In Depth - What will be China's food demand in 2025 and 2050?

In-depth Analyses
What will China's food demand be in 2025 and 2050?
While considerable research is being done to determine China's (natural) potential for producing food, less effort is being spent on assessing how much and what type of food the country will actually need. Current food demand projections for China are usually very crude. Most take into account only a medium-variant population projection and some increase in meat consumption (see the overview in Fan / Agcaoili-Sombilla, 1997). Several authors have already pointed out that China's food demand is affected by interrelated economic, demographic, social, and cultural factors with high predictive uncertainty (see, e.g., Wu and Findlay, 1997; Xiaohua, 1997; Ikegami, 1997; and Zuhui and Bolin, 1997). In this discussion we use a simple scenario method to quantify some of these uncertainties. We also use this scenario method to calculate rough estimates of China's food commodity demand in 2025 and 2050 according to three possible scenarios. For analytical purposes, it is useful to distinguish three factors that can determine a country's overall food demand. We first look into each of these factors independently and then discuss their combined effect.
Population growth
	Obviously, food demand projections have to take population growth into account. China will most likely have another 260 million people by 2025. All else being equal, this would increase the national food demand by 21%. However, we must also take into account a substantial error range in total population projections that is unavoidable because of the large size of China's population. According to the most recent projections from the UN, (mainland) China's population in 2025 might be only 1.39 billion (low variant); but it is also possible that it could reach 1.55 billion (high variant). The error range for 2050 is, of course, even larger. It is possible that China's population might be as large as 1.69 billion people in 2050 (high variant). However, it is also possible that the population with actually begin to decline after 2025, so that it would be only 1.25 billion in 2050, which is about the same size as in the mid-1990s. If we assume for the moment that per capita calorie consumption, diet preferences, and the structure of the food supply system will remain constant, we can determine China's food commodity demand using simple back-of-the-envelope calculations. According to these estimates, China's total cereal demand might increase from about 379.9 million tons in the mid-1990s to between 434.1 million tons (low variant) and 481.0 million tons (high variant) in 2025. China's rice demand (in milled equivalent) would increase from about 115.2 million tons in 1994-1996 to between 131.6 and 145.8 million tons in 2025, depending on possible lower or higher population growth. The demand for maize (the primary feed grain) would increase from 118.2 million tons in the mid-1990s to between 135.0 and 149.6 million tons in 2025. By 2050 China's wheat demand might be from 118.0 to 159.1 million tons, depending on the population projection variant (see Table 1). In other words, from the uncertainty in population projections alone, we have to expect an error range in food demand projections for China of about 12% in 2025 and of about 35% in 2050. It is also interesting that according to the low population growth variant, China's grain demand in 2050 could be lower than in 2025, because the population would actually decline after 2025.	Table 1
Increase in food calorie intake (without any change in diet patterns)
	Of course, it is unrealistic to assume that only the population will increase and nothing else will change. For instance, we certainly have to take into account that people in China will probably eat more. In our reference period of 1994-1996, China had an average food energy supply of 2,766 kcal per person per day, according to FAO estimates. This was still significantly lower than in, for instance, South Korea (3,303 kcal per person per day), Japan (2,898 kcal), Hong Kong (3,262), or the USA (3,624). With growing income we can expect that people in China will increase overall food consumption, even if they do not change their diet. (It is, by the way, a misunderstanding that an increase in food energy intake, measured in kcal per person per day, is always linked to a particular change in diet, such as an increase in meat consumption. A given level of food calorie intake can be achieved by rather different diets. For instance, Japan and Australia both have an average per capita calorie supply of about 2,900 kcal; the diet, and therefore the national commodity demand is, however, quite different.) How much would a increased food consumption in China affect overall food demand if we assume (for the moment) that all else remains constant? One simple method for estimating the possible increase is to use figures from more developed reference countries. I have therefore calculated how much China's food commodity demand would increase if the average calorie food intake were to increase to the level of Japan, South Korea, and the USA, and to the average of all developed countries. With the same level of calorie intake as in Japan, China would need only a little more food than is currently available. For instance, the total cereal demand would only increase from 379.9 to 398.0 million tons. With US levels of calorie intake, however, China's grain demand would increase to 497.7 million tons. One might argue that US levels of calorie intake are highly unlikely for China. Therefore, I have also used the average calorie intake level of all developed countries to calculate China's overall food demand. At this level, China would need some 437.2 million tons of cereals: 132.5 million tons of wheat, 145.7 million tons of rice, and some 136.0 million tons of maize (seeTable 1). In other words, if we think that China's calorie intake might increase to a level between those of Japan and the average of all developed countries, the uncertainty in demand projections would be some 10%.
Change in diet without any change in the overall level of food calorie intake
	With growing affluence people usually increase food intake and also begin to eat different types of food. There is not only the trend toward consumption of animal products (meat, fish, milk, eggs), but also a growing trend toward consumption of fruit, beer, sweeteners, and vegetables. Affluent people also tend to eat fewer pulses and roots. In general, people begin to move up the food chain, from primary products (such as grain, roots, or pulses) to secondary food commodities (such as meat, fish, or milk) and processed food (such as sweeteners). This increases the overall food demand, because much more grain is necessary when it is cycled through animals before it is eaten by humans (in the form of meat, milk or eggs), then when the grain is consumed directly by people. How would China's overall food demand increase, if people were to only change their diet? In other words, what would be the net effect of a change in diet without any increase in population and without any change in overall calorie intake? The answer to this question again depends very much on what type of diet change we assume. If China were to gradually move toward a Japanese diet, the overall cereal demand would actually decline (because with Japan's largely vegetarian diet, much less grain would be needed for feeding animals). However, if China were to move toward US diet patterns, which are based much more on consumption of animal products, a huge increase in feed grain production would be necessary. Contrary to what some authors believe, it is not clear in which direction China will move. Twenty years ago no one believed that hamburgers, ketchup, and French fries would ever become popular in France - today they are. There is currently a globalization in food preferences that is wiping out many ethnic diets. With growing affluence, China's diet might become quite Westernized. In my view, the best guess for a maximal diet change is to assume that in 2050 China will have a diet pattern similar to the current average diet pattern of all developed countries. The other "extreme" would be that China would have no diet change at all between now and the middle of next century. If China were to have a diet comparable to that of today's developed countries, its food demand would be some 55% higher due to change in diet alone (with population and food energy intake held constant). Why would there be such a huge increase, if only the diet would change? If China's diet pattern were like the current average diet pattern of all developed countries, it would have a much greater demand for meat and animal products (such as milk and milk products). This would boost the demand for feed grain. In the mid-1990s China had a total feed grain demand of about 107 million tons, mainly maize. If its average diet were equivalent to that of all developed countries, China's feed grain demand would more than triple to 365.9 million tons; the total cereal demand (including all feed and food grains) would increase to 532.3 million tons, a 40% increase to the current situation (see Table 1). Please be aware, that these calculations do not take into account population growth or a likely increase in food calorie intake; the numbers above are only the net effect of a possible change in diet.
Combined effects
So far, we have assumed that the various factors that determine China's food demand can be isolated; in reality, they cannot. China will have population growth plus some increase in food energy intake, plus a significant change in diet (which is associated with a necessary change in the food utilization structure). All three factors will likely change at the same time. To study the combined impact of these factors I have developed a simple accounting model based on the FAO food balance sheet for China. One can use it to simultaneously "apply" various population projections and different calorie intake levels, as well as several different types of diet (which determine the utilization structure of the various commodities). Moreover it is possible to change the import ratio (by commodity). One can experiment with various combinations of the three factors, such as a high-variant population projection combined with a minor diet change, or a rapid diet change with low population growth. From the large range of possible combinations, I have selected three scenarios that I consider most plausible:		Figure 1
A	Slow modernization scenario
	In this (rather pessimistic) scenario, we assume that population growth in China follows the high variant UN population projection from 1998; that is, we assume that China will have a population of 1.686 billion by 2050. Food calorie intake increases only slowly, by 5% between 1994 and 2025 and by another 5% between 2025 and 2050. The diet patterns remain unchanged. This combination of high population growth with a slowly rising level of calorie consumption yields the following results. China's overall cereal (grain) demand increases from currently 379.9 to 577.9 million tons in 2050; the domestic wheat demand increases from currently 115.2 to 175.1 million tons, and the rice demand grows from 126.6 to 192.4 million tons. In this simple linear scenario China's food commodity demand increases by 52% across the board (see Table 1).
B	Medium modernization scenario (the "vegetable and fish" variant)
	In this (probably more realistic) scenario we assume that China's population increases according to the medium variant UN population projection, which predicts a total population of 1.478 billion for the year 2050. The food calorie intake level reaches that of South Korea (3,303 kcal per person per day) in 2050. We also assume that by 2025 China's diet has become similar to that of South Korea and remains thereafter constant. South Korea has a diet typical of a developed Asian country, with relatively high consumption of fish but a lower consumption of meat than in China and high consumption of vegetables and fruit. However, contrary to Western countries, rice makes up a high percentage of direct grain consumption. As is typical for many developed countries, people in South Korea eat more sugar than in China, so that the sugar crop demand is more than three times higher. Their consumption of alcohol and milk products is also higher than in China. With growing income, people in China might very well prefer a Korean type of diet. What would be the consequences in terms of food commodity demand? China's demand for fish would more than triple, to 98 million tons in 2025 and 105 million tons in 2050. They would also need more vegetables, about 330 million tons in 2025 compared with current levels of 180 million tons. The total grain demand - taking into account feed grain - would increase moderately, from 379.9 million tons of cereals to 506 million tons in 2025 and to 538 million tons in 2050. Wheat demand would actually decline from 115 to 103 million tons in 2025. According to this scenario, China could switch to a more "developed" diet without a massive increase in feed grain demand. In fact, the total grain demand in both 2025 and 2050 would be lower than in Scenario A (the slow modernization scenario). The key to this option is lower population growth (than in Scenario A) and slightly lower consumption of meat, but significantly higher consumption of fish, vegetables, and fruits (see Table 1).
C	Rapid modernization scenario (the "meat and animal products" variant)
	It is perhaps unlikely, but we cannot rule out the possibility that China's diet might become quite meat-based in the future. If available data are correct, China has had a spectacular increase in the per capita production of livestock products (see Figure 1). Per capita meat consumption is already higher than in many other Asian countries, much higher, for instance, than in Japan or India. What would be the consequences if China were to switch to a diet typical of a developed country, with a still higher share of animal products? In this scenario we assume that China has very low population growth (low variant of 1998 UN projection). In fact, the population slowly begins to decline after 2025, so that in 2050 China has about the same population as in the mid-1990s. China's food commodity demand therefore essentially depends on the increase of per capita calorie intake and on the diet. For lack of a better choice, in this scenario we assume that China follows the average pattern (and calorie level) of all developed countries. The only modification is a change in the wheat-rice ratio. Since most of the developed countries have high levels of wheat consumption, we use China's current wheat to rice ratio in this scenario. The results of this scenario are as follows: China would need much more feed grain: the total demand for maize (food and feed) would increase from the current level of 118 million tons to 266 million tons in 2025; maize demand would then decline (because of population decline) to 251 million tons. As there would be also an increase in the demand for other coarse grains (mainly for feed), the total cereal demand in this scenario would increase to 663 million tons in 2025. The demand for wheat in 2025, however, would be almost identical to the current level of 115 million tons (because per capita wheat consumption in this diet is lower, even with the Chinese rice-wheat ratio applied). Demand for rice would even decline from the current level of 127 million tons to 85 million tons in 2025. China would also need considerably more sugar crops, but a smaller amount of vegetables. As is typical for developed countries, the demand for meat, milk, and fish would increase significantly. The domestic meat supply would have to grow from the current level of 29 million tons to 54 million tons (seeTable 1).
Conclusion: What can we learn from these scenarios?
Before we draw any conclusions, it should be made clear that this exercise is not about predicting specific levels of commodity demand in China. No one can make such predictions because there are too many political, economic, social, and cultural factors that are essentially unpredictable beyond a few years. These long-range scenarios up to 2050 should only demonstrate the relative weight of three important factors of food demand: population, level of consumption, and diet. With this modest approach, we can make the following conclusions.		Table 2
	The scenarios show that China's food demand is not as predictable as the typical single-variant models suggest. In particular, the impact of population growth and change in diet is quite different depending on which variants in population projections and diet-change assumptions are used. The food demand difference between the low and high UN 1998 population projection is in the range of 50 million tons of grain in 2025.
	Most population projections for China in recent years had to be adjusted downward. According to the most recent projection from the UN Population Division (which is typical for the newer projections), China will probably not reach the 1.5 billion mark by 2050. In the most optimistic scenario, China's population in 2050 will be about the same size as today. Even under most pessimistic assumptions, the population would remain significantly below 1.7 billion. Not too long, ago some authors predicted a total population of almost 2 billion for China. Moreover, the majority of the population increase is projected for the next 25 years; after that, China's population should level out or even decline. This would be good news for China's food prospects.
	Diet is at least as important as the number of mouths to feed. With a diet similar to that in the USA, China's grain demand would inevitably skyrocket. While China has rapidly increased its meat production in the past 15 years, its feed grain demand has grown quite moderately over the same time period. The ratio of meat to feed grain is much higher in China than in most Western countries; in other words, China produces more meat per kilogram of feed grain. Obviously, China's farmers so far have intensively used farm residues and household waste for feeding animals (especially pigs and poultry). This will most likely have to change in the future. With a further increase of meat production, China's farmers would need significantly more feed grain. However, as our scenarios show, there are alternatives. If China's diet were to include more vegetables, fruit, and fish, and less meat (as in the average diet in South Korea), the country could increase its per capita calorie intake level with only a moderate increase in overall grain demand.
If we compare these food demand scenarios with projections by other authors (see Table 2), we find that estimates from Scenarios A and B are a little lower, but well within the range of their estimates for 2020. The more extreme Scenario B (with a strong trend toward meat consumption) is some 60 million tons (or about 10%) higher than the levels projected by the World Bank.

Revision 2.0 (First revision published in 1999) - Copyright � 2011 by Gerhard K. Heilig. All rights reserved. (First revision: Copyright � 1999 by IIASA.)