and rice prices would rise 26 percent.

Slowing agricultural growth in India and China would leave 2 million more children malnourished in 2020 in each of these two Asian giants, compared with the baseline scenario. In addition, this result assumes that China and India would be willing and able to finance an enormous increase in food imports. Should these countries resist such a growing dependence on foreign suppliers and attempt instead to protect and promote domestic production, domestic food prices could increase dramatically, further worsening child malnutrition.

Easing Population Growth’s Burden on Young Children

Ever since Malthus, many have regarded rapid population growth as one of the great underlying causes of hunger and environmental destruction. Yet birthrates have been declining in recent years, pushing estimates of future population growth downward. If population grows at rates lower than projected under the baseline, what would be the impact on the world food situation? The United Nations’ (UN) low estimate of population growth results in a total world population of 7 billion people in 2020, rather than the oft-quoted medium projection of 7.5 billion - the number used in the baseline scenario. Replacing the UN medium population growth assumptions with its low population growth assumptions in the model shows how slower population growth is likely to affect food security.

In the low population growth scenario, the number of malnourished children under the age of five in the developing world declines dramatically (Table I). Not only is the population under five smaller, but per capita consumption of calories is higher thanks to lower food prices and higher per capita incomes resulting from lower population growth. Under this scenario per capita kilocalorie consumption increases by 88 kilocalories above baseline levels. The number of malnourished children in the developing world is projected to drop by an additional 29 million, to 102 million malnourished children in 2020.

While this number is still unacceptably high, a 28 percent decline in child malnourishment is truly remarkable. It reveals the extent to which high population growth in impoverished regions adds millions of children to populations that are already highly food stressed and unable to cope with the additional burden. South Asia stands out in this analysis: under the low population growth scenario, child malnutrition in that region drops 25 percent (16 million children) below baseline 2020 levels. Slower population growth does not cure all ills, by any means. In Sub-Saharan Africa, the number of malnourished children would still increase slightly between 1997 and 2020, from 33 million to 34 million, but the increase would be far less than projected under the baseline scenario.

Feeding People and Feeding Livestock

The high levels of meat demand in developed countries and the rapid growth in meat demand in developing countries have caused concern that diverting cereals to feed livestock leads to an inadequate cereal supply for direct human consumption. Would improving feeding efficiencies (the amount of cereal required to produce a kilogram of edible meat) or dramatically reducing meat consumption in developed countries release cereals for human food consumption and significantly raise nutritional status in developing countries?

From 1967 to 1982, feed ratios were constant: as meat production increased, demand for feed increased by a directly proportionate amount. Since then, however, meat producers have required substantially less feed for each additional kilogram of meat produced. The reasons are many. The most important have probably been changes in genetics and in livestock management, including hormone use. In addition, poultry, which convert feed into meat more efficiently than livestock, have accounted for an increasing share of meat production.

Table 1 - Child malnutrition under low and medium population projections, 2020

REGION	LOW UN PROJECTION	MEDIUM UN PROJECTION	DIFFERENCE
	Millions of malnourished children
SOUTH ASIA	47.6	63.3	15.7
SOUTHEAST ASIA	10.3	14.0	3.7
EAST ASIA	6.2	8.5	2.3
SUB-SAHARAN AFRICA	33.7	39.3	5.6
LATIN AMERICA	1.5	2.5	1.0
WEST ASIA/NORTH AFRICA	3.0	4.0	1.0
DEVELOPING COUNTRIES	102.3	131.5	29.2

SOURCE: IFPRI IMPACT projections, June 2001.

Some of these phenomena are declining in importance. Consumers in Europe and North America are increasing their demand for lean meat. Leaner animals are inherently less efficient converters of feed into meat. In addition, meat production in some developing countries is shifting rapidly toward commercial enterprises, which rely more heavily on cereal feed than do backyard or small-scale producers. Baseline projections for the future, therefore, assume that demand for feed will match meat production more closely than it has in recent years.

It is possible, however, that technological changes not accounted for in the baseline scenario would cause feed ratios to fall more rapidly, continuing the trend of recent years. Under such a scenario, meat production would become cheaper and consumers would eat more of these meats. In addition, total demand for grain would fall because animals would eat less feedgrain. The price of maize, instead of holding steady, would fall by 28 percent relative to the 2020 baseline scenario.

This shift in grain prices then would set off a cascade of consequences. As demand for animal feed grows more slowly, international flows of more affordable grain - particularly maize - would shift away from countries in East and Southeast Asia that import large amounts of animal feed. The grain would flow instead toward South Asia and Sub-Saharan Africa, where these cereals are mainly eaten by people. Net imports of cereals into Sub-Saharan Africa and South Asia would increase by 89 and 81 percent, respectively, compared with the baseline scenario. The less grain is demanded to feed animals in East Asia and the Middle East, the more grain is available for consumers, particularly in Africa and South Asia.

Finally, in a surprising connection between different corners of the global economy, the change in feed ratios has an effect on the number of malnourished children. A reduction of 3 million malnourished children relative to the baseline scenario is not huge, compared to a global total of 129 million malnourished children, but it is note-worthy. The effect is most significant in Sub-Saharan Africa, the region with the most entrenched difficulties. Because maize, the most important source of animal feed, is heavily used for food in Africa, less demand for animal feed would translate directly into more food available for people. Under this scenario, there would be 1.6 million fewer malnourished children in Sub-Saharan Africa in 2020 than under the baseline scenario.

Could a radical drop in per capita meat demand in developed countries also help reduce malnutrition?

Reduced demand for meat in developed countries would affect food consumption in developing countries primarily through prices. Reduced demand for meat will directly reduce the world price of meat, making meat more affordable for consumers in developing countries. This reduced meat demand and drop in meat prices will also cause a direct reduction in meat production in both developed and developing countries, which will reduce the demand for cereals for animal feed. Reduced demand for feedgrains will in turn cause a drop in prices of cereals, inducing an increase in food demand for cereals.

A 1999 study using IMPACT explored in detail the impact of a decline in per capita meat demand in developed countries by one-half between 1993 and 2020.² It found that 2020 prices for beef, pork, sheep and goat meat, and poultry would be 22-3 I percent lower than in the baseline. The effects of reduced meat demand in developed countries on cereal prices are, however, much less pronounced. Prices for maize and other coarse grains - commonly used for feed - would be I I percent and 10 percent lower, respectively, in 2020 than in the baseline scenario. Prices of wheat and rice would only be 5 percent and I percent lower, respectively. The effects on wheat and rice are particularly small because although these grains are the primary staple cereals in developing countries, they make up only a small share of animal feed. Reduced demand for maize and other coarse grains for livestock feed in developed countries does not translate into greatly reduced prices for rice and wheat in developing countries.

With the price changes that stem from cutting per capita meat consumption by half in developed countries, each person in the developing world would on average consume about 13 percent more meat and 1.5 percent more cereal than under the baseline scenario. Falling meat production in developing countries would lead to a decline of 33 million metric tons in feed demand. This decline would offset the 13 million metric tons increase in food demand, leading to a slight drop in total demand for cereals in developing countries compared with the baseline. Because of lower cereal prices, developing-country farmers would produce 27 million metric tons less cereals than under the 2020 baseline, and cereal imports would increase by I I million metric tons.

In the end, people in developing countries would consume only 40 additional calories per capita in 2020 as a result of reduced meat consumption in developed countries. The number of malnourished children in developing countries would decline by 3.6 million children in 2020 and by 1.2 million in Sub Saharan Africa. This improvement in food security in developing countries is much smaller than many observers have predicted.

Productivity Growth and Food Prices

As already described, crop yields have been growing ever more slowly around the world. Many yield gains in recent decades have resulted from one-time advances, such as higher crop planting density, multiple harvesting, introduction of strains with greater fertilizer responsive-ness, and better management practices. Crop yields may be approaching their physical limitations in some high-yield systems, primarily in developed countries, and the maximum yield potentials of rice and maize have changed little over the past three decades. Environmental constraints add a layer of uncertainty to any prediction regarding future yields of intensively cultivated crops. Some of the world’s most productive agricultural areas now suffer environmental repercussions from intensive use of fertilizer, irrigation, and pesticides.

What would happen if a combination of increased environmental problems and reduced investment in agricultural research caused crop yields to grow more slowly than in the baseline? Alternatively, what if the world food situation seized the attention of governments, international organizations, and private firms, unlocking resources for agricultural research and irrigation such that crop yields increased more rapidly than in the baseline projection?

Figure 15. Rice prices under alternative scenarios, 1997-2020

SOURCE: IFPRI IMPACT projections, June 2001.

Figure 16. Maize prices under alternative scenarios, 1997-2020

SOURCE: IFPRI IMPACT projections, June 2001.

The low-yield scenario assumes that irrigation does not grow at all and that yield growth rates for meats, milk, and all crops decline by 50 percent from the baseline level in the developed world and by 40 percent in the developing world. The high-yield scenario assumes that irrigated area expands by I percent per year over the baseline growth rate and that yield growth rates rise by 20 percent from the baseline level in the developed world and by 40 percent in the developing world.

With slower growth in yields, supplies of food would be unable to keep pace with population and income growth at prevailing prices. Food would become scarce, resulting in sharply higher prices. In the low-yield scenario, rice would be 46 percent more expensive in 2020, compared with baseline projections, and maize prices 34 percent higher (Figures 15 and 16). Conversely, faster growth in yields would produce abundant food and falling prices. Maize prices would fall by 30 percent compared with the baseline, and rice prices would fall by 47 percent.

Changes in the growth of crop yields have dramatic effects on international cereal prices. Rice prices are particularly sensitive to slower yield growth because of the high proportion of rice produced in the developing countries that are most affected under this scenario. How fast investments in agricultural research and infrastructure grow over the next few decades - and thus how fast crop yields grow - will fundamentally determine the price of food for the poor.

Trade Liberalization Is Important

Most governments have been unwilling to turn food production over to the forces of the free market. They intervene in agriculture in countless ways to promote domestic food production, to keep domestic food prices low, or to reduce dependence on foreign suppliers. Many experts have demanded at least the partial abolition of these measures, arguing that they produce inefficiencies that leave most people worse off. Reducing agricultural trade distortions has been a major thrust of recent trade negotiations.

It is possible to simulate many of the effects of removing all agricultural subsidies and trade barriers, at least on agricultural production. Although this simulation does not capture the possible effects of trade liberalization on other parts of the economy, such as industry, it can provide some hints regarding the potential importance of trade liberalization.

Because most subsidies and trade barriers promote higher domestic production, full trade liberalization would generally cause agricultural production to fall somewhat, especially in the developed world. Full liberalization would cause moderate increases in world cereal prices above the projected baseline level in 2020. World rice prices would increase the most, by 14 percent above the baseline level in 2020, followed closely by wheat and maize, with price increases of 8 and 9 percent, respectively. Meat prices would respond to full trade liberalization with even sharper price increases above baseline levels in 2020, with sheep and goat prices rising 19 percent and beef prices rising 18 percent. Pork and poultry are less affected, with price increases of 13 percent for both commodities.

More important, trade liberalization would generate significant net economic benefits. Taking into account the benefits to producers and consumers and the tax savings resulting from removals of subsidies, liberalizing trade for the 16 commodities included in the model would generate global benefits of US$ 35.7 billion in 2020 (Table 2). Both developed regions and developing regions benefit, with the former gaining US$ 14.2 billion and the latter US$ 21.5 billion. Although these gains are not large compared with GDP, in many regions they are significant compared with the value of agricultural production. In proportion to their agricultural sectors, the biggest gainers would be Japan and South Korea (the latter is included in Other East Asia in the table). But the biggest absolute gains would be in Sub-Saharan Africa, at US$ 4.4 billion, or 10 percent of the value of production of the commodities examined here. This gain arises partly because African farmers face less competition from subsidized exports from Europe and other developed countries. It also, however, results from the removal of taxes that most African governments impose on food production and consumption. These taxes tend to discourage investment by farmers and make food more expensive for consumers.

This simulation captures only a portion of the potential effects of trade liberalization. If removing trade barriers stimulated overall economic growth by boosting industrial production, it would also have large effects on agriculture and food consumption.

Policy Choices Make a Big Difference

Most people, if given a moment to think about it, probably could assemble a reasonable list of the factors that determine how much food humankind has available to eat. Some of these forces are not under human control, such as the weather. Many others are, at least to a degree. They include the speed of population growth, the level of investment in technologies that allow increased food production, and the ability of farmers to take advantage of available methods for growing food.

Table 2 - Effects of global trade liberalization on welfare, 2020

REGION/COUNTRY		GAINS FROM GLOBAL TRADE LIBERALIZATION
		TOTAL VALUE	SHARE OF VALUE OF AGRICULTURAL PRODUCTION	SHARE OF GDP
		(BILLIONS OF US$)	(%)	(%)
World		35.7	3.0	0.07
Developed countries		14.2	3.0	0.04
	United States	4.3	2.5	0.03
	Europe (EU-15)	4.2	3.0	0.03
	Japan	3.0	22.3	0.04
Developing countries		21.5	3.0	0.14
	Latin America	3.0	2.1	0.07
	West Asia/North Africa	2.3	5.9	0.13
	Sub-Saharan Africa	4.4	10.4	1.03
	China	3.6	1.3	0.1 1
	Other East Asian countries	2.4	36.7	0.18
	India	2.1	1.9	0.14
	Other South Asian countries	1.3	3.3	0.36
	Southeast Asia	2.5	3.5	0.15

SOURCE: IFPRI IMPACT projections, June 2001.

NOTE: Data represent net welfare effects if trade in the 16 commodities in IMPACT were liberalized.

But how much of a difference do these humanly controlled factors make? Put another way, how would the world be different in 2020 as a result of a concentrated effort to improve the global food situation? How different would that be from a world in which governments, civil society, the private sector, and indeed all stakeholders devoted little attention to such issues?

Compare two alternative futures, an optimistic scenario characterized by increased attention to key drivers of food security and a pessimistic scenario characterized by relative neglect of these key drivers. In the first, economic growth accelerates by 25 percent, compared with the baseline projection, and population growth rates decline. The number of people with access to clean water and the number of women with access to secondary education both increase by 10 percent. Depending on the region, agricultural yields increase between 10 and 20 percent faster than anticipated by the baseline scenario. The area of irrigated land increases substantially.

The second alternative foresees a reversal of these developments. Instead of accelerating, economic growth slows, while populations grow more rapidly. Access to clean water and to education declines. Agricultural yields slow down, and no additional land is irrigated.

Global cereal production in the more optimistic scenario increases by 5 percent relative to the baseline scenario, whereas it decreases by 4 percent in the pessimistic scenario. Moreover, in the first scenario, food becomes much cheaper. Rice, for instance, falls in price by 44 percent compared with the 2020 baseline projection. In the pessimistic scenario, by contrast, its price rises by 45 percent. These shifts in food production and prices reverberate through world food markets. If food production should increase substantially, Asian countries would become less dependent on imports for their food supplies because they would be able to grow more of their own food. Sub-Saharan Africa, on the other hand, would import more, because food would become more affordable. If food production stagnated, on the other hand, South Asia’s net imports of food would rise by 90 percent. Sub-Saharan Africa would increasingly become priced out of world food markets; the region’s consumers would need just as much food but would not have the income to buy it.

The most dramatic result of this simulation is also the most vital for humanity. The two alternatives for the future show enormous differences in the ability of families to feed their children. Under the optimistic scenario, the number of malnourished children in developing nations would decline from 166 million in 1997 to 94 million in 2020, well below the 132 million of the baseline scenario. In China, this number would drop by 80 percent over two decades. Latin America could virtually eliminate the scourge of malnutrition. In Sub-Saharan Africa, where childhood malnutrition is now on the rise, progress would be less dramatic but still substantial. Instead of increasing, as predicted by baseline estimates, the number of malnourished children would decrease by 3 million (Figure 17). The central causes of this improvement in child malnutrition are threefold: broad-based and rapid agricultural productivity and economic growth, reduction in population growth rates, and increased investment in education and health. Each factor accounts for about one-third of the improvement in childhood malnutrition.

Figure 17. Malnourished children in 1997 and 2020 under alternative scenarios

SOURCE: IFPRI IMPACT projections, June 2001.

The pessimistic scenario, on the other hand, depicts a slowly unfolding catastrophe. Instead of substantial declines in childhood malnutrition over the next 20 years, the problem would become worse, especially in Sub-Saharan Africa, where the number of malnourished children would increase from 33 million to 49 million. On a global scale, the developing countries would be home to 178 million malnourished children, compared with 94 million in the optimistic scenario. This is the human price of economic and agricultural failure.