Economic Development and Cultural Change fall once again to the point where China exported grain from 1992 to 1994. In the mid-1990s, the cycle repeated. Net imports reached historic highs of nearly 20 MMT in 1995 but were projected to decline in 1997. Unlike aggregate imports, trends for specific commodities have been steadier. Despite a primary reliance on domestic sources, China im ported more wheat than any other country in the world since the mid- 980s. Averaging 10 MMT of wheat per year means China accounts for 10%0-15% of the world trade(table 2, row 2). In contrast, except for 1995 and 1996(and 1989 for rice), China has exported maize and rice (rows 3 and 4). These recent trade patterns continued in 1997 IlL. Alternative Projections of Grain Demand and Supply in China Various attempts at projecting future trends in Chinas grain imports and exports have been published or are currently being used and periodically updated. The most striking feature of the projections of grain surpluses and deficits is their wide range. At one extreme, China is predicted to become a net exporter of grain. The Chinese Academy of Agricultural Sciences forecasts that China will have the capacity to export 47 MMT in the year 2000. L. Chen and A. Buckwell construct a scenario where ey argue that China can move from being an importer of about 10 MMT in the mid-1980s to a net exporter of 17 MMT in the year 2000. Other analysts believe China will eventually become a net importer of grain, some predicting imports will rise gradually, others, more sharply. The medium-term forecasts of the Economic Research Service of the United States Department of Agriculture predict China will be a moderate importer through 2005. K. Anderson et al. predict Chinas grain imports will rise to 33 MMT in the early twenty-first century. In contrast, another set of researchers predict China's grain imports will increase significantly. Other than Brown, who predicts imports could exceed 350 MMT by 2030, R. Garnaut and g. Ma project that, at per capita income growth rates of 6% to 7. 2%(rates under those experi enced between 1992 and 1994), China will require imports of between 50 to 90 MMT by 2000. C. Carter and F Zhong predict that consump- tion will outpace production, leaving a food balance deficit of more than 100 MMT by 2000. 0 Chen and Buckwell arrive at a high-growth sce- nario where China imports 59 MMT by 2000 Since all analysts are essentially forecasting from the same general base period, the predicted changes in the relative rates of growth of grain Brown projects actual declines in grain production of 0.6% per year(or a 20% decline by 2030), most of which comes from a nearly 50%o fall in sown area. Carter and Zhong project zero growth in production, while all other estimates of grain production growth are positive, ranging from 1. 1% to 1.8% for baseline or slow growth scenarios and to 2.9% for rapid growth scenarios Copyright 1999 All rights reserved
Copyright © 1999. All rights reserved
Jikun Huang, Scott Rozelle, and Mark W. Rosegrant 743 Variation in demand projections is similar. Several projections of demand growth are in the range of 1.0%0-1.7% per year, but demand growth rates well in excess of 2% are projected by Garnaut and Ma and hen and Buckwell. 5 The long-term projections from the Economic Re search Service predict much higher cereal imports; a number of the alter native scenarios forecast food balance shortfalls of 100 MMT. 6 Given the significant variation in both supply and demand projections, it is not surprising that projected net imports differ widely. The largest import projections result from highly pessimistic supply projections(Brown Carter and Zhong) and high demand projections( Garnaut and Ma; the Economic Research Service). 7 Projections by commodity are much less common. S. Fan, G. cra mer, and E. Wailes predict that rice exports will continue, mainly because of their assumption that demand elasticities will continue to be positive and rice production growth will continue at current levels. Unpublished projections by Fan and M. Agcaoili and recent long-range projections by the Economic Research Service forecast rising imports for wheat. Their estimates, however do not take into consideration some of the structural changes facing Chinas food economy. The most difficult part of evaluating current projections results that the sources of the parameters of the forecasting models, and forces behind the changes in important state variables(e.g, population and in- come growth) are not transparent. The parameters on which all of these grain projections are based (except for Carter and Zhong; and Fan, Cramer, and Wailes) are either partly or wholly synthetic. There is also little scope for assessing the effect of policy variables. With the excep- tion of the Economic Research Service and Fan and Agcaoili models, no other model can be used to systematically assess the effect of govern- mental policy. Fundamental forces in the economy, such as urbanization and market development, are ignored. Given the rapid structural change in Chinas economy-in-transition and the importance of policy in China, the omission of such important variables reduces the robustness of dictions from currently available models IV. Structural Change and Government Intervention in Chinas agriculture As Chinas economy continues to change and grow, one of the main uestions facing policy makers is how future patterns of utilization can e met most effectively. China is a country in rapid transition from a socialist system to one in which an increasing proportion of its goods and services, including food, are allocated by market forces. 2 It also is a country that is rapidly developing. There are many forces arising from these development and transition processes that will affect China's food economy. Any attempt to accurately forecast future supply and demand trends of food must account for these major economic forces Copyright 1999 All rights reserved
Copyright © 1999. All rights reserved
744 Economic Development and Cultural Change Demand Shifters: Income, Market Development, and Urbanization On the demand side, recent changes in the urban economy have made urban consumers almost entirely dependent on markets for their con- sumption needs. In this sector, prices and income changes most likely will be the fundamental force driving consumption pattern changes. Real income per capita for urban residents has risen rapidly in recent years jumping an average of more than 5% annually between 1985 and 1995 At the current average level of income for most urban residents, rice and wheat consumption rises very little with new increments in income; meat consumption, on the other hand, is still very much influenced by income Rural residents live in a different environment than their urban counterparts and exhibit different demand behavior. While rural income have grown more slowly since the mid-1980s, the demand for food grains and meat products has increased as incomes have risen. The av erage rural consumer, however, will spend less of the additional income on rice, wheat, and other staple food grains as he or she becomes richer during the development process Rural consumption markets are also less complete, but as transition improves the market environment, dietary habits may change. Farmers in many areas face limited choices in their consumption decisions since many of the products they desire on a daily basis, such as meat and fresh fruit, are not always available, even as incomes rise. In a sample of households that we drew from the national household income and expen diture survey, we found a strong and significant correlation between the level of consumption of primarily purchased goods, such as meat and fruit, and the level of market development, holding income and prices constant."Discontinuous free markets, lack of refrigeration, and gener ally high transaction costs for procuring food in rural areas affect rural consumption patterns in China. Even with rapid changes in rural mar- kets, in 1992 China's farmers still purchased only 46% of their food. As markets develop and activity on rural consumption markets increases apart from changes in income and prices, consumption patterns will be Across Asia, the behavior of consumers changes dramatically as countries urbanize 26 Urban dwellers consume less rice and demand higher levels of meats, milk products, and fish than do their rural coun- terparts, even after accounting for the differences in income and prices The ratio of urban to rural residents in China is changing fast. The urban population has grown from 19% of total population in 1980 to 28%in 1992. The effects of this population shift on consumption in China have been documented. While structural transformations of the economy should be accounted for in any predictions of future consumption pat- terns, few projections explicitly consider the differences in the consump- tion between rural and urban consumers Copyright@ 1999. All rights reserved
Copyright © 1999. All rights reserved
Jikun Huang, Scott Rozelle, and Mark w. Rosegrant 745 The case of wheat in China may unfold in an d patte when compared to China's Asian neighbors, although the dietary changes from migration will differ depending on what part of the country one is examining. Other countries in East and Southeast Asia always have experienced rising wheat demand with migration, since farmers in these countries produce and consume few wheat products. Migrants from southern rice-producing areas may be expected to follow this path since their current production and consumption patterns resemble those in neighboring countries. In contrast, city-bound migrants from north China consume very high levels of wheat, about 200 kilograms per cap ita in many northern provinces, levels that exceed or approach those of traditional wheat-producing, bread-eating nations, such as Pakistan, Tur- key, and Egypt. North China migrants will cut their consumption of wheat dramatically as they adopt the dietary patterns of urban dwellers Since the nations average rural consumption level exceeds that of urban areas(table 1), if the same magnitude of migration occurs in both the dampening net effect on wheat demand, in contrast to its other urbaniz- ing Asian neighbors 29 Supply Shifters: Technology, Investment, and Environmental Stress On the supply side, many sharp transitions are also under way. Above all, technological change needs to be considered explicitly, since it has been the engine of Chinas agricultural economy. 0 Chinas technological base grew rapidly during both the prereform and reform periods. A breakthrough in the engineering of hybrid rice, for example, pioneered by Chinese rice scientists in the 1970s, led to significantly increased yields in many parts of the country, and its use rapidly spread to nearly one-half of Chinas rice area by 1990. Wheat and maize enjoyed similar technological transformations. 3China's robust growth in the stock of re- search capital has been significantly responsible for these dramatic changes. Recent work has shown that the contribution of technology to crop growth equaled or exceeded that of the Household Responsibility System in the early reform period. Technological change contributed al most all crop growth by the late 1980s and early 1990s. 3 There is concern, however, that Chinas system may be suffering from neglect after more than a decade of reform. Real annual expendi tures on agricultural research fell between 1985 and 1990 before resum- th in 1990.The slowdown in growth in annual invest- ments in the late 1980s will result in slower growth in the overall stock of research in the 1990s and may affect production. Historic patterns of research expenditures as well as China's invest ment plans affect how the supply of rice, wheat, and maize will respond to future research expenditures. Agricultural planners have traditionally invested most heavily in rice, wheat, and maize research.36 Technological Copyright 1999 All rights reserved
Copyright © 1999. All rights reserved
746 Economic Development and Cultural Change breakthroughs and greater extension efforts in the production of rice and wheat have pushed their yields closer to their frontiers than in the case of maize. Interviews with breeders from multinational seed corporations ommonly reveal that the yield potential in all crops still exists, includ ing rice and wheat, but that it is higher in maize. A number of other factors similarly will affect future supply ment in agricultural infrastructure, especially irrigation, has been important determinant of Chinas agricultural growth in recent decades, 37 Irrigation investment and the stock of facilities have followed patterns similar to those for research, falling in the early reform period before recovering in recent years. Trends in environmental degradation, includ ing erosion, salinization, and loss of cultivated land show that it may be that considerable stress is being put on the agricultural land base. 3Ero- sion and salinization have increased since the 1970s, although in a some hat erratic pattern, and these factors have affected output of rice, wheat, maize, and other agricultural products. 9 V. A Framework for Forecasting Chinas Grain Supply and Demand The major components of this article's forecasting framework include a supply model for the rice, wheat, maize, other grain, and cash-cropping sectors of the agricultural economy, while demand models are specified separately for rural and urban consumers for rice, wheat, other grain, and six animal products. Real world price projections are generated by IMPACT, a partial equilibrium global trade model developed by M. Rose grant, M. Agcaoili, and N. Perez. o Grain supply is assumed to respond to the crop's own price, prices of other crops, quasi-fixed inputs, and the off-farm wage. Output also is a function of the stock of agricultural research, the stock of irrigation infrastructure and three environmental factors -erosion salinization and the breakdown of the local environment 4 The full set of results and detailed discussion of the model can be found in J. Huang, M. Rosegrant, and S. Rozelle. 42 Grain consumption is divided into two parts: grain that is directly consumed for food and that which is fed to animals and consumed indi- rectly. Direct food equations are divided into rice, wheat, and other Food Grain Demand Rural and urban food grain demand are modeled separately for several reasons. Consumption patterns are inherently different for rural and ban consumers. 4 Income differentials, expenditure growth, rates of hange of population, and other demographic factors also vary dramati ally between rural and urban regions. The effect of urbanization is ac counted for by multiplying per capita grain projections for each sector Copyright 1999. All rights reserved
Copyright © 1999. All rights reserved