The growth of demand will limit output growth for food over the next quarter century

The growth of demand will limit output growth for food over the next quarter century

D. Gale Johnson*

Department of Economics, University of Chicago, 1126 East 59th Street, Chicago, IL 60637

Abstract

The rate of growth of world food demand will be much slower for 1990–2010 than it was for the prior three decades. The major factor determining the increase in food demand is population growth. Income growth has a much smaller effect. From 1960 to 1990, population growth accounted for approximately three fourths of the growth in demand or use of grain. For 1990–2010, it is anticipated that population growth will account for nearly all of the increase in world demand for grain. The rate of population growth from 1990 to 2020 is projected to be at an annual rate of 1.3% compared with 1.9% for 1960 to 1990—a decline of more than 30%. World per capita use of grain will increase very little—perhaps by 4%. The increase in grain use is projected to be 40% less than in 1960–1990. It is anticipated that real grain prices will decline during the period, although not nearly as much as the 40% decline in the previous three decades. Concern has been expressed concerning the deterioration of the quality and productivity of the world’s farmland. A study for China and Indonesia indicates that there has been no significant change in the productive capacity of the land over the past 50 years. Contrary to numerous claims, the depth of the topsoil has not changed, indicating that erosion has had little or no impact.

Footnotes

↵ * To whom reprint requests should be addressed. e-mail: dg-johnson{at}uchicago.edu.
This paper was presented at the National Academy of Sciences colloquium “Plants and Population: Is There Time?” held December 5–6, 1998, at the Arnold and Mabel Beckman Center in Irvine, CA.
↵ † The same difficulty in interpreting an average applies among regions in developing countries or even between rural and urban areas in a given country. Whenever there are differences in consumption levels that are correlated with changes in the rates of growth of population, changes in per capita averages will transmit little information and may, in fact, be misleading.
↵ ‡ The per capita grain consumption in the developing countries has increased at a somewhat higher rate than production because net imports have increased at a faster rate than population. Whereas the developing countries were net importers in both 1980 and 1990, the percent increase in net imports was only moderately greater than the percent increase in population. Net grain imports per capita in 1980 were 18 kilograms and in 1990, 21 kilograms. If net grain imports of developing countries increase to 160 million to 210 million tons as projected in the three studies referred to earlier, per capita grain imports would increase to 28–36 kilograms per capita by 2010 (6). To put this amount in perspective, imports would account for about 12–15% of grain use in the developing countries in 2010.
↵ ¶ These are rough estimates of the income elasticity of demand, with 0.25 being for the developing countries. An annual growth of real per capita income of 2% would result in annual increases in the per capita consumption of grain of 0.5%. Because of the uncertainties concerning developments in grain production and incomes in the former Soviet Union it is much more difficult to estimate the income elasticity of demand for grain in the developed countries. Pinstrup-Andersen et al. (8) projected a very small increase in world per capita demand for grain from 1993 to 2020—less than 2% for the entire period, but with a significant increase in the per capita consumption of grain in the developing countries.
↵ ‖ The medium projections may well be on the high side for years some distance in the future. The reason is that in the two projections noted in the text, the demographers who make the projections must decide what to assume for the countries that now have fertility rates below replacement level. They resolved this issue by assuming that fertility will actually increase to the replacement level in the not too distant future. For example, Germany is assumed to increase its fertility level to 2.1 (actually 2.076) by 2035 compared with its recent level of 1.3 (4). China is thought to have a fertility rate of 1.9 in 1995–2000 and this is projected to increase to 2.127 in 2025–2030. I report this aspect of the two projections to make it clear that the projected levels of population do not assume that countries that have fertility below replacement levels will continue to follow the path of recent trends. Quite the contrary, it is assumed that in these countries fertility will increase over the next three decades or so and in some cases by significant percentages. So far as I know, there is no foundation for this assumption but some assumption had to be made to complete the projections.
↵ ** Per capita utilization of grain would differ somewhat from per capita production because of projected imports of grain by the developing countries. In 1990, developing countries imported 21.5 kilograms of grain per capita and they are projected to import 28–36 kilograms in 2010. If developing country imports continue to grow between 2010 and 2020 at approximately the same rate as from 1990 to 2010, then per capita consumption might be 285–290 kilograms. Given the shift in population weights, the increase in world per capita utilization would be a little less than 4%.
↵ ‡‡ I have used the U.S. wholesale price index rather than the World Bank’s index of the prices of manufactured products exported by the industrial countries to the developing countries. Over time that index increases much more than the U.S. wholesale price index and if it is used as a deflator, the declines in the real prices of grains are much greater than what is reflected in Table 1. For example, if the World Bank’s index is used, the real international price of corn declined by 48% between 1950 and 1990. In Table 1, the decline is 37%. The U.S. wholesale price index has a much broader commodity coverage than the World Bank index. The U.S. index includes oil and coal, for example.
↵ †† Erosion occurred long before man was a factor in affecting the world’s environment. And not all erosion has adverse effects on the productivity of the land. An example is the southwestern part of the state of my birth, Iowa. This part of Iowa has some of the deepest topsoil in the world, and much of it was created by erosion; it was transported by wind from Texas and Oklahoma. The soil is much more productive where it now is than if the erosion had not occurred.
↵ §§ The title of the section was “Judging Soil Quality Trends Without Measuring Them?”
↵ ¶¶ Lindert (14) adds the following: “Most importantly, the proffered data on soil quality trends are neither data nor trends. Rather they are experts’ predictions from a single snapshot, derived by combining data on slope, climate, and land use with what happens to such soils under experimental conditions. Sometimes it is refined into ”expert opinion,“ as in the GLASOD map, but it is still not based on any observation before the mid-1980s. That it is not a real history does matter, since the complexity of human soil interventions over a large countryside can defy simulation on experimental plots. Farm populations react to the soil itself with complex mixtures of crop rotations, amendments, fertilizer application, investments in water control, and sometimes neglect and mismanagement. To know the soil impact of recent human interventions, we need data on actual practice over long time spans and large areas.”
↵ § It is not implied that the 30 million tons annually has gone to importers; some part of it remained in the exporting countries. The only point is that the decline in grain production in the former U.S.S.R. has not imposed a reduction in grain utilization in the rest of the world.