Jikun huang, Scott rozelle and Mark w. rosegrant 747 by the projected changes in rural and urban populations, including the nticipated flows of rural residents into the cit Econometrically estimated parameters also are used for this part of the analysis. Using an Almost Ideal Demand System framework and stimated the demand mated coefficients and elasticities are discussed in detail in two articles by Huang and Rozelle, and one by Huang and H. Bouis. Expenditure elasticities are estimated so that they may vary according to the level of income. As projected incomes rise throughout the projection period, in come elasticities fall. Urban food grain income elasticities become zero in 2000 and turn negative in 2010; those for rural residents become zero in2010. Indirect grain consumption is imputed from the underlying demand equations for pork, beef and mutton, chicken, fish, eggs, and milk. De- mand parameters for the products are estimated for rural and urban resi dents. Different sets of parameters are estimated for different types of cities. These estimates are used for the first 10 years of the projection period. Following the experience of the rest of Asia, it is assumed that after 10 years the income-demand relationship for meat by rural residents will be similar to the current expenditure pattern of small town resident Similarly, during the first decade of the next century, demand patterns of urban consumers in small-and medium-sized cities will become more like those of consumers in large cities in the 1990s. Once the demand for meat and other animal products is known, the implied feed demand(and hence the overall demand for grain) is calcu- lated by applying a set of feed conversion ratios. The feeding efficiency of hogs is expected to increase slightly over time. Meat is assumed to be produced in China and to be sufficient to satisfy the demand for animal products, an assumption that is relaxed later in the analysis VI. Baseline Assumptions All simulations begin from 1993-95, the base period, Base-period data on production and utilization(discussed above)are 3-year averages cen- tered on 1994. Summaries of demand and supply factors that potentially affect the future development of China's food supply are in appendix A and B. a complete detailing of the structural elasticities and projected nographic structure of the economy can be found in Huang, Rozelle, Demand Side Assumptions nants of future food balance. Population growth peaked in China ii mi- Income growth and population growth will remain important deter late 1960s and early 1970s. Since then, fertility rates and the natural rate Copyright 1999. All rights reservedCopyright © 1999. All rights reserved