Can China Feed Itself? Main Arguments.

This is a matrix for systematic screening and discussion of scientific arguments on how trends in various sectors of society will affect China's food security. Contrary to approaches that focus on biogeophysical conditions of food production - such as arable land area, water resources, and climate conditions - this matrix also points to demographic, political, economic, and administrative factors.

		Trends	Impact	Data quality	Prediction error	Intervention possibilities	Intervention costs
	Population	Massive growth unavoidable	Drives food demand	Good	Very large	Possible, but limited	Direct low, indirect high
	Change in diet	Data show rapid change in diet	Drives feed crop demand	Medium	Large	Difficult or impossible	Not very cost-effective
	Urbanization	Rapid urbanization is likely	Commercial agriculture & food industry	Very poor	Large, but overall trend is clear	Very difficult	High
	Arable land/soils	Scarcity, highly unreliable land-use data	Decline, degradation, soil loss	Poor	Large	Possible but difficult	Very high
	Water resources	Growing demand, regionally very scarce	Deficit, flooding, pollution	Mixed	Large	Possible	Very high
	Agricultural policy (grain import, land ownership, market access, food prices)	Liberalization of agriculture since 1978	Increases efficiency	Uncertain	Very large	Possible	Low
	Science and technology (in agriculture and in the food industry)	Improved science and technology in agriculture	Increases productivity	Adequate	Large	Possible	Medium
Click on links for more detailed arguments.

Color coding
	Behavioral factors	Policy / legislation
	Physical resources	Science / technology

Explanation

This matrix is an attempt to answer six important questions for each of the factors that determine China's food security:

What are the major trends in a particular sector? For instance, we discuss China's main demographic trends because population growth is a core factor of future food demand.
What is the impact of those trends on China's food security? For example, we know that diet preferences in China have changed significantly in recent years, with people eating considerably more meat, vegetables, and fruit. How will these trends affect overall food demand, and can they threaten China's food security?
How reliable are the empirical data needed to assess the impact of a particular trend? For instance, population statistics for China are known to be relatively reliable as confirmed by independent tests of the census, whereas land-use statistics are known to be unreliable due to underreporting.
How large is the error that can be expected from projecting the current trends of a particular factor for the next 50 years? For example, in the case of China, population projections could have a very large absolute error margin (in the range of 200 million people) due to the huge initial population size. On the other hand, the error range for projections of meat consumption should only be medium (or even small), because per capita meat consumption can only vary within the range of the human physiology: no one can eat 30 kg of meat per day.
Can policy measures affect a given factor easily or not? For instance, can China's planners and politicians do something to increase the area of arable land? They probably can, because from statistical data and satellite images indicate that there are still a few land reserves in remote areas (such as in the Northeast). However, it will be very difficult to build the necessary infrastructure, resettle farmers, market the product, etc. Is it possible to increase feed grain imports? It may be a hard fight within the political establishment to make the decision, but once made, it is certainly possible for China to import substantial amounts of (feed) grain.
What are the costs of policy interventions? Here the discussion should include not only economic, but also social and political costs. A drastic example explains the difference. A program to slow population growth that uses pressure for sterilization is certainly very cost-efficient in economic terms, but it can be extremely costly from a social and political point of view. On the other hand, legislation (including adequate enforcement) against urban sprawl does not cost much and could be quite effective at preventing loss of arable land.

This matrix was developed as a tool for decision makers who need to consider and evaluate the many dimensions of China's food problem, which are linked to various interest groups both inside and outside of China (farmers, consumers, industry, scientists, administrators, and politicians).
The matrix tries to avoid the "tunnel-visioned" approach of recent publications on China's food security, which have focused almost exclusively on just two issues: the scarcity of agricultural resources (arable land and water) and environmental degradation. Much less thought has been given to demand-side factors of the food equation or to the reliability problems of Chinese statistics. There is also a remarkable lack of imagination in the literature on China's food prospects when it comes to technological innovation or improvement in agricultural management (which could be a significant source for production increases).

The matrix's main intention is to promote a more adequate, multidisciplinary approach to China's food problems. I hope the reader will use this matrix as a stimulus to explore the associated data and extensive references.