Jikun Huang, Scott Rozelle, and Mark w. Rosegrant jection for the Agricultural Sector of the Peoples Republic of China"(paper presented at the postconference workshop on"China's Food Economy in the 2lst Century, at the annual meetings of the American Agricultural Economics Association, Toronto, July 31, 1997). There is still much uncertainty about the current estimates of meat demand parameters, mainly because of data problems. In the published data, production statistics report a level of pork output that is more than twice as great as the level of pork consumption as reported in China's income and expenditure data. The discrepancy probably has a number of compo chers in China belie cant part of family member consumption that occurs out of the household(e.g dining in restaurants, etc. ) It is also suspected that production figures are in- flated, in part because of statistical problems(mostly a double-counting prob- lem) and in part because local officials may have an incentive to overstate pork oduction, since unlike grain, monitoring of livestock production is much more ficult and the probability of being caught for exaggerating production num n. 7 above) and are consistent with estimates used by Chinese agriculturists found in handbooks used by ministry of Agriculture officials. Officials, how- ever, told us that they believed these rates were too high. This would mean that the demand for feed and imports are overestimated. However, commercialization of China's livestock industry is occurring rapidly, which would mean version rates should increase over time (since farmers tend to feed se other nongrain feedstuffs to hogs). Hence, any overestimation in the should be eliminated at some point during the study period. Our current research 48. Huang, Rosegrant, and rozelle(n. 33 above) 49. The baseline assumptions for population growth rates in the three study decades implies an overall projection period population growth rate of 0. 89,a level slightly higher than that assumed by rosegrant, Agcaoili, and Perez(0.74) There are many reason believe that, with increasing reform, the govern ment's ability to control fertility may lessen, and future rates of popt growth may be greater than the baseline rates. Rosegrant, Agcaoili Ise an alternative rate of 1% per year. In this study's high-population scenario, it is assumed the growth rate in the first decade is 1.413; the second, 0.932; and the third, 0.844; implying an overall study period growth rate of 1.06 We present results that show the sensitivity of the conclusions to the choice of population growth rat Ine, so. According to J. Huang and C. David, " Price Policy and Agricultural ntive in China, a report submitted to the Food and Agriculture Organiza tion of the United Nations(Rome, 1995), while once far out of line with world agricultural prices, in recent years China's market prices have converged with those in international markets In an initial set of runs. we assumed constant real prices. We then simulated projected growth rates in production and demand(and us net imports of rice and other grain) in the International Food Policy Re search Institutes IMPACT model to generate projected world prices with China ntering as a significant importer. We then used these projected world prices as le baseline projections for the China projections model: world grain prices hould fall by 0.5% annually throughout the projection period. Meat prices should follow a similar trend. In this sense, the assumption is consistent with Chinas entry into the World Trade Organization where in the long run Chinese lucers will not be protected or taxed by border restrictions. Since China's urrent grain prices are almost the same as world market prices, there is also no obvious one time effect from liberalization The case would be different if Chin Copyright 1999 All rights reserved.Copyright © 1999. All rights reserved