Food Consumption in Rural China: Preliminary Results from Household Survey Data 1

Transcription

1 Food Consumption in Rural China: Preliminary Results from Household Survey Data 1 Zhang-Yue Zhou a, The University of Sydney Yan-Rui Wu b The University of Western Australia Wei-Ming Tian c China Agricultural University 1 Revised version of a paper presented at the International Conference on China in the New Era: Strategies for Sustainable Economic Growth and Business Responses to Regional Demands and Global Opportunities, held in RMIT University, Melbourne, Australia, 2-3 October We wish to thank the delegates of the Conference and the referees for their useful comments. This paper presents some preliminary results from our study and further work is being carried out which will be reported in a later stage. We welcome any further comments to help us improve our work. a,b,c Correspondence to: Zhang-Yue Zhou a Yan-Rui Wu b Asian Agribusiness Research Centre Department of Economics The University of Sydney The University of Western Australia Orange, NSW, 2800 Perth, WA., 6009 AUSTRALIA AUSTRALIA zzhou@orange.usyd.edu.au yanrui.wu@uwa.edu.au Wei-Ming Tian c College of Economics and Management China Agricultural University Beijing P.R. CHINA tianwm@sina.com Household Survey Data

2 Abstract The increase in consumer income in developing countries induces changes in the composition of food consumption. Such changes have important implications for policy makers and for food marketers. For a large developing country such as China, changes in food consumption have even much greater implications, not only for China itself but also for the rest of the world, due to the sheer size of its population. As a result of its rapid economic growth and the resulting increase in consumer income, expansion and diversification of food consumption in China have been most notable in the past two decades. It is believed that there is considerable latitude for food consumption growth in China and China will become one of the fastest-growing markets for food products in the coming years. Using household-level survey data of representative provinces, this study examines the changes in the proportion of food consumption expenditure out of total household expenditure and identifies patterns of change in food consumption in rural China by region and by income group. Major factors that affect consumption behaviour in rural China are investigated. Implications of the changes in China s rural food consumption are discussed. 1. Introduction China s population is reaching 1.3 billion. Feeding so many mouths has always been a challenge for the Chinese government. First, the large population needs to have an adequate amount of food to eat. Then, after they have had enough to eat, they may want to have better and more nutritious food to eat. Next, they may want to do less to prepare their food and thus they may want to consume more processed or semi-processed food. Indeed, there have been notable changes in food consumption by Chinese consumers in the recent times. For example, during the two decades between 1981 and 2001, direct consumption of feedgrains at the national level has dropped by 7% in rural areas and 45% in urban areas. On the other hand, the consumption of meat, eggs, and aquatic products in rural areas has increased by 85%, 278%, and 222%, respectively. For urban areas, the increases are 29%, 113% and 69%, respectively. Recent studies have also shown that consumer demand for processed food is on the increase (Yuan 2001, Wang and Yang 2003). Household Survey Data

3 Given the sheer size of the Chinese food market, changes in food consumption in China have important implications for the agri-food industries not only for China but also for many other countries that have an interest in supplying to the Chinese market, such as Australia. As such, the recent developments in China s food consumption have attracted much attention from researchers. However, despite the efforts of earlier studies, there are still some issues and questions that remain untouched or unanswered. This paper adds to our understanding of the dynamics of China s rural food consumption and their implications. 2 In the next section, we review earlier studies. Section 3 describes the data used in this study. Section 4 presents the results of preliminary analyses of the data. The findings are discussed in Section 5 with concluding comments. 2. Review of Literature A number of factors can cause changes in household consumption. They can be economical (e.g., income increase), social (e.g., changes in dietary culture due to urbanisation and foreign influences), market development that makes new foods available, and health concerns that lead to the preference towards, or deliberate avoidance of, some foods. Among these factors, however, income is seen to be the most influential one affecting household food consumption. Studies show that the levels of income do affect the composition of food consumption (Regmi et al. 2001; Jones et al. 2003). The increase in consumer income in fast growing developing countries, such as China and Malaysia, tends to induce drastic changes in the composition of food consumption (Cranfield et al. 1998, Coyle et al. 1998, Regmi 2001, USDA 2002, Ishida, Law and Aita 2003). Changes in food consumption in developing countries have important implications for policy makers in regard to a country s food industry strategic planning and for food suppliers in terms of market development. For a large developing country such as China, changes in food consumption have even much greater implications, not only for China itself but also for the rest of the world, due to the sheer size of its population. As a result of the rapid economic 2 Chinese rural and urban residents are two distinct consumer groups. It is useful to analyse the consumption dynamics of the two groups separately. We will discuss food consumption in urban China in another paper. Household Survey Data

4 growth and the resulting increase in consumer income in the past two decades, expansion and diversification of food consumption have been notable in China. It is believed that there is considerable latitude for food consumption growth in China and China will become one of the fastest-growing markets for food products in the coming years, due to its high prospective rates of economic growth (Wu 1999, p. 24). Because of the significant and wide implications of China s food consumption changes, researchers from both within and outside China have paid increased attention to examining this issue from various perspectives. A number of studies have used the popular Almost Ideal Demand System (AIDS) to identify factors that affect consumption behaviours and to derive parameters such as price and income elasticities; for example, Halbrendt and Tuan (1994), Fan, Wailes and Cramer (1995), Wu and Li (1995), Huang and Rozelle (1998). Wan (1998), using a nonparametric framework developed by Sakong and Hayes (1993), examines taste changes in food demand in rural China. Focusing on the structural change in the impact of income on food consumption in China, Guo, Mroz and Popkin (2000) followed a two-step process proposed by Haines, Guilkey and Popkin (1988). To identify determinants of expenditures on food for at-home consumption, Gould (2002) uses an econometric model specification that consists of (1) the system of share equations in which the shares are based on at-home food expenditures and (2) an auxiliary at-home food expenditure function. When reliable price information is unavailable, Engel functions may be used (Wu 1999). Obtaining data for China s consumption analyses used to be a challenge. This situation seems to have been improved in recent years although gaining quality data remains to be a major challenge. In the earlier attempts, researchers often resorted to cross-sectional data (e.g., Halbrendt and Tuan 1994, rural household survey data collected by China s State Statistical Bureau (SSB), 2560 households of Guangdong province for 1990; Huang and Rozelle 1998, 433 rural household surveys of SSB in Hebei province for 1993) or highly aggregated data (e.g., Fan et al. 1995, provincial aggregates for ; Wu and Li 1995, aggregated data from 33 cities for 1990; Wan 1998, provincial aggregates for ). In the more recent studies, Guo et al. (2000) used the data (for 1989, 1991, and 1993) from the China Health and Nutrition Survey data; the survey is conducted biannually and covers eight provinces with a little over 3000 households. On the other hand, Gould (2002) managed to secure three Household Survey Data

5 consecutive years of urban household survey data ( ) from SSB for Jiangsu, Shandong and Guangdong provinces. Earlier studies have contributed importantly to the understanding of the consumption behaviours in China. Most of them generated important parameters such as income and price elasticities of demand for various food items. Some provided other useful observations such as (1) demand is price-inelastic for most commonly consumed food items; (2) there is a tendency to shift away from rice to wheat consumption; and (3) Chinese consumers will consume more meats as their income increases. However, there have been conflicts between findings of earlier studies. Besides the different analytical methods used, data (i.e., the use of cross-sectional or very aggregated data) are likely a major source contributing to the different findings. To analyse consumer behaviours, cross-section and time-series data at the household level are most ideal. Such micro-level panel data allow us to construct and test more complicated behavioural models than purely cross-section data (Baltagi 1995, p. 5). Many variables can be more accurately measured at the micro level and biases resulting from aggregation over individuals are eliminated (Blundell 1988; Klevmarken 1989). Further, the time span of the panels used by Guo et al. (2000) and Gould (2002) seems to be short. When panels involve annual data covering only a short span of time, this means that asymptotic arguments rely crucially on the number of individuals tending to infinity (Baltagi 1995, p. 7). It is also noted that data used in the existing studies are somewhat outdated. Consumption behaviours have experienced rapid changes in the past decade in China, driven chiefly by increased consumer income and external cultural influences. Related to income growth are the likely changes in income elasticities. For some food items such as animal products income elasticities may increase when consumers incomes increase from a very low level. However, after income has reached a certain high level, income elasticities may decline with further increase in consumers income. For example, Shono et al. (2000) note that income elasticities decreased as income and consumption levels of some meats increased over time. Zhou (2001) reveals that the income elasticity of demand for milk tends to decline when income rises further from a very high level. Recent empirical evidence by Chu (2003), where he modelled urban meat consumption patterns, shows all animal products tend to exhibit a Household Survey Data

6 trend of declining income elasticities. As such, researchers have argued that some parameters derived based on data of a decade ago are less adequate for policy purposes in China and new estimates based on recent data are called for (He and Tian 2000, Zhou et al. 2003). This current study makes an additional contribution to the literature concerning China s rural food consumption in several ways. First, it uses the most up-to-date household-level survey data with a longer span of time ( ). Second, for better representation, it selects three provinces of different levels of economic development (Guangdong represents an economically developed region and Hubei a medium developed region while Yunnan an economically less developed region). Third, while the SSB household survey data have been used fairly extensively for study on China s food consumption, the household survey data collected by the Research Centre for Rural Economy of the Ministry of Agriculture remains little exploited. This study makes use of this data source. Finally, in addition to examining those common determinants of food expenditure, this study will also explore the likely impacts of some other factors on food consumption such as family type, family major business operations, tax and fee burden, and influences of family members not working on the land; these factors have been so far largely left untouched in the literature. 3. Data Household-level survey data are regularly collected by the Research Centre for Rural Economy (RCRE) of the Ministry of Agriculture in all regions in China. The survey first began in 1986 and was carried out annually except in 1992 and In each village, some 100 households were surveyed. The survey instruments have evolved over the years. Those used for were the same (with 312 variables), but they were expanded for the 1993 survey (with 394 variables) and further expanded for the surveys since 1995 (with 439 variables). Data between 1995 and 2002 are used in this study to ensure consistency of variables over years. The three provinces, Guangdong, Hubei and Yunnan, are selected for this study. They represent different levels of economic development; namely, Guangdong representing the most developed, Hubei the medium developed and Yunnan the least developed. Three Household Survey Data

7 villages were then selected from each province according to the level of economic development as measured by per capita net income. Brief information about the region where the chosen villages are located is given in Table 1, together with per capita income in each village. [Table 1 here] In the surveys, there are nine categories of major family business activities: 1) cropping, 2) forestry, 3) animal husbandry, 4) fishery, 5) industry, 6) construction, 7) transportation, 8) tertiary services, and 9) others. In this study, activities belonging to categories 1-4 are considered as farming and the rest as non-farming (i.e., categories 5-9). 3 To minimise errors in the data, we applied various criteria to a number of variables to cross check the data. Whenever doubt arose or when an observation has an un-legitimate entry (e.g., the number of family members is zero), the relevant observations are eliminated. 4. Preliminary Analysis Results of preliminary analyses of the survey data are reported below. (1) Food expenditure still accounts for a large portion of total living expenditure Table 2 shows that over 55% of total living expenditure is spent on food. This proportion is higher for the poorer province, Yunnan. In Table 2, the proportion of Guangdong, more economically developed province, is not significantly lower than that of Yunnan due to the following reasons. (1) The income level of Yunnan s Village 1 is very high, thus pulling down the provincial average (see Table A1). (2) Although the income level of Guangdong s Village 1 is very high, that of the other two villages is low, pushing up the provincial average. At the village level, about 66% of total living expenditure were spent on food in these two villages of Guangdong in 2002 (see Table A1). Table A1 also shows that in the poorest 3 In 1996, one further category was added: no family business operations. Currently, the number of families without any business operations is small. Household Survey Data

8 village, Village 2 of Yunnan, by 2002, there is still some 75% of total living cost spent on food. [Table 2 here] While the proportion of expenditure on food is slowly declining, the proportion of expenditure on other items is increasing, chiefly, houses, tertiary services, and culture. However, there are distinct regional differences. In the poorer province, Yunnan, the proportion of expenditure on culture is small and also increases at a very slow pace. This same proportion in other two provinces is almost three times that in Yunnan (Table 2). (2) Proportion of food expenditure on staple food is declining but that on away from home consumption is increasing At the all-three province level, over the past decade, expenditure on staple food (chiefly foodgrains in China) among total food expenditure is declining (from 39% in 1995 to 35.3% in 2002) while expenditure on away from home consumption has steadily increased (from 7% in 1995 to 11.7% in 2002) (Table 3). Expenditure on non-staple food declined slightly from 54% in 1995 to 53% in [Table 3 here] (3) Consumption of foodgrains is declining and that of food of animal origin is slowly increasing Table 4 indicates that, as rural income increases over the past decade, there is a clear declining trend in direct consumption of foodgrains. However, at the all-three-province level, the increase in the consumption of food of animal origin is not very notable except in the case of poultry meat where there is a 28% increase in 2002 compared with At the more disaggregated levels, a slow increase in the consumption of most animal products can be established (see, Table 4 at the provincial level and Table A2 at the village level). [Table 4 here] Household Survey Data

9 However, while there is a general trend in reduced consumption of foodgrains and increased consumption of animal products, there are clear regional differences both in the speed of changes in consumption composition and the level of consumption of various animal products. For example, Table A2 shows that in Village 2 of Yunnan province, the poorest village among the chosen nine villages, consumption of animal products is very low and in some cases declining. Consumption of aquatic products and poultry eggs is almost nil. The consumption of animal products in the two villages in Guangdong province, Villages 2 and 3, also experienced stagnation or even reduction. On the other hand, the increase in the consumption of animal products in the two richer villages, Village 1 of Guangdong and Village 1 in Yunnan, is most notable. The level of consumption of various animal products in these two villages is also much higher than that in other villages (Table A2) (Higher level consumption of aquatic products in the villages of Guangdong is also related to local dietary habits). The above tends to confirm that, in addition to some regional influences on dietary habits, consumer income is a most import determinant that affects the composition and the level of food consumption. Analysis between the income level and food consumption that follows provides further confirmation. (4) Income level is the major determinant on the composition and the level of food consumption Table 5 clearly shows that as income increases, the proportion of living expenditure spent on food declines. For those living under poverty, a very high portion of the living expenditure, being almost 80%, is spent on food. 4 This same proportion is only a little over 30% for those very wealthy residents with a per capita income of more than yuan. On the other hand, as income increases, the Chinese rural residents spend more on improving their dwelling conditions and obtaining more tertiary services. Proportionally, the expenditure on culture tends to increase first and then start to decline (Table 5). 4 In 1995, the poverty line for rural residents in China was 530 yuan (1US$= in 1995). In 1999, this was raised to be 625 yuan (1US$= in 1999). The poverty line set by the United Nations for developing countries is US $31.67 per capita per month, or roughly 1US$ per day. Village-level per capita income data show that in Village 3 of Yunnan province many households live near or under the poverty line as set by the Chinese government. Most worrisome is that the per capita income in this village sometimes declines excessively. Household Survey Data

10 [Table 5 here] Table 5 further shows that, as income increases, the proportion of food expenditure on staple food, chiefly foodgrains in China, declines rapidly while that on non-staple food and on away-from-home consumption increases, suggesting that the composition of their food consumption changes and diversifies. This is clearly reflected in Table 6. An increase in income from very low level initially leads to increased direct consumption of foodgrains. This is not surprising as the poor have difficulty in getting enough to eat and they will consume more foodgrains before they commit their limited resources to consume other food of more a luxury nature. However, as the income further increases, the consumption of foodgrains starts to decline while that of other food items continue to increase (Table 6). [Table 6 here] While most of other food items, chiefly, animal products, further increases as income goes up, some other food items do not follow the same pattern. Among the latter, the consumption of animal oil increases and then declines most likely due to health concerns. The consumption of vegetable exhibits a similar patter to animal oil perhaps due to increased consumption of fruits. Sugar consumption tends to stabilise as income reaches a certain level again likely due to health concerns. On the other hand, the consumption level of other food items by higher income groups of rural residents (namely, pork, beef and mutton, milk, poultry meat and eggs, and aquatic products) is significantly higher than that of those lower income groups of rural people. (5) Other factors that affect rural consumption Some other factors may affect food consumption, though indirectly, through their influence on the level of income. In Table 7, family heads of those with lower per capita income tend to have lower education level. Table 7 also indicates that a family s ability to earn income off farm has a very strong impact on their income level and hence their consumption patterns. Those families with very high per capita income derive a very small portion of their income from agriculture. Their living expenditure is over 35 times that of the poorest group. It is also interesting to note that those very poor families proportionally pay much higher taxes and Household Survey Data

11 charges. These further eat into their very bare means of surviving. Cross tabulation was carried out to examine whether family type would have any impact on the patterns of food consumption in rural China. No clear relationships between the two could be established although those extended families do tend to have slightly higher milk consumption. This is likely due to the fact that such larger families tend to have more children and elderly and thus higher milk consumption. Family size and the age of family head are also found to have little impact on the patterns of food consumption. [Table 7 here] The preliminary analyses of the data revealed some useful trends in relation to food consumption in rural China. These trends confirm that important changes are taking place in the front of rural food consumption. The preliminary analyses also pinpointed some factors that may have contributed to the changes in food consumption patterns in rural China. An understanding of such factors is of great relevance to government food policy formation and food marketing strategies. Major factors that may affect the level and composition of food consumption in rural China may be placed into the following groups: (1) Prices of food (2) Per capita net income (3) Tax and fee burdens 5 (4) Regional differences (5) Other non-price and non-income factors, chiefly, family characteristics, including: 1) family size 2) family type 6 3) age of household head 4) education level of household head 5 Tax and fee burden affects net income. We examine this factor separately because we are interested in finding out whether the higher proportion of tax and fees paid by the lower income group of people negatively affect their food consumption. 6 1) and 2) are related. Family type will affect the size of the family. They may affect food consumption patterns, e.g., a larger family is more likely to have children and elderly and hence has to cater for the needs of the children and elderly. Household Survey Data

12 5) proportion of non-rural labour out of total family labour 6) proportion of income from non-farming sources 7 The RCRE household survey data only collect the quantity of various food items consumed by households without the corresponding consumption expenditure. This makes it impossible to infer any price information. Various other alternatives have been attempted but they can help little to derive any meaningful price information. The absence of price data is rather unfortunate and it limits our ability to analyse various effects related to price changes such as income effect and substitution effect. Nonetheless, our database contains much other useful household information that will still entail us to discover the effects associated with many other factors as enumerated in the above. Our econometric modelling will focus on such other factors and will be carried out in our next stage s work. 5. Some Concluding Comments In spite of the absence of empirical results based on econometric modelling, some important observations can still be derived based on some preliminary analyses. While food expenditure still accounts for a large proportion of total living expenditure, this proportion is declining. The proportion of expenditure on housing, tertiary services and education is on the increase. This trend will continue into the near future for China s rural areas. As such, although the proportion of income spent on food will decline, total expenditure spent on food will continue to rise due to increased consumer income. On the other hand, the market potential for non-food consumer goods will be enormous. Direct consumption of foodgrains has started to decline in rural China. At the all-three province level, the proportion of food expenditure on staple food is declining while that on away from home consumption has steadily increased. Expenditure on non-staple food 7 5) and 6) are also related and both are also related to income. Generally if they are higher, we can expect the family has higher per capita income. However, they can be used as proxies to capture the influence of other two factors on consumption, i.e., urbanisation and influence of outside culture. With more labour engaged in nonfarming/non rural work, they get more exposed to influences from elsewhere which in turn will affect their family consumption patterns. Household Survey Data

13 declined slightly. This suggests that there is much room for food catering industry development. While direct consumption of foodgrains is declining, there is an increase in the indirect consumption of grains; that is, through increased consumption of animal products whose production consumes grains. However, according to the survey data, at the all-three province level, the increase in animal product consumption by the Chinese rural residents is rather slow over the past decade. It implies that unless there are drastic changes (e.g., rapid increase in consumer income), the increase in the consumption of animal products in China s rural areas is likely to be small. 8 However, there are significant regional differences in the composition of food consumption. Per capita consumption of aquatic products in Guangdong is much higher than that in Yunnan. Apart from income differences, availability of some foods is a key determinant that leads to different dietary habits in different regions. Unlike the markets in developed economies, many agricultural markets in China are not well integrated and some food items are simply not available in the vast rural areas in China. It may be anticipated that, as the Chinese agricultural markets get further integrated, the total demand for some food items will increase. Income plays a major role in determining the composition and the level of food consumption. As income increases, the proportion of food expenditure on staple food, chiefly foodgrains, declines rapidly while that on non-staple food and on away-from-home consumption increases. This suggests that the composition of their food consumption changes and diversifies. The consumption level of pork, beef and mutton, milk, poultry meat and eggs, and aquatic products by higher income groups of rural residents is significantly higher than that of those with lower income. Future increase in rural income will have a great impact on food consumption in China. 8 It is noted that faster increase in the consumption of some animal products took place in the early 1990s, largely attributable to a faster increase in income. However, there has been a slowing down in consumer income increase in rural areas in the past several years. During , per capita income growth in rural areas was 3.2%, down from 25.2% during the previous five-year period ( ). Household Survey Data

14 A family s ability to earn income off farm has a very strong impact on their income level and hence their consumption patterns. Those families with very high per capita income derive a very small portion of their income from agriculture. Apart from the fact that more family labour working off farm earns the family higher income, it is also likely that those working off farm have great exposure to external influences and subsequently influence home food consumption. As such, increased rural-urban exchanges and urbanisation will impact on China s rural food consumption. Very poor families proportionally pay much higher taxes and charges. Government s efforts in reducing the poor s tax burden will help them to improve their food intake. In addition, heads of those poor families tend to have lower education level. This is another area where government action is needed. Household Survey Data

15 References Baltagi, B.H. (1995), Econometric Analysis of Panel Data, Wiley, Chichester. Blundell, R. (1988), Consumer behaviour: Theory and empirical evidence A survey, The Economic Journal, Vol. 98, pp Chu, Y.T. (2003), An analysis on China s feed market, Ph.D. Dissertation, China Agricultural University, Beijing. Coyle, W. Gehlhar, M., Hertel, T.W., Wang, Z. and Yu, W.S. (1998), Understand the determinants of structural changes in world food market, Staff Paper 98-05, GTAP Centre, Purdue University. Cranfield, J.A.L., Hertel, T.W., Eales, J.S. and Preckel, P.V. (1998), Changes in the structure of global food demand, Staff Paper 98-05, GTAP Centre, Purdue University. Fan, S.G., Wailes, E.J. and Cramer, G.L. (1995), Household demand in rural China: a twostage LES-AIDS model, American Journal of Agricultural Economics, Vol. 77, pp Gould (2002), Household composition and food expenditure in China, Agribusiness, Vol. 18, pp Guo, X.G., Mroz, T. A. and Popkin, B.M. (2000), Structural change in the impact of income on food consumption in China, , Economic Development and Cultural Change, Vol. 48, pp Haines, P.S., Guilkey, D.K. and Popkin, B.M. (1988), Modelling food consumption decisions as a two-step process, American Journal of Agricultural Economics, Vol. 70, pp Halbrendt, C. and Tuan, F. (1994), Rural Chinese food consumption: The case of Guangdong, American Journal of Agricultural Economics, Vol. 76, pp Household Survey Data

16 He, X.R. and Tian, W.M. (2000), Livestock consumption: diverse and changing preferences, in Yang, Y.Z. and Tian, W.M. (eds) China s Agriculture at the Crossroads, Macmillan Press, London, pp Huang, J.K. and Rozelle, S. (1998), Market development and food demand in rural China, China Economic Review, Vol. 9, pp Ishida, A., Law, S.H. and Aita, Y. 2003, Changes in food consumption expenditure in Malaysia, Agribusiness, Vol. 19, pp Jones, E., Akbay, C., Roe, B. and Chern, W.S. (2003), Analyses of consumers dietary behaviour: An application of the AIDS model to supermarket scanner data, Agribusiness, Vol. 19, pp Klevmarken, N.A. (1989), Panel studies: What can we learn from them? Introduction, European Economic Review, Vol. 33, pp Regmi, A., Deepak, M.S., Seale, J.L. and Bernstein, J 2001, Cross-country analysis of food consumption patterns, in Regmi, A. (ed.), Changing Structure of Global Food Consumption and Trade, ERS WRS No. 01-1, USDA, Washington, D.C. Regmi, A. (ed.), Changing Structure of Global Food Consumption and Trade, ERS WRS No. 01-1, USDA, Washington, D.C. Sakong, Y. and Hayes, D.J. (1993), Testing the stability of?references: a nonparametric approach, American Journal of Agricultural Economics, Vol. 75, pp Shono, C., Suzuki, N. and Kaiser, H.M. 2000, Will China s diet follow western diets? Agribusiness, Vol. 16, pp Wan, G.H. (1998), Nonparametric measurement of preference changes: The case of food demand in rural China, Applied Economics Letters, Vol. 5, pp Household Survey Data

17 Wu, Y.R. and Li, E. (1995), Food consumption in urban China: An empirical analysis, Applied Economics, Vol. 27, pp Wu, Y.R. (1999), China s Consumer Revolution, Edward Elgar, Cheltenham. Yuan, X.G. (2001), A study on the consumption of animal products in China, Ph.D. Dissertation, Chinese Academy of Agricultural Sciences, Beijing. Zhou, J.L. 2001, A study on the dairy market in China, Ph.D. Dissertation, China Agricultural University, Beijing. Zhou, Z.Y., Tian, W.M., Liu, X.A. and Wan, G.H. (2003), Studying China s feedgrain demand and supply: Research methodological issues, in Zhou, Z.Y. and Tian, W.M. (eds), China s Regional Feedgrain Markets: Developments and Prospects, Grains Research and Development Corporation, Canberra. Household Survey Data

18 Table 1: Brief Description of Sample Regions Province and Rural Per City from which a Sample Villagelevel Per Brief Description of the Region Capita Income Village is Capita ( ) Drawn Income ( ) Hanyang (Village 1) 3253 A relatively developed region within the province with good transportation Hubei Hanchuan 3049 Similar to Hanyang (2352) (Village 2) Changyang (Village 3) 1838 A hilly area with a less developed economy and market Dongguan (Village 1) Highly industrialised area close to Guangzhou and Hong Kong Guangdong (3770) Dianbai (Village 2) 2651 Relatively less developed area within Guangdong province Wuchuan 2333 Similar to Dianbai but in a coastal area (Village 3) Yuxi (Village 1) 6638 A wealthier area with much injection from a highly developed tobacco industry Yunnan (1534) Lanchang (Village 2) 509 Very remote and mountainous area, one of the poorest regions in the province and in China Dali (Village 3) 2070 Very remote and mountainous area, one of the poorer regions but with tourist resources Note: Per capita net income at the village level for 2001 is reported here to be comparable with the provincial average in 2001, the latest available. Sources: SSB (2002); RCRE survey data. Household Survey Data

19 Table 2: Changes in the Composition of Living Expenditure in Rural China (%) Year Per Capita Income ( ) Food Cloth House Fuel Daily Consumables Services Culture Other All Three Provinces Hubei Guangdong Household Survey Data

20 Yunnan Note: The per capita income of Guangdong suddenly almost doubled in This is caused by an abrupt increase in per capita income (more than doubled) in Village 1 of Guangdong in that year. Source: calculated from RCRE survey data. Household Survey Data

21 Table 3. Composition of Food Expenditure Year Staple Food Non-Staple Food Away from Home Consumption Source: calculated from RCRE survey data. Household Survey Data

22 Table 4: Trends in Per Capita Consumption of Major Food Items in Rural China (kg) Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Oil Animal Oil Pork Beef and Milk Mutton Poultry Meat Eggs Aquatic Fruits Sugar Products All Three Provinces H ubei Household Survey Data

23 Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Oil Animal Oil Pork Beef and Milk Mutton Poultry Meat Eggs Aquatic Fruits Sugar Products Guangdong Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Oil Animal Oil Pork Beef and Milk Mutton Poultry Meat Eggs Aquatic Fruits Sugar Products Yunnan Household Survey Data

24 Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Oil Animal Oil Pork Beef and Milk Mutton Poultry Meat Eggs Aquatic Fruits Sugar Products Source: calculated from RCRE survey data. Household Survey Data

25 Table 5: Consumer Income and Proportion of Living Expenditure on Food and Other Items in Rural China Expenditure on Food Consumption (%) Expenditure on Other Items (%) Income Group Total Food Among which (%) Cloth House Fuel Daily Expenditure Culture Other ( ) Expenditure Staple Food Non-Staple Food Away from Home Consumables Consumption on Tertiary Services > Source: calculated from RCRE survey data. Household Survey Data

26 Table 6: Consumer Income and the Level of Consumption of Different Food Items in Rural China Income Group ( ) Foodgrains (unmilled) (kg) Foodgrains (bought) (kg) Vegetables (kg) Vegetable Oil (kg) Animal Oil (kg) Pork (kg) Beef and Mutton (kg) Milk (kg) Poultry Meat (kg) Eggs (kg) Aquatic Products (kg) Fruits (kg) Sugar (kg) > Source: calculated from RCRE survey data. Household Survey Data

27 Table 7: Relationships between Income and Other Family Attriutees Income Group ( ) No. of Obs % of total Per Capita Education Non-Agric. Per Capita Income Level of Income out Living ( ) Family Head of total Income Expenditure ( ) Tax/Fee Tax/Fee Payments Payments out Per Capita of Income Per ( ) Capita (%) (%) > Source: calculated from RCRE survey data. Household Survey Data

28 Appendix Table A1: Changes in the Composition of Living Expenditure at the Village Level in Rural China (%) Per Capita Food Cloth House Fuel Daily Services Culture Other Year Income Consumables Hubei, Village Hubei, Village Hubei, Village Household Survey Data

29 Year Proceedings of the 15 th Annual Conference of the Association for Chinese Economics Studies Australia Per Capita Income Food Cloth House Fuel Daily Consumables Services Culture Guangdong, Village Other Household Survey Data

30 (Table A1 continued) Per Capita Food Cloth House Fuel Daily Services Culture Other Year Income Consumables Guangdong, Village Guangdong, Village Yunnan, Village Yunnan, Village Household Survey Data

31 Year Proceedings of the 15 th Annual Conference of the Association for Chinese Economics Studies Australia Per Capita Income Food Cloth House Fuel Daily Consumables Services Culture Yunnan, Village Source: calculated from RCRE survey data. Other Household Survey Data

32 Table A2: Trends in Per Capita Consumption of Major Food Items in Rural China at the Village Level (kg) Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Animal Pork Beef and Milk Poultry Eggs Aquatic Fruits Sugar Oil Oil Mutton Meat Products Hubei, Village Hubei, Village

33 Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Animal Oil Oil Pork Beef and Milk Poultry Eggs Aquatic Mutton Meat Products Hubei, Village

34 (Table A2 continued) Year Per Capita Income ( ) Foodgrains (unmilled) Vegetables Vegetable Animal Pork Beef and Milk Poultry Eggs Aquatic Fruits Sugar Oil Oil Mutton Meat Products Guangdong, Village Guangdong, Village