3.8 Issues of food-population balance beyond the year 2010

Contents - Previous - Next

General considerations

Concerns are often expressed about the possible adverse developments in the longer-term world food-population balance in the face of: (a) ever increasing population; (b) the need for food supplies to increase faster than population in order to ensure growth in per caput supplies of the countries and population groups with very low and nutritionally inadequate levels; (c) finite agricultural resources, meaning ever declining per caput land and water quantities; (d) degradation of the productive potential of these resources and other adverse environmental effects associated with the process of an expanding and more intensive agriculture; and (e) uncertainties regarding the progress of technology and its potential to generate further growth in yields.

Addressing these issues requires a notion of the magnitudes involved, above all what is the total agricultural production which would be required to ensure that, by a given future year, all countries will have per caput food supplies compatible with an "acceptable" degree of food security. Moreover, estimates of such future magnitudes are needed at a fairly disaggregated level in order to capture the essence of the problem which is the need to increase food supplies in the countries and regions with high population growth rates and low levels of per caput food supplies at present or projected for 2010. This is the subject of this section.

The time horizon of this study is far too short for addressing how the longer term food-population balance may evolve. But some of the findings may help put the examination of the relevant issues in a proper perspective.

Relevant insights from the analyses to 2010

Slowdown in the growth rate of agriculture

The main relevant finding is that world demand for the products of agriculture may grow in the future less rapidly than in the past. The reasons why this may be so are: (a) the progressive slowdown in world population growth; and (b) the fact that a higher proportion of world population is today well fed compared with the past, thus allowing for less scope for further increases in their per caput consumption.

In addition to these two universal forces, the period to 2010 exhibits some particular characteristics which play a role in the projected demand slowdown at the world level. These are: (a) the potential savings from more efficient utilization of total food supplies in the ex-CPEs; (b) the prospect that the most populous region of the world, East Asia, would not need to maintain in the future the very fast growth of consumption it experienced in the last 20 years; and (c) many countries and population groups with low per caput food consumption levels are unlikely to experience sufficient improvements in incomes that would allow them to increase their demand as fast as required for achieving nutritionally adequate levels of food consumption by the year 2010.

Since at the world level demand equals production, the latter's growth rate is also required to be lower in the future compared with the past (1.8 percent p.a. in 1988/90-2010, compared with 3.0 percent in the 1960s, 2.3 percent in the 1970s and 2.0 percent in 1980-92). The projected growth rate in world production is disaggregated as follows: from 1.4 percent p.a. in 1970-90 to 0.7 percent p.a. in 1988/90-2010 for the developed countries as a whole; and from 3.3 to 2.6 percent for the developing countries. It was concluded that for the developed countries, a growth rate half as large as that of the past is unlikely to put much strain on their productive potential; indeed there exists the possibility that production growth would tend to run ahead of that of effective demand unless policy reforms allowed a larger role for market forces, rather than support and protection policies, in determining the volumes of production. There is greater uncertainty on these issues for the developing countries. It is for this reason that a detailed evaluation and analysis of their potential for increasing production in terms of land and water resources and of growth in yields, all for the different agroecological zones, was undertaken. The conclusion is that a combination of land expansion not unlike that of the past (bringing into crop production some 90 million ha out of the existing 1.8 billion ha of land with rainfed crop production potential but not in agricultural uses), mainly in the land-abundant regions of sub-Saharan Africa and Latin
America/Caribbean, and significantly lower growth rates than in the past for irrigation, yields and fertilizer use would deliver the 2.6 percent p.a. growth rate of production.

Table 3.20 Developing countries, probable evolution of net agricultural trade balance

(million $)
(million $)
Net balance
(million $)
Likely changes in
net balance, in real
terms, 1988/90-2010
Coffee 8110 566 7 544 +24%
Oilseeds, veg. oils, oilmeals 13416 9776 3640 +50%
Sugar 7636 4 392 3 244 decline
Rubber 4382 1458 2924 +35%
Cocoa 2871 660 2 211 +24%
Citrus 2257 598 1659 +10 20%
Bananas 2091 164 1927 +33%
Other fruit 4086 2097 1989 +100 150%
Vegetables 3676 1920 1756 +50-70%
Tea 2163 1 108 1055 +20%
Spices 1150 580 570 increase, modest
Cassava/other roots 1424 525 899 - 40%
Veg. fibres, excl. cotton 208 117 91 zero or decline
Tobacco 3696 3 688 8 perhaps zero
Other products        
(unspecified) 10495 7997 2498  
Sub-total 67661 35646 32015  
Cereals 5977 21939 - 15962 +80%
Dairy products 457 5 805 - 5 348 + 55%
Meat, eggs 6025 7162 - 1 137 + 100% or more
Animal fats 40 729 - 689 increase
Hides+skins, excl. leather 745 2292 - 1547 prod.
Wool, excl. wool textiles 504 1421 - 917 increase, probably
Cotton, excl. cotton textiles 3 797 4062 - 265 large
Beverages (mostly alcoholic) 1377 2329 - 952  
Pulses 926 1120 -194 +100%
Sub-total 19 848 46 859 - 27011  
Grand total 87509 82505 5004  

Production growth, environmental resources and sustainability

Lack of comprehensive empirical data on the relationship between the growth of output and the generation of adverse environmental effects precludes the making of any definitive statements in answer to the question whether this growth rate of production and the associated land expansion and more intensification are compatible with sustainability. The conclusion is that more production means more pressures on environmental resources, because almost by definition production of more food for the growing population and for increasing per caput supplies cannot be achieved while leaving the environment intact; for example, converting natural habitat to grow crops or graze livestock or cementing it over to create rural infrastructure subtracts from the total stock of environmental resources. Whether this threatens sustainability, i.e. the potential for increasing food production to the level when all people are well fed and maintaining them at that level, notionally ad infinitum, is another question. In principle, it is possible to have reduction of environmental resources without threatening sustainability as defined above. It all depends on relative magnitudes, i.e. on the relative reduction of environmental resources that may come about in the process of transition from the present situation to a steady state of world agricultural production when maintenance of total output rather than continuous growth is required. The issue is whether this transition can be achieved at acceptable environmental costs, that is costs in terms of losses of environmental resources that would not threaten: (a) the sustainability of agricultural production per se; and (b) other essential functions of the environment that have an inherent life-support role (e.g. carbon cycle, biodiversity) or have a high ranking in society's hierarchies (e.g. amenity value or mere existence value).

In considering these aspects of long-term development, account must be taken of the fact that, in a fundamental sense, food production is different from other economic sectors because per caput consumption is subject to physiological limits. Once all people have "acceptable" levels of consumption (from the standpoint of nutrition, tastes, social considerations, etc.) world output would need only grow at the rate of population. If the latter is ever declining and tends to zero, so will the growth rate of world agricultural output. The key issue from the standpoint of the world's food production capacity, the environment and sustainability is whether that state of no growth in world output can be reached with: (a) enough environment resources left intact for them to perform their essential life-support functions, notionally ad infinitum; and (b) whether even a non-growing world food output much larger than at present (passing through phases requiring it to be some 45 percent above present levels by 2010 and perhaps 75 percent by 2025, see below) is a sustainable proposition. For it will probably be produced with much more intensive production methods using significantly greater quantities of external inputs like fertilizer and pesticides whose supply is finite and/or whose continuous use at high intensities is environmentally damaging.

Those who consider that even present production levels and methods are unsustainable (e.g. Pimentel et al., 1994; Ehrlich et al., 1993) will find no comfort in the proposition that the world could eventually settle in a steady state position as regards agricultural production. For the future steady-state output will need to be much larger than that of today. In parallel, the notion that the increased production can be achieved in a sustainable manner by carefully managed expansion and intensification, albeit at the cost of some reduction in environmental resources, must be supported by sufficient analysis of the path of getting from here to there. In both cases a notion of future output is necessary. The relevant estimates are attempted in what follows.

Supplies of agricultural products in the year 2025 for given per caput consumption levels

Future output will be the product of future population times per caput consumption. Given population projections, future per caput consumption can be specified by: (a) using some normative yardstick because the issue is the volume of total supplies compatible with progress in food security for all; and (b) accounting for the fact that when per caput consumption in terms of calories grows, the volume of agricultural products used up to generate these calories grows at a faster rate. This happens for two reasons: firstly, higher value products increase their share in the basket of foods consumed directly and, secondly, indirect use of agricultural products in the form of animal feed also increases more than proportionally as diets shift to more livestock products.

It can, therefore, be expected that the achievement of any target that would raise average per caput calories to "acceptable" levels would require more than proportional growth in the total availabilities of agricultural products in the countries concerned and, at the world level, also of production. However, attaching numerical values to these concepts is no simple matter. The following considerations are relevant.

1. The findings of this study project the "most likely" situation for the year 2010. This would still leave the regions of sub-Saharan Africa and South Asia with per caput food supplies of 2170 and 2450 calories respectively, that is levels which are grossly inadequate for food security. As noted earlier, this lack of sufficient progress in raising per caput food supplies in these regions is one of the reasons why world demand and production is projected to continue to slow down. However, if these year 2010 outcomes are at all credible, they must be taken as the basis from which to start measuring the further growth required to attain per caput food supply levels compatible with some target for the longer term.

2. Population projections by country are available only to the year 2025 (UN, 1993b). Therefore, an exploration of the future in the required country detail can only go to year 2025. Population projections at the country level are required because targets for future per caput food supplies can only be specified in relation to the existing gaps between present (or year 2010) levels and those that would be approximately compatible with significant reductions in chronic undernutrition. The rather arbitrary, and certainly not overgenerous, rules used here to define the targets for after 2010 and up to 2025 are as follows: in 2025 no country should have per caput calories under the current world average of 2700; countries which are projected at or above that level by the year 2010 should have further de-escalating increases subject to a maximum of 3050 calories by the year 2025; and countries which are projected to be above that maximum in 2010, just maintain their 2010 levels up to the year 2025.

3. The resulting per caput calories for 2025 are shown by region in Table 3.21, Column 6. It can be seen that under these assumptions little further growth in per caput food supplies (in terms of calories) is required at the world level (from 2880 in 2010 to 3000 in 2025) and even the required increase for the average of the developing countries is modest (from 2740 to 2900). But a quantum jump is required for sub-Saharan Africa and a smaller one for South Asia.

4. How do these targets for per caput calories translate into growth for the volume of total supplies of agricultural products measured by the conventional volume index presented earlier? In the first place, these targets must be multiplied by the year 2025 projected population which is shown in Table 3.21, Column 3. Then a notion is needed of by how much the volume of agricultural products per calorie may increase for the reasons given earlier. On this point, the empirical evidence shows that the amount of agricultural products used (all domestic uses, all agricultural products) per calorie in direct human consumption varies widely among countries. Putting the present world average at 100, the data for the individual countries span a range from around 200 in most developed countries with high average per caput calories and high shares of livestock products in their diets through about 75-100 in most developing countries with around 3000 calories but much lower livestock products component (e.g. countries in the Near East/ North Africa region) down to about 50-60 in the countries with low calories and insignificant amounts of livestock products in the diet, e.g. most countries in sub-Saharan Africa and South Asia. Taking all these country data into account, the implied average relationship over all countries of the world between per caput calories and products volume per calorie yields an elasticity of 1.6, meaning that a 10 percent rise in per caput calories would be associated with a 16 percent rise in the products used up per calorie. In principle, this relationship would represent the long-run evolutionary pattern.

However, this relationship may not be entirely appropriate for the period to 2025 examined here, and perhaps not even for well beyond this year, because it implies that if, for example, countries in South Asia were to make the transition from the present 2200 calories to that of the developed countries (3400 calories) they would be adopting the consumption patterns of the latter of over 80 kg of meat and some 630 kg of cereals (for food and feed), compared with present levels of 4 kg and 180 kg, respectively. Those developing countries which have made a good part of this transition in the last three decades, have not adopted the typical developed country basket of agricultural products. For example, a country with a typical cereals-based diet, Egypt, moved from 2290 calories in 1961/63 to 3310 calories in 1988/90 but its per caput consumption of meat grew from 10 kg to 18 kg, and that of cereals (all uses) from 250 kg to 360 kg. The same pattern is exhibited by other countries in the Near East/North Africa region. Even a fast growing country like Korea (Rep.) made the transition from 1960 calories to 2820 calories while increasing per caput consumption of meat from 4 kg to 20 kg and of cereals (all uses) from 190 kg to 350 kg.

This evidence does not, of course, mean that further structural change in favour of livestock products may not be forthcoming in these countries even if their per caput calories were not to increase appreciably. This may happen if economic growth were to raise per caput incomes towards the levels of the lower tier of the developed countries. The experience of Southern European countries, which moved to high levels of meat consumption in the last three decades, is instructive. But this type of transformation may not be in prospect for some time, if ever, for the majority of the developing countries for which the growth path to 2025 must be defined under the assumptions for per caput calories presented earlier. The more relevant experience appears to be that of those countries which achieved quantum jumps in per caput calories in the three decades 1960-90. There are 24 countries in this class, including Greece, Spain and Portugal. Their average per caput calories (simple average) increased from 2130 in 1961/63 to 3020 in 1988/90. The average amount of agricultural products per calorie was 86 in 1961/63 (always measured with world average in 1988/90 = 100) and had risen to 102 by 1988/90. Taking into account all the 30-year annual average estimates for this group of countries, it results that every 10 percent increase in per caput calories was associated with about 5.7 percent increase in the amount of agricultural products per calorie, an elasticity of about 0.6. This relationship is used here to translate the increases in direct consumption in terms of calories to increases in terms of the volume of total domestic use of agricultural products.


Contents - Previous - Next