Contents - Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Economic aspects of food protein supplies in the world
Kyowa Hakko Kogyo Company, Ltd., Tokyo, Japan
The food crisis in 1973 was not just a shortage of food at a specific location but was the first truly global crisis. Since then, great efforts have been made to increase food production. However, in spite of the effort, significant improvement in the world's nutrition situation has not occurred. This could be attributed to a population increase that offsets that in food supply and to the failure to improve income distribution in developing countries.
The world-wide average supply of calories was 2,442 per capita per day in 1966-1968 and 2,617 calories in 19781980, an increase of 0.6 per cent per year. Meanwhile, the average protein supply in the world increased from 65.5 grams per capita per day to 69.8 grams, an increase of only 0.55 per cent. The ratio of animal to vegetable protein showed only a slight increase, from 33.1 per cent to 34.4 per cent (1).
On the other hand, in 1978-1980, regional differences in calorie supply ranged from a high of 3,624 in North America to a low of 2,120 in the Far East. For protein supply, the regional difference between the highest and lowest in these same regions rose by 2.11 times, and the regional difference in animal protein supply was an extremely large 9.86. What is even worse, over the years from 1966 to 1980, these regional differences have shown almost no improvement. Market prices for protein foods are higher than those for other foods, and animal protein foods are particularly expensive.
RECENT PROGRESS IN POPULATION, GROSS DOMESTIC PRODUCT, AND PROTEIN SUPPLIES
The energy and food crisis in 1973 had a great impact on trends in people's standard of living. Table 1 shows the population (2), gross domestic product (GDP) (3), and protein supply (1) in the world at intervals of three years during the period from 1966 to 1980. There was no important change in the annual growth rate of about two per cent in population after 1972-1974. However, the annual growth rate of GDP per capita per year decreased sharply from 3.40 per cent to 1.49 per cent in 1972-1974, as stagnation of the world economy continued.
TABLE 1. Population, GDP, and Protein Supplies in the World, 1966-1980 (at Three-Year Intervals)
|world total (billion $)||2,987||3,497||4,083||4,486||5,028|
|per capita per year ($)||865||958||1,057||1,100||1,155|
|world total (million tons)||82.5||88.9||94.5||101.6||110 3|
|per capita per day (g)||65.5||66.7||67.0||68.2||69.4|
|world total (million tons)||27.3||29.9||32.6||35.0||37.8|
|per capita per day (g)||21.7||22.4||23.1||23.5||23.8|
|world total (million tons)||55.2||58.9||61.9||66.6||72.5|
|per capita per day (g)||43.8||44.2||43.9||44.7||45.6|
|Animal protein ratio (%)||33.1||33.5||34.5||34.5||34.3|
a. In US dollars at 1975 prices.
Animal protein production has been at a standstill since 1972-1974. Hence, any trend towards an increase in the animal protein ratio has been choked off as well. This seems to reflect the stagnation of GDP per capita per year around the world. But, due to an increase in the vegetable protein supply, total protein supply per person per year has been maintained. Thus, in most regions except in parts of Africa, the increase in total food production is keeping up for the present. On the other hand, in 1978-1980, distribution of protein supplies per capita per day among countries and GDP per capita per year varied greatly, as shown in figure 1.
About one-third of the world's population, with a GDP per capita of less than US$400 at 1975 prices, is still receiving less than 50 grams per person per day of total protein supplies and only 10 grams of animal protein. Another third earning over US$1,000 in GDP per person per year averages 60 to 70 grams of total protein, of which 10 to 20 grams are animal protein. It should be noted, however, that the distribution discussed so far is based on the aggregate average levels for each country; if the distribution of calories and protein could be calculated on the basis of the distribution within each country, the situation would be seen to be far worse.
CORRELATIONS OF GDP WITH TOTAL PROTEIN SUPPLY AND WITH ANIMAL PROTEIN RATIO
Semi-logarithmic correlations of GDP per capita per year in US dollars at 1975 prices with total protein supply per capita per day and the animal protein ratio (1) were computed by pooling the data from 164 countries at intervals of three years during the 15 years from 1966 to 1980. The computation was done by the least-squares method, putting weight on population. The estimated correlation formulas are shown in table 2 with related coefficients.
Figure 2 presents linear regressions of total protein supply and GDP as a world standard shown in table 2 and also plots the shift of total protein supply in 29 countries with more than 25 million in population. Figure 3 shows linear regressions of animal protein ratio to GDP as another world standard. In the same manner as figure 2, it plots the shift of animal protein ratio in 29 countries. Both figures show that there are not large differences among five correlation lines at three-year intervals. Also, each correlation coefficient in table 2 shows a good fit. This means that, except for inflation during those years, the real protein price at 1975 prices has not increased by much. However, the slope of linear regression lines tends to slide to the left for this time period. This tendency is much clearer in the correlation of animal protein ratio with GDP than for total protein supply and GDP. This indicates that the demand for animal protein, the most expensive, is high, but world production does not meet it. Projecting such a tendency to the years 1990 and 2000, correlation formulas are obtained as shown in the lower section of table 2.
TABLE 2. Correlation Formulas and Related Coefficients 1966-1980 (at Three-Year Intervals)
|Y- X Formula||Z- X Formula|
X - GDP (in US dollars per capita per year at
Y - total protein supply (grams per capita per day)
Z = animal protein ratio (%)
Y= (log X - b) / a
Z = (log X - d) / c
FIG. 1. Population Distribution of Protein Supplies and GDP Per Capita in the World, 1978-1980 (Based on Average Levels in Each Country)
FIG. 2. Correlations between GDP and Total Protein Supplies, 1966-1980 (at Three-Year Intervals)
The shifts plotted in figures 2 and 3 for 29 countries during this 15-year period are varied indeed, reflecting national situations. For example, in middle-income countries such as Korea, nutritional improvement has progressed rapidly along world standard lines. In general, every country's shift seems to have a tendency to move closer towards the world standard as well. Countries on the left side of the world standard correlation lines have relatively high real prices for total protein or animal protein, while countries on the right side have lower real prices. This also reflects the situation in each country.
Thus, applying the same computation procedure to the world-wide pooling analysis, the correlation of protein supply with GDP in each country could be computed to reveal trends in the protein supply of the country. Meaningful correlations are also obtained for provinces, districts, households, and individuals of both sexes and all ages. All of these may be useful in analysing nutritional problems related to protein.
FIG. 3. Correlations between GDP and Animal Protein Ratios, 1966-1980 (at Three-Year Intervals)
PROPERTIES OF CORRELATION FORMULAS
Linear regressions for the world, which are statistically based on the average GDP per capita per year and protein supplies, may not reflect the situation in countries that have an extremely low per capita GDP per year. This may also be true for those countries with stagnant economies. Once an increase in GDP occurs, nutritional improvement will progress along world standard correlations.
Standard values of protein supplies for 1966-1968, 19781980, and towards 2000 computed by the world correlation formulas are shown in table 3. The table shows that total protein supply and animal protein supply both increase as GDP per capita per year at 1975 prices rises. A higher GDP increases animal protein supplies more than vegetable protein supplies. Vegetable protein supply reaches its peak when per capita (;DP is between US$800 and US$1,000 at 1975 price levels. After that, vegetable protein supply decreases. Therefore, this means that people prefer more animal and fish protein with improving economic development.
TABLE 3. Standard Values of Correlations between GDP and Protein Supplies
|Total Protein (g)||Animal Protein Ratio (%)||Animal Protein (g)||Vegetable Protein (g)|
When per capita GDP per year reaches US$2,000 to US$3,000 at 1975 prices, the animal to plant protein ratio amounts to 50 per cent, and the animal protein supply becomes greater than vegetable protein supplies. This occurs around the world. But the level of real' GDP per capita that corresponds to an animal protein ratio of 50 per cent tends to go up every year. Its value was US$1,859 in 1966-1968 and US$2,318 in 1978-1980, but in the year 2000 it will be US$3,451. This is serious for those who need the nutritional benefits of improved protein intake.
FUTURE PROSPECTS FOR PROTEIN DEMAND
Correlations of protein supplies with GDP provide one statistical basis for predicting present and future trends in protein supplies. They provide estimates of future protein supplies and the demand for them in order to help solve protein problems. These results are given in table 4, which shows the protein supplies in each region estimated from the change in GDP from 1971 to 1980 (3) and world population prospects (4) using correlation formulas towards 1990 and 2000 in table 2. The figures for economic classes are pooled for the related regions, and the figure for the world is given by the sums of economic classes.
The table shows two sets of estimates for protein supplies in 2000, based on differing assumptions: Case I assumes an exponential increase in world standard correlations, and case P assumes a parallel increase in the world standard correlations. Therefore, actual future protein supply and demand functions are likely to fall somewhere in between.
The total protein supply demanded for the future is estimated to increase from 108.2 million tons in 19781980 to about 180 million tons in the year 2000 for both cases I and P. Out of this figure, vegetable protein supply will increase from 71 million to 100 million tons or more, about a 45 per cent increase. Animal protein supply and demand, however, will increase from 37 million tons to about 70 million tons or more, which is nearly double. This will result in an increase in the animal protein ratio from 34 per cent to 40 per cent or more. Therefore, if the trend of the period studied were maintained until the year 2000, nutritional improvement at the world average might change more or less for the better, withstanding an increase in population. But in the Far East and especially Africa, where nutritional improvement is most needed, it is estimated that, even by the year 2000, the nutritional situation will be inadequate.
In general, demand for animal protein will increase steadily. By contrast, the world's fish catch levelled off in the 1970s, and the limit of this resource seems to have been reached, unless aquaculture is vastly expanded. Increase in livestock production will be difficult because of limited pasture lands. Thus, a future increase in animal protein production will depend mainly on the use of compound feeds. These will require land for production of coarse grains and oilseeds in competition with production of plant proteins for direct human feeding. For this, agricultural systems and yield will require drastic improvements, a global challenge.
TABLE 4. World Protein Supplies and Demand, 1978-1980 and Towards 2000
|1978-1980 Actual||2000 Estimateda|
|Total protein||Vegetable protein||Animal protein||Animal protein ratio (%)||Total protein||Vegetable protein||Animal protein||Animal protein ratio (%)|
|Case I||Case P||Case I||Case P||Case I||Case P||Case I||Case P|
|Developed market economies||27.7||11.4||16.3||58.8||36.8||34.8||12.2||11.3||24.6||23.5||66.8||67.5|
|Other developed MEs||4.6||2.4||2.3||48.6||7.5||6.6||2.7||2.7||4.8||3.9||64.2||58.7|
|Developing market economies||43.8||34.3||9.6||21.9||88.5||85.5||64.0||59.0||24.6||26.4||27.8||30.9|
|Centrally planned economies||36.7||25.5||11.2||30.5||53.8||57.7||30.7||32.0||23.1||25.7||42.9||44.5|
|Eastern Europe & USSR||13.7||6.8||6.9||50.5||16.7||18.5||6.7||6.5||9.9||12.0||59.6||64.8|
The principal figures in the protein columns are totals for the indicated regions in millions of tons. Figures in parentheses are grams per capita per day.
a. The estimates in case I allow for demand rising with rising affluence of the world population. Case P assumes demand parallel to the world increase in population.
This paper attempts to analyse statistically the economic aspects of food protein supply in the world at three-year intervals for the years 1966 to 1980. Correlations of total protein supplies and animal protein ratio with GDP per capita per year at 1975 prices are computed and given by high correlation coefficients over these years. This means that improvement in world protein supplies is closely related to an increase in real GDP per capita per year. The differences among respective correlations during three-year intervals are slight, reflecting stability of real protein prices beyond the food crisis of 1973.
Demand for animal protein in 2000 will be about double the supplies in 1978 to 1980 because of population increase and the rise in GDP per capita per year. While the increase in real price of protein may be small, it will adversely influence world protein nutrition, especially for people with a low GDP per capita per year.
Correlations of protein supplies with GDP, obtained from statistical data from 164 countries, illustrate annual shifts of protein supplies in individual countries from a global economic viewpoint in a dynamic manner. Such a statistical analysis suggests that efforts to shift the correlation slope between protein supplies and GDP to the right through
improved intra-country distribution of purchasing power will be essential to nutritional improvement in the world. If this can be achieved, increased per capita production of protein foods, especially those of animal origin, will be demanded.
I wish to express my thanks to Professor Nevin S. Scrimshaw, of the Massachusetts Institute of Technology and the United Nations University, for contributing special advice for this paper, and to the Statistics Division of FAO, the Statistical Office of the United Nations, and the Economic and Social Data Division of the World Bank for providing access to recent statistical information on food balance and economics.
1. FAO Production Yearbook, vols. 31-35, 1977-1981 (FAO, Rome).
2. Demographic Statistics, 1966-1980 (United Nations, New York).
3. Statistical Information, Statistical Office, United Nations, New York, and Economic and Social Data Division, The World Bank, Washington, D.C.
4. World Population Prospects as Assessed in 1980 ((United Nations, New York).
Hoshiai, K. "Protein Demand toward the 21st Century in Correlation with GDP." Shigen (bulletin of the Japan Resources Association), no. 200, p. 65 (1977).
- "The World Protein Demand in Future." Chem. Econ. Engin. Rev., 10 (9): 17 11978).
- "Imbalance of Essential Amino Acids: An Approach to the Problems of Protein Shortage in the World." Chem. Econ. Engin. Rev., 12 (4): 14 (1980).
- "Imbalance Situation of Essential Amino Acids in the World." Chem. Econ. Engin. Rev 13 (9): 16 (1981).
- "Present and Future Demand for Animal Feeding." Paper presented at the APRIA International Symposium on Single-Cell Proteins, 28-30 Jan. 1981, Paris.
- "Imbalance Situation of Essential Amino Acids in the World." Paper presented to the Twelfth International Congress of Nutrition, 16-21 Aug. 1981, San Diego, Calif., USA.
- "Economic Aspects of imbalance of Essential Amino Acids in World Protein Supplies." Paper presented to CHEMRAWN II, 6-10 Dec. 1982, Manila, Philippines.
Perissé, J., F, Sizaret, and P. François." The Effect of Income on the Structure of the Diet." Nutr. Newsletter, 7 (3): 1 (1969).
Contents - Previous - Next